Michael R. Baye Dennis W. Jansen

MONEY, BANKING *
FINANCIAL MARKETS
n eonomics
Baye / Jansen
Wr Hi) q 6? 14% 1•«*n4%+’ft'

i
(3/4 96(1 6 2 <3 60 15% 3 65ft 1 55ft — 3; 58 50 i
' 3 054 6? q 170 17 16519 !6!/a + Vs, 67 56 4.
>!■/» 78 6? q 64 12>/8 125ft 12*78- '(<1 27iq 241)81
ftft 1 tls 72 q 375 (5% 1$% 153» + He' (03ft 9-VjJ
■A S3 6? 13 133 t4Jft 14 14 - 1 k mk 24%
%l€Uft *!
vi? " . ’
2 Wb 19W04
.11% 81)4 10'.
9% 8 ..Mh
(37%) 10*fe1 OB*'
W< 651® Off
2i'l4 9 4:
liift 12%' 12 ,uw
25V4 2004 80 | ,mSGiZK
'iiWi. ..iHIIb ?4«i
533)4 447/8.32 S'S

52% 50 1.25/5.1
271)4 251/8.# 7<
13V» 113)4/-c i
1/8+ 1/4 32VI 23% •.?• yS
I’Vi'-i’ft-
73'tt 2:
80%64«16 2B2S 79 781ft

’Sift + 'ftT 26% 23’* 03 84
19% 14'ls 23 1 5 '8 6183)4 1894'83/4+ 3)9 54 32% 12 32
343.(4 193/4 24 0 7 7a 1345ft 341/8
341/4 ?9 23'<6-51 2.0
191/8 151/4! 178 74 q jo 16 163(4
( 53ft- >ft
('90 263ft »'+ ( 3ft 27 18 0% 60
15 1298101 78 5 -13 12')8 12%
838 17% 17% 17%
14 11% 87 75 d h 11 ;/8 115ft..>'5ft- U. 20% 1594 60 32
184 483ft 489s 483ft
m •+% ’+• 9994 44l)»' 1’ 1 4 IS &*’“• 7,f^k X Aj^,! «
Money, Banking, and
Financial Markets:
An Economics Approach
'
V
.
Money, Banking, and
Financial Markets:
MICHAEL R. BAYE
Pennsylvania State University
DENNIS W. JANSEN
Texas A & M University
Houghton Mifflin Company Boston Toronto
Geneva, Illinois Palo Alto Princeton, New Jersey

\
To my parents, Firmin and Jaynet Baye,

for their loving support over the past 36 years.
MRB
In memory of Elmer Jansen and William Hennessey.

DWJ
Sponsoring Editor: Denise Clinton

Basic Book Editor: Karla Paschkis
Senior Project Editor: Susan Westendorf
Associate Production/Design Coordinator: Caroline Ryan-Morgan
Senior Manufacturing Coordinator: Marie Barnes
Credits
Cover design: Judy Arisman

Cover image: William Whitehurst, NYC
Copyright © 1995 by Houghton Mifflin Company. All rights reserved.
No part of this work may be reproduced or transmitted in any form or by any means, electronic or
mechanical, including photocopying and recording, or by any information storage or retrieval
system without the prior written permission of Houghton Mifflin Company unless such copying is
expressly permitted by federal copyright law. Address inquiries to College Permissions, Houghton
Mifflin Company, 222 Berkeley Street, Boston, MA 02116-3764.
Printed in the U.S.A.

Library of Congress Catalog Card Number: 94-76473
ISBN: 0-395-64395-3
ISBN EXAM: 0-395-71667-5
12 3 4 5 6 7 8 9-DH-98 97 96 95 94
Brief Contents
PART ONE Introduction to Money and Banking i
CHAPTER 1 Money and Banks 2

2 Financial Markets, Financial Institutions and
Instruments, and Money 31
3 Money, Inflation, and Interest 61
PART TWO The Microeconomics of Banking and Financial Markets 91
CHAPTER 4 Supply and Demand in Financial Markets 92

5 The Bank as a Firm: Loans 130
6 The Bank as a Firm: Deposits 166
7 The Banking Industry 201
PART THREE The Valuation of Financial Assets 235
CHAPTER 8 The Time Value of Money and Asset Pricing 236

Risk, Uncertainty, and Portfolio Choice 269
The Term Structure of Interest Rates 306
11 Foreign Exchange Markets 330
Rational Expectations and Efficient Markets 369
PART FOUR The Creation of Money 405
CHAPTER 13 The Federal Reserve System: History, Modern Structure, and Policy
Tools 406
14 The Money Supply Process 446
The Demand for Money and Equilibrium in the Money Market 489
PART FIVE Money and the Macro Economy 525
CHAPTER 16 A Simple Macroeconomic Model 526

17 Open Economy Macroeconomics 586
18 Money and Economic Activity: A Look at the Evidence 625
PART SIX Special Issues in Monetary Economics 665
CHAPTER 19 An Introduction to Monetary Policy 666

Monetary Policy: Rules, Discretion, and Games 708
Government Spending and Finance: the Deficit Problem 746
v
The Authors
4.
Michael R. Baye (Penn State) received his Dennis W. Jansen (Texas A&M) received his
B.S. in economics from Texas A&M Univer¬ A.B. in economics and mathematics from St.
sity in 1980 and his Ph.D. from Purdue Louis University in 1978 and his Ph.D. from
University in 1983. He has taught at Purdue the University of North Carolina at Chapel
University, where he received the Outstand¬ Hill in 1983. He has taught at the University
ing Graduate Instructor award, as well as the of North Carolina, North Carolina State Uni¬
University of Kentucky, Texas A&M Univer¬ versity, Indiana University, and Texas A&M
sity, and The Pennsylvania State University. University. His primary research interests are
His primary research interests are game theory monetary and macro economics. He is the au¬
and industrial organization. He is the author thor of two books and over 30 articles pub¬
of three books and 30 articles published in lished in economics journals, including The
economics journals, including American Eco¬ Economic Journal, Review of Economics and
nomic Review, Journal of Political Economy, Statistics, Journal of Money, Credit, and
Econometrica, and the Review of Economic Banking, and Journal of Econometrics. Since
Studies. He has been honored with a Fulbright 1992, Jansen has been a Research Fellow at
Lecturer/Research Scholar Grant as well as a Texas A&M University’s Center for Research
research grant from the National Science in Free Enterprise; he has also been a Visiting
Foundation. Scholar at the Federal Reserve Bank of St.
Louis, and on the visiting faculty at Catholic
University of Leuven and Erasmus University
of Rotterdam.
VI
Contents
Preface xxiii
part one Introduction to Money and Banking i
chapter i Money and Banks 2
Money 3
Functions of Money 5
The Origin of Money 8
Types of Money 9
Physical Properties of Money 12
The Evolution of Banks 14
Banks as Depositories 15
Demand Deposits 15
Bank Notes 16
Fractional Reserve Banking 18
Banks as Financial Intermediaries 22
Conclusion 27
Box 1.1 The Data Bank Components of the Money Supply 4
Box 1.2 International Banking Money on the Yap Island and Gold
Transfers Among Nations 10
Box 1.3 Inside Money Cigarettes as Money in a Prisoner of War Camp
Box 1.4 Inside Money Where Are All the Dollars? 17
chapter 2 Financial Markets, Financial

Institutions and Instruments, and Money 31
Financial Markets 31
Debt Markets 33
Equity Markets 33
Financial Service Markets 33
Financial Institutions 35
Depository Institutions 35
Nondepository Institutions 36
VIII Contents
Financial Instruments 41
Money Market Instruments 41

Capital Market Instruments 45
International Financial Instruments 52
Official Definitions of the Money Stock 54
Ml 55
M2 55
M3 55
L 56
Conclusion 57
Box 2.1 The Data Bank Changes in the Relative Size of Financial
Institution Assets, 1976-1990 38
Box 2.2 Inside Money Money Market Rates from The Wall Street
Journal 46
Box 2.3 Inside Money A Lexicography for Bonds 48

Box 2.4 Inside Money Are All Bonds Created Equal? 53
CHAPTER 3 Money, Inflation, and Interest 6i
Distinguishing Between Inflation and Economic Growth 61
The Price Level and Real Output 62

Inflation and Economic Growth 64
The Relationship Between Inflation and Economic Growth 65
Causes of Inflation 67
Interest Rates 72
What is Interest? 73
The Time Value of Money 75
How Inflation and Taxes Affect Interest Ra tes 80
The Real Interest Rate 80

The Aftertax Interest Rate 85
Historical Data on Interest Rates 86
Conclusion 88
International Banking Inflation Rates Around the World 66
Box 3.2 Inside Money Points and Mortgage Interest Rates 74
Box 3.3 INSIDE Money 15-Year Versus 30-Year Mortgages 76
Box 3.4 International Banking Real Interest Rates Around the World 84
Contents ix
PART TWO The Microeconomics of Banking and Financial Markets 9i
CHAPTER 4 Supply and Demand in Financial Markets 92
Basic Supply and Demand Analyis: A Brief Review with Applications

to Fee-Based Financial Services 92
Demand for Fee-Based Financial Services 93

Supply of Fee-Based Financial Services 96
Equilibrium in the Market for Fee-Based Financial Services 98
Changes in Equilibrium: Applications to the Insurance Market 100
The Market for Loanable Funds 104
The Demand for Loanable Funds 104

The Supply of Loanable Funds 110
Equilibrium in the Market for Loanable Funds 115
Applications of the Loanable Funds Model 116
The Tax Reform Act of 1982 and Interest Rates on Consumer Credit 116
Inflationary Expectations and the Mortgage Interest Rate 117
Returns on Real Estate Investments and Corporate Interest Rates 118
Recessions and the Prime Interest Rate 120
The Federal Deficit and Interest Rates on Government Bonds 122
Tax Policy and Interest Rates on Municipal Bonds 123
Interest Rate Ceilings 123
Interest Rate Floors 124
Conclusion 127
Box 4.1 The Data Bank Actual and Expected Rates of Inflation 108
Box 4.2 Inside Money Tax Rates and the Aftertax Interest Rate
in Two States 113
Box 4.3 Inside Money Nominal Interest Rates and Expected

Inflation Rates 119
CHAPTER 5 The Bank as a Firm: Loans no
Purely Competitive Banks 130
Market Demand and the Demand for an Individual Bank’s Loans 132
The Purely Competitive Bank’s Loan Decision 134
Banks with Market Power 139
Sources of Market Power for Banks 139

Profit-Maximizing Loan Decisions for Banks with Market Power 143
X Contents
Oligopoly Banks 147
The Nature of Oligopoly Interdependence 148

Using Game Theory to Model Oligopolistic Interdependence 148
Repeated Interaction 151
Banks with Imperfect Information 153
Symmetric Information 153

Asymmetric Information and Adverse Selection 154
Bank Strategies for Countering Asymmetric Information 158
Conclusion 163
Box 5.1 1 he Data Bank Economies of Scale in Banking 142
Box 5.2 International Banking Concentration Ratios in Banking 149
Box S3 Inside Money Going for Broke in the Savings and Loan
Industry 162
CHAPTER 6 The Bank as a Firm: Deposits m

Deposits as an Input in Producing Loans 166
Total and Marginal Product 167

A Simple Technology for Producing Loans 168
The Demand for Deposits by Purely Competitive Banks I7i
Market Supply and the Supply of Deposits to an Individual Bank 171

The Value Marginal Product of Deposits 173
The Demand for Deposits by Banks with Market Power 176
Market Power in the Loan Market Only 177

Market Power in Both the Loan and Deposit Markets 181
Uncertainty and Bank Deposits 185
Bank Withdrawals and the Law of Large Numbers 185

Bank Panics 190
Moral Hazard 193
Conclusion 197
Box 6.1_1 he Data Bank Loan and Deposit Interest Rates 174
Box 6,2 Inside Money The Law of Large Numbers and Deposit
Withdrawals 188
Box 6.3 International Banking The Bank Panic of 1907 192

Contents xi
Box 6.4 Inside Money Failures of Insured Thrifts and

Banks, 1934-1989 196
CHAPTER 7 The Banking Industry 201
Dual Banking 201
Acquiring a Bank Charter 202
Supervision and Examination of Banks 202
Status of the Loan Portfolio 204

Bank Capitalization 204
Overlapping Responsibilities 205
The Federal Reserve System 206
The FDIC 207
The S&L Crisis and Changes in the FDIC 209
An Overview of Banking Regulation 211
How Branching Restrictions Shaped the Banking Industry 212
Bank Holding Companies 212

Electronic Banking 214
Current Status 214
How Glass-Steagall Shaped the Banking Industry 214
Restrictions Against Investment Banking 215

Interest Rate Restrictions 216
The Economic Impact of Banking Regulations 220
Bank Charters 220

Branching Restrictions 222
Interest Rate Regulations 224
Uniform Reserve Requirements 226
Deposit Insurance 228
Bank Examination as a Solution to Moral Hazard and Adverse Selection 229
Conclusion 230
Box 7.1 INSIDE MONEY Requirements for Obtaining a Bank Charter 203
Box 7.2 International Banking Banking in the European Community and

the Principle of Mutual Recognition 213
Box 7.3 The Data Bank Employment in Banking and ATM Machines 215
Box 7.4 Inside Money The Long Phaseout of Regulation Q 217

XII Contents
PART THREE The Valuation of Financial Assets 235
CHAPTER 8 The Time Value of Money and Asset Pricing 236
The Valuation of Debt Instruments 238
Treasury Bills and Other Debt Instruments Sold on a Discount Basis 239
Coupon Bonds and Other Interest-Bearing Debt Instruments 244
The Equilibrium Price and Quantity of Bonds 25 i
Bond Prices: The Loanable Funds Approach 252

Bond Prices: The Supply and Demand of Bonds Approach 253
Valuing Stock and Other Assets 257
Income Stocks 259

Growth Stocks 260
Equilibrium Price and Volume of Stock Transactions 262
Conclusion 265
Inside Money How to Read Information on Treasury Bills in the

Wall Street Journal 241
Inside Money How to Read Information on Treasury Bonds and

Notes in the Wall Street Journal 246
Box 8.3 Inside Money How to Read Information on Corporate Bonds in

the Wall Street Journal 250
The Data Bank The Inverse Relationship Between Bond Prices

and Interest Rates 254
Box 8.5 Inside Money How to Read Information on Stocks in the Wall
Street Journal 258
CHAPTER 9 Risk, Uncertainty, and Portfolio Choice 269
Uncertainty and Risk 269
Expected Value 271

Variance 272
Risk Preference 273
Valuing Risky Financial Assets 274
Default Risk 275

Liquidity Risk 282
Interest Rate Risk and Call Risk 285
Contents xiii
The Inflation Premium and Inflation Risk 289

Tax Risk 291
Minimizing Financial Risk 294
Diversification 295
Future Markets 297
Options Markets 299
Conclusion 303
Box 9.1 INSIDE Money What Bond Ratings Mean 277
Box 9.2 The Data Bank Interest Rates on Aaa, Baa, and U.S. Treasury
Bills, 1971-1992 283
Box 9.3 Inside Money What Are Junk Bonds and Convertible Bonds? 284
Box 9.4 Interna tional Banking The Beginning of World War II and
Interest Rates 293
Box 9.5 INSIDE Money How to Reach Information on Future Markets in

the Wall Street Journal 298
Box 9.6 INSIDE Money How to Read Information on Futures Options

Markets in the Wall Street Journal 301
CHAPTER 10 The Term Structure of Interest Rates 306
The Yield Curve 306
Explanation of the Term Structure of Interest Rates 309
The Expectations Hypothesis 311

The Segmented-Markets Hypothesis 319
The Preferred Habitat Hypothesis 322
Conclusion 327
Box 10.1 Interna tional Banking Yield Curves for Different
Countries 310
Box 10.2 INSIDE Money How to Read Information on the Treasury Yield
Curve in The Wall Street Journal 312
Box 10.3 The Data Bank Calculating Expected Future Interest Rates 318
CHAPTER 11 Foreign Exchange Markets 330
Fundamentals of Foreign Rates 330
The Market for Foreign Exchange 335
Long-Run Exchange Rate Determination 336

Exchange Rates in the Short Run 344
XIV Contents
Spot and Forward Exchange Rates 351
Participants in the Forward Market 352
Fixed and Flexible Exchange Rate Systems 357
Flexible Exchange Rates 357

Central Bank Intervention 357
Fixed Exchange Rates 359
A Brief History of the Foreign Exchange Rate System 362
The Classical Gold Standard 362

Bretton Woods 363
Post-Bretton Woods 364
Conclusion 366
Box 11.1 Inside Money Exchange Rate Information in the Wall Street
Journal 332
The Data Bank Exchange Rate Movements between

the U.S. Dollar and Four Major Foreign Currencies 334
Box 113 International Banking Purchase Power Parity; the Big Mac
Standard, and the Economist Standard 345
International Banking The European Monetary System 365
CHAPTER 12 Rational Expectations and Efficient Markets 369
The Role of Expectations in Financial Markets 369
Three Theories of Expectation Formation 370
Markov Expectations 371

Adaptive Expectations 372
Rational Expectations 379
A Critique of the Views of Expectations Formation 390
Efficient Markets 393
The Efficient Markets Hypothesis 393

Rational Expectations and Efficient Markets 395
Versions of the Efficient Markets Hypothesis 396
Weak-Form Market Efficiency 396
Efficient Markets and Expectations 399
Conclusion 400
Box 12.1 Inside Money Error in the Inflationary Expectations

of "Experts" 380
Box 12.2 Inside Money The Importance of Using the Correct

Model When Forming Rational Expectations 387
Contents xv
Box 123 The Data Bank The Random Nature of Stock Returns,
Interest Rates, and Inflation 397
PART FOUR The Creation of Money 405
CHAPTER 13 The Federal Reserve System: History, Modern

Structure, and Policy Tools 406
A Brief History of Central Banking in the United States 407
Banking in the United States Before 1791 407

The First U.S. Central Bank: First Bank of the United States 408
The Second U.S. Central Bank: Second Bank of the United States 409
The Absence of a Central Bank: 1836-1913 409
The Third U.S. Central Bank: The Federal Reserve System 412
The Structure and Role of the Modern Federal

Reserve System 417
The Institutional Structure 417

The Roles of the Fed 422
Tools of Monetary Policymakers 424
Open Market Operations 424

Reserve Requirements 428
Discount Window Policy 432
Other Monetary Policy Tools 434
The Debate over Federal Reserve Independence 437
Conclusion 438
Box 13■ 1 Inside Money Declining Bond Prices and Bank Failures
During the Free Banking Era 411
Box 13.2 Inside Money Making the Fed More Responsive to the
Electorate: The Rebirth of an Old Idea 422
Box 133 The Data Bank Open Market Operations and Churning 430
Box 13.4 International Banking The Benefits of Central Bank

Independence: A Look at Inflation Around the World 439
CHAPTER 14 The Money Supply Process 446
The Balance Sheet of Banks and Multiple Deposit Creation 447
How a Bank Responds to a Change in Reserves from Within the

Banking System 448
How the Banking System Responds to Infusions and Contractions
of Reserves 455
XVI Contents
A General Model of Money Creation 465
The Monetary Base 466

The Complete Deposit Multiplier 467
The Complete Currency Multiplier 469
The Complete Money Multiplier 470
Determinants of the Money Supply 474
Federal Reserve Determinants of the Money Supply 476

Banking System Determinants of the Money Supply 477
The Public’s Determinants of the Money Supply 479
The Money Supply Curve 482
Exogenous Money Supply Curve 483

Endogenous Money Supply Curve 484
Conclusion 486
Box 14.1 The Data Bank The Balance Sheet of the Commercial
Banking Sector 449
Box 14.2 INSIDE Money Reserve Requirements of the United States 462
Box 14.3 Inside Money The M2 Multiplier 472
CHAPTER 15 The Demand for Money and Equilibrium in the

Money Market 489
The Classical View of Money Demand 490
The Simple Quantity Theory 490

The Cambridge Theory 493
Keynes's View of Money Demand 495
The Modified Cambridge Theory 496

The Inventory Approach to Money Demand 498
The Portfolio Approach to Money Demand 502
The Modern Quantity Theory of Money Demand 506
Factors that Influence the Opportunity Cost of Holding Money 508
Money Demand: An Historical Note and Current Thought 509
An Historical Note 509

Current Thought 510
Equilibrium in the Money Market 5 io
Equilibrium with an Exogenous Money Supply 512

Equilibrium with an Endogenous Money Supply 515
Contents xvii
Keynesian and Monetarist View on Money Demand 518
Using the Loanable Funds Model to Analyze Monetary Policy 520
Conclusion 522
Box 15.1 Interna tional Banking Money Demand in Mexico 494
Box 15.2 The Data Bank Velocity and Interest Rates 497
Box 15.3 Inside Money Currency Substitution 505
Box 15.4 The Data Bank Money Demand in the United States 511
part five Money and the Macro Economy 525
chapter 16 A Simple Macroeconomic Model 526
Aggregate Demand 521
Why the Aggregate Demand Curve Slopes Downward 528

The Components of Aggregate Demand and Their Determinants 530
Determinants of Consumption Demand 530
Adding Up the Components to Obtain Aggregate Demand 540
Why Aggregate Demand Does Not Depend on Income 541
Changes in Aggregate Demand 543
Short Run A ggrega te Suppl y 546
Why the Short-Run Aggregate Supply Curve Slopes Upward 546

Determinants of Short-Run Aggregate Supply 549
Short Run Macroeconomic Equilibrium 550
Achieving Equilibrium 553

Changes in Equilibrium 553
Vertical Long-Run Aggregate Supply and Long-Run Equilibrium 556
Fiscal Policy and the Government Budget Restraint 561
What is the Government Budget Restraint 561

Government Purchases Financed by Taxes 562
Government Purchases Financed by Borrowing: Deficit Finance 567
Government Purchases Financed by Money Creation 570
Monetary Policy: Open Market Operations 573
Conclusion 576
Box 16.1 The Data Bank Components of Aggregate Demand 531

XVIII Contents
Box 16.2 The Data Bank Components of Government Spending, 1962,

1977, and 1992 539
Inside Money Income Taxes, Indentation, and Inflation 564
Box 16.4 Inside Money Did Indexation Cause the Budget Deficits
of the 1980's? 571
CHAPTER 17 Open Economy Macroeconomics 586
The Balance of Payments 587
The Current Account 588

The Capital Account 590
Recent U.S. Balance of Payments Accounts 591
The Demand for Net Exports 593
Open Economy Macroeconomic Equilibrium 601
Aggregate Demand in an Open Economy 601

Short-Run Aggregate Supply in an Open Economy 603
Long-Run Aggregate Supply in an Open Economy 604
Fixed Exchange Rates, Flexible Exchange Rates, and Policy

Effectiveness 607
Fiscal Policy in an Open Economy 608
Tax and Tariff-Financed Increases in Government Purchases 608

Government Purchases Financed by Borrowing: Deficit Finance 611
Government Purchases Financed by Money Creation 614
The Twin Deficits: The Budget and Current Account Deficits 614
Investment, Saving, and the Budget Deficit 615

The Supply of Loanable Funds in an Open Economy 617
Conclusion 622
Box 17.1 The Data Bank Net Exports and the Exchange Rate 598
Box 17.2 Inside Money The Impact of NIFTIER on Output and

Employment 600
Box 17.3 Interna tional Banking Fixed Versus Flexible Exchange Rates:
The European Experience 609
CHAPTER 18 Money and Economic Activity: A Look at

the Evidence 625
The Long-Run Relationship Among Money, Prices,
and Real Output 625
Contents xix
Theoretical Relationships Between Money and Economic Activity 625

U.S. Evidence on the Long-Run Relationships Between Money and Economic
Activity 628
International Evidence on the Long-Run Relationships Between Money
and Economic Activity 632
The Short-Run Relationship Among Money; Prices,

and Real Output 638
Theoretical Relationships Between Money and Economic Activity 638

Empirical Evidence on the Short-Run Relationship Between Money
and Economic Activity 639
Reconciling the Long-Run and Short-Run Results 643
Inflation, Money Growth, and Interest Rates 650
Real Money Balances, Velocity, and Interest Rates 654
Real Money Balances, Real Income, and the Interest Rate 654
Inflation, Velocity, and Real Money Balances 657
Conclusion 660
Box 18.1 International Banking Money Growth, gdp Growth, and

Inflation in the United Kingdom: 1 636
Box 18.2 Inside Money Does Money Create Income, or Does

Income Create Money? 644
Box 183 International Banking Money Growth, gdp Growth, and

Inflation in the United Kingdom: 2 649
part six Special Issues in Monetary Economics j565___ __
chapter 19 Art Introduction to Monetary Policy 666
Targets and Instruments 666
Monetary Policy Goals or Targets 668

Monetary Policy Instruments 669
The Number of Independent Instruments and Targets 671
The Link Between Money and Policy Targets 674
Intermediate Targets 679
A Money Aggregate as an Intermediate Target 681

An Interest Rate as an Intermediate Target 682
The Choice Between Interest Rates and Money as an Intermediate
Target 684
Other Suggested Intermediate Targets 686
XX Contents
Federal Reserve Operating Procedures 692
Federal Funds Rate Targeting 694

Nonborrowed Reserves Targeting 695
Borrowed Reserves Targeting 696
The 1990’s: A Return to Interest Rate Targeting 696
Problems in Monetary Policymaking 697
Lags 700
Instrument Instability 703
Inaccurate Macroeconomics Models 704
Conflicting Goals 704
Conflict Among Policymakers 704
Conclusion 705
Box 19.1 International Banking Monetary Policy Gone Awry:

Hyperinflation in Yugoslavia 677
Box 19.2 The Data Bank Does the Fed Hit Its Money Targets? 683
Box 19.3 Inside Money P-Star: A New Target for Monetary Policy? 698
CHAPTER 20 Monetary Policy: Rules; Discretion,

and Games 708
Should the Fed Pursue an Active or Passive Monetary Policy? 708
The Case for Active Monetary Policy 709

The Case for Passive Monetary Policy 717
So Who’s Right? 721
Should Monetary Policy Be Guided by Rules or by Discretion? 723
The Case for Discretionary Monetary Policy 723

The Case for Monetary Policy Rules 724
Why Does the Fed Have Discretion? 729
Passive Versus Active Rules 733
A Change in Aggregate Demand 733

A Change in Aggregate Supply 735
The Pros and Cons of Activist Rules 737
Anticipated Versus Unanticipated Changes in the Money Supply 738
Unanticipated Changes in the Money Supply 739

Anticipated Changes in the Money Supply 740
Relevance for Monetary Policy 742
Conclusion 742
Contents xxi
Box 20.1 The Data Bank The Path of the Price Level and
Output Over Time 710
Box 20.2 Inside Money Inflation and Politics 718
Box 20.3 Interna tional Banking International Policy Coordination 203
CHAPTER 21 Government Spending and Finance:

The Deficit Problem 746
An Historical Look at Government Spending and Its Finance 747
The Federal Government Budget 747

State and Local Budgets 750
The Aggregate Government Budget 751
The Federal Debt 754
The Deficit and the Growing Debt: Are They Bad for the Economy? 756
Deficits Will Lead to Inflation 757

A Burden on Future Generations? 760
Crowding Out 763
Ricardian Equivalence and Fiscal Policy 765
Is the Ricardian Equivalence Idea Correct? 766

The Impact of the Ricardian Equivalence Idea on Our Macroeconomic
Model 767
Can the Federal Deficit Continue to Grow? 770
The Primary versus the Total Budget Deficit 770

The Impact of Growing Deficits: An Illustrative Example 771
The Bottom Line for the United States 773
Conclusion 111
Box 21.1 Interna tional Banking The Government Budget Constraint

in Action: The Case of Mexico 758
Box 21.2 The Data Bank The Government Budget Constraint in
Action: The United States 761
Box 21.3 Inside Money Is Debt a Problem? 774
Glossary G-l
Solutions S-l
Index 1-1
,
■
Preface
As a result of banking deregulation over the past decade, banks today are
like most firms in the economy in that they must set their own interest rates
and vigorously compete with one another for depositors and loan customers.
These recent changes in the U.S. banking system led us to write a money
and banking text with a firm foundation in basic economic theory. Our book
uses economic tools like supply and demand analysis, the theory of the firm,
the economics of information, game theory, rational expectations, and
AD/AS to analyze the workings of the banking system and the effects of
Fed policies on it.
When we were undergraduate students of money and banking in the late
1970s, the U.S. banking system was highly regulated and banks had little
discretion in setting interest rates. Consequently, texts at the time were
institutional in nature and T-accounts were the primary tool used to look at
bank operations. We left wondering why our money and banking texts didn’t
reflect what we learned in our introductory economics courses about supply
and demand and firm behavior. We also remember times when the facts in
the text were no longer accurate, due to daily changes in the rules and
regulations of the banking system. These changes made it extremely difficult
for us to learn, and for our instructors to teach, a coherent framework for
analyzing the banking industry. We believe the situation is much better
today.
Our use of basic economic theory that is accessible to undergraduates
means that instructors no longer need to center their lectures on institutional
details of the banking system—details that change as frequently as the
weather. Each chapter in this book develops timeless economic models—
models we use to analyze the economic consequences of past and present
regulations on the banking system and the overall economy. A quick glance
through the book reveals that we rely heavily on simple graphs and economic
models to analyze the real-world workings of the banking system, rather
than using the accounting or institutional approach. We believe this makes
using our money and banking text easier, more fun, and of more lasting
value than memorizing what a competing institutional text says is the most
recent piece of banking legislation. It is inevitable that some institutional
changes will occur after this book is published, but the theoretical foundation
and real-world focus of this book gives students the tools to analyze them
too.
XXIil
XXIV Preface
Content and Organization
Since banks are firms, we believe a modem money and banking course
should contain elements of both microeconomics and macroeconomics. Is¬
sues such as how banks set interest rates, attract deposits, or ration credit,
and how the banking industry is subject to the pitfalls of adverse selection
and moral hazard are all microeconomic in nature. Accordingly, Chapters 5
and 6 provide self-contained models of the microeconomic decisions of
individual banks, while Chapter 7 looks at the implications of banking
regulations for the entire banking industry. The remainder of the book pre¬
sents models of the more traditional financial and macroeconomic issues.
Those instructors who choose not to concentrate on the microeconomics of
banking can skip Chapters 5, 6, and 7 without loss of continuity.
A noticeable feature of our book is that it contains 21 chapters instead
of the 30 or so that many other money and banking texts include. Recogniz¬
ing that few instructors actually have time to cover the 30 chapters and based
on reviewer comments, we grouped related topics into fewer chapters. Our
hope is that more complete, self-contained chapters will make it easier to
use our book. Students gain a deeper, less fragmented sense of money,
banking, and the economy overall. Each chapter includes extensive cross
references to point readers to related topics elsewhere in the text.
We divide Money and Banking into six parts. Part I Introduction to
Money and Banking, Chapters 1-3, provides a broad overview of money
and banking and some of the key concepts and terms that underlie analyses
throughout the book. These chapters explain the rudiments of money and
various financial instruments, discuss how they evolved, and provide prac¬
tical definitions of the money multipliers, Ml, M2, M3, and L. They also
provide an overview of money creation, present value analysis, and the
relation between monetary growth and inflation.
Part II The Microeconomics of Banking and Financial Markets,
Chapters 4-7, begins with a simple supply and demand framework for
loanable funds that builds on what students learned in their introductory
course. Chapters 5, 6, and 7 show how individual banks operate much like
other firms in the economy, using the tools of microeconomics and industrial
organization to look at banks and the banking industry. Again, Chapters 5,
6, and 7 are self-contained, giving instructors who stress macroeconomic
issues the flexibility to omit them if desired.
Part III The Valuation of Financial Assets, Chapters 8-12, first studies
the time value of money in Chapter 8. Chapters 9 and 10 examine the impact
of risk on asset prices and explain the term structure of interest rates. Chapter
11 develops short and long-run models of foreign exchange rate determi¬
nation setting the state for the open economy material in Part V. Chapter 12
presents basic models of expectation formation like rational expectations.
These chapters, particularly Chapters 11 and 12, begin a shift toward the
macroeconomic issues that are covered in Chapters 13-21 of the text.
Parts IV and V, The Creation of Money, Chapters 13-15, and Money
Preface XXV
and the Macro Economy, Chapters 16-18, present the core material of
money and the macroeconomy. While the introductory material in Part I
acquaints students with basic macroeconomic terminology, these chapters
develop simple macroeconomic models that help them to understand the
macro economy more fully. Part IV begins with a brief overview of the
Federal Reserve System in Chapter 13, then turns to a discussion of the
money supply process (Chapter 14) and the demand for money (Chapter 15).
Part V focuses on key macroeconomic models: Chapter 16 develops a closed
economy macro model and discusses both the workings of the model and
the impact of monetary and fiscal policy. Based on reviewer comments, the
main text of Chapter 16 relies exclusively on an AD/AS model, while an
Appendix includes the IS/LM framework for those instructors who choose
to integrate it into their courses. Chapter 17 presents the open economy
model as an extension of the closed economy model that allows an endog¬
enous determination of the exchange rate and net exports. Chapter 18 closes
this action with a presentation of some of the empirical evidence on mac¬
roeconomic propositions linking money to macroeconomic variables.
Part VI Special Issues in Monetary Economics, Chapters 19-21, pro¬
vides more detailed, applied treatments of monetary issues for those instruc¬
tors who have the time (or preference) to cover one or more of these chapters.
Chapter 19 looks at the theory of monetary policy, including a discussion
of targets and indicators, while Chapter 20 looks at monetary policy games
and the rules versus discretion debate. The material is presented at a level
accessible even to students who have not been previously exposed to game
theory. Finally, Chapter 21 is devoted exclusively to the government budget
constraint and provides an economic analysis of the dilemma facing Wash¬
ington today regarding how to finance the growing federal budget deficit.
While many of the issues covered in Part VI are treated elsewhere in the
book, the comments of numerous reviewers suggest that many instructors
want to cover some of these topics more deeply, as in these chapters.
Special Features
This text has many features designed to enhance the learning and teaching
experience.
Boxed Examples. Each chapter includes boxes that use real-world

examples to show how the theory developed in the text relates to the actual
workings of the economy. These boxes are organized into three broad cat¬
egories. The Data Bank boxes examine published economic and financial
data through the lens of the models used in the text (see The Data Bank 4.1:
Actual and Expected Rates of Inflation). Inside Money boxes focus on
special topics or issues related to money and banking that most students will
find relevant (see Inside Money 3.3: 15-Year Versus 30-Year Mortgages).
Finally, International Banking boxes acquaint students with money and
XXVI Preface
banking around the world (see International Banking 5.2: Concentration

Ratios in Banking).
Summary Exercises. Each chapter contains numerous summary ex¬

ercises with answers immediately following. This lets students verify that
they have mastered material in a section before reading on and, based on
our own experience, considerably reduces the frequency with which students
need to find the time for one-on-one consultation with their instructors.
End-of-Chapter Questions and Problems. The end of each

chapter contains numerous questions and problems. Selected answers appear
in the back of the book.
Key Terms. Each chapter includes an extensive list of key terms, all of
which are defined in the end-of-text glossary.
Selections for Further Reading. Each chapter includes a list of

articles accessible to students and instructors who desire further information
about topics included in the chapter.
Glossary and Index. A glossary defines all key terms highlighted in

the text, and an extensive index aids students and instructors in cross-refer¬
encing material in the book.
Support Package
To help both students and teachers using this textbook, Houghton Mifflin
offers an accompanying support package. The Study Guide includes chapter
synopses and a wide range of practice questions and exercises with answers
and helpful hints. The Instructor’s Resource Manual/Test Bank contains
sample course outlines and, for each chapter, helpful classroom hints, a
summary of key concepts, and solutions to text questions and problems. The
Instructor’s Resource Manual/Test Bank also contains over 800 test ques¬
tions, including multiple choice, essay, and discussion questions, many of
which have accompanying graphs. These test questions are also available in
a Computerized Test Bank for IBM-PC and compatible microcomputers.
A CKNO WLEDGMENTS
Our product has been improved by the feedback and thoughtful comments
of numerous students, colleagues, and professionals at Houghton Mifflin.
Denise Clinton, the economics Sponsoring Editor of Houghton Mifflin, man¬
aged the project and was instrumental in molding our initial view of this
Preface XXVII
text into what it is today. Her support and constructive criticism throughout
the development process, but most importantly her vision and her patience
with hardheaded authors, is deeply appreciated.
We owe a special debt to our two Basic Book editors at Houghton
Mifflin—Tom Mancuso, who helped us during the early stages of the book,
and Karla Paschkis, who helped us enliven the manuscript and forced us to
better organize our thoughts. Our Senior Project Editor, Susan Westendorf,
did a superb job managing production and those responsible for producing
the final manuscript. Countless others at Houghton Mifflin—Greg Tobin,
Sue Wame, and Mike Ginley, to name just a few—challenged and encour¬
aged us during day long development meetings in Boston. Thanks, folks.
Numerous reviewers across the country provided feedback on our man¬
uscript, not only correcting errors but helping us better see their own needs
as well as the needs of the market.
These reviewers included:
Thomas Bonsor Dennis Fixler Daniel Houser

Eastern Washington University George Washington University University of Minnesota
& Bureau of Labor Statistics
Kathleen Brook Harry Huizinga
New Mexico State University Peter Frevert Stanford University
University of Kansas
Colleen Callahan Charles W. Johnston
Lehigh University Lance Girton University of Michigan—Flint
University of Utah
Arthur E. Kartman
Mitch Charkiewicz
John Grobey San Diego State University
Central Connecticut State
Humboldt State University
University Richard H. Keehn
Nell Gullet University of Wisconsin—
David Craig
University of Tennessee at Parkside
Westark Community College
Martin
and Arkansas State University James Kelley
Scott D. Hakala Texas A&M University
Eleanor Craig Southern Methodist University
University of Delaware Elizabeth Sawyer Kelly
Bassam E. Harik University of Wisconsin—
Mario Crucini Western Michigan University Madison
Ohio State University
Gay Hatfield Jane Knodell
A1 Culver University of Mississippi University of Vermont
California State University at
Thomas Havrilesky James Lynch
Chico
Duke University Robert Morris College
Andrew Dane Emily Hoffnar Rich MacDonald
Angelo State University University of North Texas St. Cloud State University
Stan Eakins Lora Holcombe David MacPherson
East Caroline University Florida State University Florida State University
Peter J. Ferderer Randall G. Holcombe W. Douglas McMillan

Clark University Florida State University Louisiana State University
XXVIII Preface
Robert M. Mulligan Calvin Siebert Thom Thurston

Providence College University of Iowa City University of New York
K
Charles Murray Harinder Singh Anne P. Villamil

Northeast Missouri State San Diego State University University of I lino is at
University Urbana—Champaign
Neil Skaggs
Douglas K. Pearce Illinois State University John C. Wassom
North Carolina State University Western Kentucky State
Paula Smith University
Malcolm Robinson
University of Central Oklahoma
University of Cincinnati Eugene White
John Soper Rutgers University
Charles Sawyer
University of Southern John Carroll University James Wible
Mississippi Robert Sorenson University of New Hampshire
Amin U. Sarkar University of Missouri—St. Raymond Zelder

S UNY—Fredonia Louis Western Michigan University
Richard T. Selden Frederick Thum

University of Virginia University of Texas
To all of you, thanks.

We owe our special thanks to the text’s dedicated accuracy reviewers:
Daniel Houser, University of Minnesota; James Kelley, Texas A&M Uni¬
versity; and Thom Thurston, City University of New York.
We are also grateful to John Wells and Dennis Powers for the careful
job they did checking problems, collecting data, and proofing our work at
various stages of the project. Jim Kelley also assisted with the manuscript,
and our assistants, Sue Bryant and Tamara Ariens helped make this project
manageable. Finally, we thank our children and wives (actually, we have
only one each) for putting up with busy fathers and husbands who have yet
to learn how not to overcommit.
MRB University Park, Pennsylvania

DWJ College Station, Texas
Money, Banking, and
Financial Markets:
4,
.
.
,
PART ONE
Introduction to Money
and Banking
1
Money and Banks
2
Financial Markets, Financial Institutions and Instruments, and Money
3
Money, Inflation, and Interest
CHAPTER
Money and Banks
his book is about money and banking, two subjects closely intertwined
with our daily lives and with the everyday functioning of our economy. On
a typical day, you encounter money and banks in many forms—from the
obvious contact made when you use dollar bills to buy lunch or “go to the
bank’ ’ to withdraw funds to the almost invisible contact via the investment
of your school’s endowment. When you write a check, withdraw funds at
the corner ATM, or use a credit card to purchase this book, you interact with
the banking system. If you work, you may be paid in the form of a check
or a direct deposit into your bank account. If your parents send you money
from home, it may arrive as cash delivered in person, as a check mailed
directly to you, as a check deposited directly into your account, or as a wire
transfer. If you have some savings, you may invest them in a savings account,
a money market mutual fund, or various forms of bonds or stocks. All of
these situations are interactions between you and the banking and financial
system.
Taking the larger view, money and banks, and the financial system as a
whole, play a vital role in the national economy. The total quantity of money
in the economy and the rate at which that quantity grows over time have
important implications for interest rates, inflation rates, and the economy’s
overall functioning. In later chapters, we will see how the federal government
has appointed a central bank to regulate and control the quantity of money
and oversee the health of the banking system and, to some extent, of the
entire financial system. We will also see how the central bank accomplishes
its tasks and how well it achieves its goals of low inflation rates and full
employment.
At this point, you may have little knowledge of how banks work or what
money really is. At the end of this course, however, you will have a much
greater understanding of the roles money and banks play, both on the indi¬
vidual level and in the economy overall. This chapter begins with an over¬
view of these roles and discusses the evolution of money and banks as we
know them today. We will see that money is not simply bills and coins; it
is anything widely used to pay for goods and services. Throughout the history
of humankind, numerous things have served as money, including shells,
stones, beads, sacks of grain, gold, cigarettes, paper bills, and checks. We
will also see that money is not synonymous with “wealth’’ or “income.”
Money 3
An individual’s wealth refers to the stock of assets the individual owns, less
his or her debts. For example, your wealth includes the total value of your
holdings of real estate, stocks and bonds, cash on hand, money deposited in
banks or other financial institutions, cars, and even textbooks, minus the
amount you owe to others. In contrast, income refers to a flow of earnings
over some given time interval, say, a week, month, or year. Your wealth is
like an “inventory” of everything you own minus debts owed to others,
while your income represents the “change” in the inventory that occurs
during a given time period.
With these concepts in mind, we will now look more closely at the
nature and role of money in our economy and at the origin of money and
banks.
Money
Money is anything generally accepted as a medium of exchange. A medium

of exchange is virtually anything used to pay for goods and services or settle
debts. Thus, the distinguishing feature of money is that society widely ac¬
cepts it to settle transactions. In the United States, money takes many forms.
Obviously you use the dollar bills (more technically, Federal Reserve notes)
and coins in your pocket to pay for things like compact discs and restaurant
meals. Thus, Federal Reserve notes and coins minted by the U.S. Treasury
are money. In popular use, the term money is often a synonym for these two
media of exchange. Coins and dollar bills, however, are called currency and
comprise only part of the money supply. As Box 1.1 shows, traveler’s checks
and checks written against checking accounts at banks, savings and loans,
and credit unions are also commonly accepted as payment and are therefore
money.
Other financial assets, such as savings account balances, are sometimes
considered money. However, savings account balances clearly are not a
medium of exchange, since it is difficult to literally exchange your savings
account balance for a meal at a restaurant. Instead, you must first convert
your savings account into another asset, such as currency or a check, and
then exchange the currency or check for a meal. Some assets—your house,
for instance—cannot be quickly and easily converted into a medium of
exchange. It often takes substantial real estate commissions and time to
convert a house into currency or checkable deposits. Your automatic teller
machine (ATM) card is not money either, since you cannot directly purchase
a meal by handing over the card. You can, however, use your ATM card to
obtain money from your checking or savings account. Thus, ATM cards are
instruments that allow the user to convert checking or savings account bal¬
ances into currency.
Liquidity is the term economists use to describe how cheaply and easily
an asset may be converted into a medium of exchange. Savings account
4 Chapter 1 Money and Banks
The Data Bank Box 1.1
Components of the Money Supply
If you wished for all the money in the U.S. econ¬ Total U.S. Money Supply (Ml) in January 1993:
$1.04 Trillion
omy and your wish came true, what would you
get? The following chart, which is based on a
definition of the money supply the Federal Re¬ Other Checkable Deposits
($392.6 Billion)
serve System calls Ml, provides the answer:
Si .04 trillion. As you can see, money consists of
much more than the dollar bills and coins in
your pocket. The Ml money supply includes cur¬
rency in the hands of the public, demand depos¬
its (checking account balances at commercial
Traveler's Checks
banks), other checkable deposits (money market
($7.8 Billion)
deposit accounts and various checkable deposits
at savings and loans, credit unions, and mutual
savings banks), and traveler's checks. There are
also other official Federal Reserve System defini¬
tions of the money supply, such as M2 and M3,
which we discuss in more detail in Chapter 2.
Currency Demand Deposits
Of the roughly $1 trillion of Ml money in
($293.6 Billion) ($346.2 Billion)
the United States, only 28 percent is in the form
of currency. Demand deposits and other checka¬
ble deposits together comprise the lion's share
(71 percent).
Source for data: Board of Governors of the Federal Re¬

serve System, Federal Reserve Bulletin, various issues, and
Citibase electronic database.
balances are a liquid asset, since they can be quickly and easily converted
into a medium of exchange. Real estate, on the other hand, is not quickly
and easily convertible, and therefore is an illiquid asset. Dollars (i.e., cur¬
rency) and checkable deposits, the primary media of exchange in the United
States, are the most liquid of all assets.
Most dictionaries include variations of one or more of the following defi-

Exercise 1,1 I nitions of money: (1) portable pieces of metal used as a medium of exchange;
Money 5
(2) currency and coins; (3) wealth. Comment on which, if any, of these
definitions corresponds with an economist’s definition of money.
Answer: These definitions are all deficient to an economist. The first

definition identifies money as a medium of exchange but limits money to
coins. This seems to exclude paper currency as well as checking accounts.
The second definition mentions coins and currency but does not mention
checkable deposits. The third definition treats money as being synonymous
with wealth; for example, “He’s got a lot of money” is a popular way of
saying “He’s wealthy.” Economists, however, do not use money as a syn¬
onym for wealth.
Functions of Money
You now have an overview of what money is and recognize the distinctions
among money, income, and wealth. We now examine four roles of money
in an economy. As we will see, money serves as a medium of exchange, a
unit of account, a store of value, and a standard of deferred payment.
Medium of Exchange. As mentioned earlier, the primary function of

money in an economy is to serve as a medium of exchange. This simply
means that when you buy goods, services, or financial instruments (such as
stocks or bonds), you pay with money. In the United States, people almost
always use money in the form of currency or checks as the medium of
exchange. Barter—the exchange of goods and services without the use of
money—is a possible but rare method of making transactions.
It is hard to envision life without money as the medium of exchange.
Imagine a barter economy—an economy that has no money in which goods
are traded directly for other goods. If a baker wants shoes, it is not enough
to find a shoemaker; the baker must find a shoemaker willing to trade shoes
for bread. Barter transactions require a double coincidence of wants: Each
individual must have something the other desires. If this happens, exchange
will take place; the shoemaker will get bread, and the baker will get shoes.
But if the baker does not want shoes, or vice versa, no exchange will take
place. In this case, the two parties will have to continue to search for someone
who wants what they have for trade.1
Thus, we see that barter is highly inefficient, since the baker would have
to spend considerable time searching for someone willing to accept bread as
payment for some other good. This search time is a transactions cost: a
cost borne in making an exchange. When money is used as the medium of
exchange, the baker can use money to buy shoes from the shoemaker even
if the shoemaker does not want bread. The shoemaker, in turn, can use the
1 Of course, intermediate exchanges might occur in which the baker first trades for apples and then
trades for shoes. Indeed, when one good becomes an acceptable intermediate exchange in all
transactions, that good has become money!
money received to buy whatever he or she desires from a third party. In

effect, in a monetary economy all individuals find money useful (and thus
want money); thus, money satisfies the double coincidence of wants. There¬
fore, one main advantage of a monetary economy—an economy that uses
money as a medium of exchange—is that it eliminates the problem of finding
a double coincidence of wants; that is, it reduces the transactions costs of
exchange. As we will see, monetary economies have other advantages.
Unit of Account. Money also serves as a unit of account: The values

of goods and services are stated in units of money, just as time is measured
in minutes and distance in feet. Accountants record the revenues and costs
of companies like AT&T in terms of money. Similarly, individuals budget
their expenditure and income flows in terms of money.
In a monetary economy, the medium of exchange nearly always also
serves as the unit of account. Because the medium of exchange is common
to virtually all transactions, it is convenient to state the prices of goods and
services in terms of the medium of exchange. In the United States, the unit
of account is the U.S. dollar, which is also the medium of exchange.
The use of money as the unit of account reduces the amount of infor¬
mation individuals need to make purchase decisions. In a monetary economy,
prices are quoted in terms of the unit of account (be it dollars, yen, or shells).
If there are 1,000 goods for sale, there are 1,000 prices—one for each good.
Each price specifies how many units of money must be given up to receive
one unit of each good. In a primitive monetary economy, one loaf of bread
might cost 1 shell, one cow might cost 100 shells, and so on. In the modem
U.S. economy, the U.S. dollar is the unit of account, and these prices are
quoted in dollars: $1.59 for one loaf of bread, $1,250 for one cow.
In the absence of a common unit of account, there would be more prices
than goods in the economy! To see this, imagine you live in a simple barter
economy with only four goods: apples, oranges, shoes, and bread. Apple
sellers would have to quote three different prices. One price tag would
indicate how many oranges it takes to buy an apple, a second price tag would
state how many pairs of shoes it takes to buy an apple, and a third price tag
would reveal how many loaves of bread it takes to buy an apple. Similarly,
orange sellers would have to specify how many apples, shoes, or loaves of
bread it takes to receive an orange. Of course, if millions of goods were
available, as in our economy, there would be millions of price tags to put
on each item.2 Imagine how costly it would be for you to figure out whether
you could afford to buy an item when each item had millions of price tags!
2 If there are n goods, the number of prices in a barter economy is given by the formula n X
(.n — l)/2. In our four-good economy, there are six prices. If n = 1,000, the number of prices is
499,500.
Money 7
Fortunately, we live in a monetary economy, and we use money as a

unit of account. All prices are stated in terms of dollars, and thus each good
has a single price tag. The presence of money as a unit of account reduces
the amount of price information you need to buy goods in the marketplace,
and this too reduces the transaction’s costs associated with exchange.
Store of Value. Money also serves as a store of value: It is a means

of storing today’s purchasing power to purchase, say, a house or a car
tomorrow. In the absence of money or other assets as a store of value,
individuals and companies would have to maintain stocks of goods to use
to trade in the future. This approach would be inefficient for two reasons.
First, some commodities, like fruit and milk, are perishable and would be of
little or no value if stored for future use. Second, even when a commodity
is not perishable—a car, for instance—it can be very costly to use it to store
value over time. General Motors would find it very costly to maintain in¬
ventories of extra cars to use to pay their workers. GM workers, in turn,
would find it useless to be paid in engine parts or transmissions, for they
would have a difficult time finding someone willing to accept an engine part
or a transmission as payment for food or housing.
The absence of money as a store of value would also lead people to
hoard a variety of goods—even goods they did not personally wish to
consume (liver or licorice, perhaps?)—in the hope of locating another person
who offered something they wanted and required one of those goods as
payment. In contrast, in a monetary economy, people hold inventories of
money—not transmissions, liver, or licorice—to use to pay for goods and
labor services.
Of course, money is not the only store of value. Indeed, many other
assets—savings accounts, stocks, and bonds, to name just a few—are often
better stores of value than money. These assets pay interest or dividends,
whereas currency and many checkable deposits do not. Thus, while money
provides a convenient store of purchasing power, it is not wise to use money
as a store of value over long periods of time. However, money is unmatched
by other assets in its liquidity, which gives it an advantage over other assets
as a temporary store of value.
Standard of Deferred Payment. Finally, money serves as a stan¬

dard of deferred payment; that is, a payment that is deferred to the future
is usually stated as a sum of money. For example, if you owe money to a
friend and plan to pay it back in a week, you usually state the amount you
owe in money terms, such as $5. In the absence of money, you would have
to plan on making payment in terms of some other good. Having a common
standard for deferred payments, which is the same as the medium of
exchange and the unit of account makes it relatively easy to determine
exactly how much a deferred payment will be. There are efficiencies in
thinking of payments today and payments tomorrow or next year in terms
of a common item—money. However, we will see that just as money is a

store of value, but not the best store of value, money is a standard of deferred
payment, but not necessarily the best standard for all purposes.
The Origin of Money

The origins of money are little known. Historians believe money originated
as religious objects of value, which ultimately evolved into a medium of
exchange. By the ancient Babylonian, Greek, and Roman civilizations, mon¬
etary systems included coins.
After the fall of Rome, the medieval economies of Europe moved toward
a barter system. The feudal system was built largely on barter, with debts
between lord and vassal paid in terms of goods and/or services—most fre¬
quently labor services. Thus, a noble would owe military service to his lord,
and a serf would owe labor to his lord. Taxes were often collected in the
form of agricultural products or labor services.
However, barter was neither a permanent nor a worldwide phenomenon.
Indeed, in Byzantium (now Constantinople) the Eastern Roman Empire
survived until the eleventh century and with it the nomisa gold coin, which
was the standard currency of the Mediterranean world. In Europe, the feudal
system and the use of barter eventually diminished, and a monetary economy
gradually reemerged. By the eleventh century, there was a revival of trade
and a renewed growth of towns, and gold emerged as a medium of exchange.
After William the Conqueror won the Battle of Hastings in 1066 and became
ruler of England, he allowed only royal coinage to be used in his kingdom.
The presence of a monetary economy in Europe was not without con¬
troversy. Phillip IV allegedly debased the French currency during his rule
(1285-1314) by minting coins that contained a lower amount of gold or
silver than promised and substituting small amounts of lead to maintain the
weight and feel of the declared quantity of precious metal.3 Despite episodes
such as this, however, Europe gradually moved closer to the monetary econ¬
omy of today.
Summary Credit cards are often used to pay for items purchased at retailing outlets.
Exercise 1.2 Are credit cards “money”?
Answer: Believe it or not, this is a controversial question. To answer,

note that a credit card is not money; you do not give someone your plastic
card in exchange for a good. Instead, a credit card is a money substitute.
3 Debasement of coinage meant that the monarch could issue more coins from a given supply of
gold or silver and by so doing could buy more goods and services from the general public than
would have been possible otherwise. Thus, debasement was an attempt by Phillip IV to tax the
general public without their knowledge.
Money 9
Using the card is taking out a loan from a preapproved line of credit. The
merchant receives a signed voucher from the buyer that can be exchanged
for money. If you paid the merchant with dollar bills or coins, the merchant
would directly receive money and could instantly use the bills or coins as a
medium of exchange in another transaction. In contrast, a credit card voucher
gives the merchant the right to collect a specified amount of money from
the credit card company at the end of the month. Thus, the credit card
voucher the merchant receives is an IOU that cannot be quickly converted
into a medium of exchange to use in another transaction. For this reason,
economists typically do not consider credit cards to be money.
Types of Money
Historically two types of money have been used as a medium of exchange.
The first type was commodity money—money backed by stores of a com¬
modity such as religious relics. Much later, a second type of money emerged:
fiat money. Marco Polo observed the Chinese using fiat money during his
famous travels in the thirteenth century. Fiat money is money that is not
“backed” by a commodity such as animal pelts or gold but is valuable as
money because of government pronouncement or fiat. Fiat is a Latin word
that means literally the pronouncement “let it be done.”
Commodity Money. The first type of money used by civilization-

religious relics—was a form of commodity money. Commodity money is
a physical commodity that is used as money but also has alternative, non¬
monetary uses. More recent examples of commodity money are gold and
silver coins. Gold and silver are used in jewelry and for other decorative
purposes and thus have value independent of their use as money. But when
they circulate as money, they are valuable as a medium of exchange, that is,
to purchase other commodities.
In addition to religious relics, gold and silver, numerous other goods
have served as commodity money in human history: live animals, sacks of
corn, and even huge stone wheels (See Box 1.2). The common characteristic
among all forms of commodity money is that something valuable to society
for its physical properties is chosen to perform a second function as the
medium of exchange.
A type of commodity money in which the commodity itself circulates
as money is called full-bodied money. As the term implies, the monetary
value of full-bodied money exactly equals its nonmonetary value. Animal
pelts, for example, were full-bodied money.
The most recent example of full-bodied money is the gold coins that
were minted by the U.S. Treasury until 1914. The value of the gold in each
coin was equal to the value of the coin as money. For instance, if the price
of gold was $1 per ounce, a $1 gold coin contained one ounce of gold. The
coin could then be used to purchase $1 worth of candy, or it could be melted
to extract one ounce of gold for use in jewelry.
International Banking Box 1.2
Money on the Yap Island

and Gold Transfers among Nations
The ancient inhabitants of the Yap island in the limited to the Yap Island; even today most gold
Pacific Ocean had a most peculiar full-bodied transfers among nations are conducted in un¬
medium of exchange: a large stone wheel called derground vaults such as those at the Federal
a fei. The fei was sculpted from limestone ob¬ Reserve Bank of New York. When a transaction
tained on an island hundreds of miles from Yap. that requires the transfer of gold from, say, the
The value of a fei depended in part on its size; it Netherlands to Germany occurs, the gold is not
had a diameter that often exceeded the height actually shipped from one country to the other.
of an average man. Smaller fei were carried on Instead, it is merely moved from the vault that
wooden poles placed through the hole in the stores the Netherlands' gold to the vault that
middle, but large fei were seldom actually stores Germany's gold. This journey of a few
moved. Instead, when an islander made a pur¬ meters helps keep the ownership record intact
chase that required the payment of a huge fei, but is otherwise similar to the practice of the
the fei would remain in a fixed location and the Yap islanders, who merely transferred the ac¬
new owner would merely accept the acknowl¬ knowledgment of ownership rather than the
edgment of ownership as payment for the physical location of large fei.
goods.
You might find it odd that the fei served as
Sources: Norman Angell, The Story of Money (New York:
a medium of exchange without physically Garden City Publishing Company, 1929), 88-89; Journal
changing hands. This peculiarity, however, is not of Political Economy, 90 (August 1982), back cover.
In contrast to full-bodied money, representative full-bodied money is

paper money that represents a claim to a specific quantity of some commod¬
ity. The actual commodity is held as an inventory in a bank vault or other
depository and does not circulate. During the early part of this century, the
U.S. Treasury issued representative full-bodied money called gold certifi¬
cates. Gold certificates were pieces of paper that could be redeemed for a
specified amount of gold. For instance, if the price of gold was $35 per
ounce, a $1 gold certificate represented a claim to 1/35 ounces of gold.
Anyone holding such a gold certificate could convert it into 1/35 ounces of
gold on demand.
The advantage of representative full-bodied money is that the commodity
itself does not have to circulate; only the paper claim to the commodity does.
The innovation of representative full-bodied money significantly reduced the
Money 11
need to carry around heavy bags of gold to use in exchange. Paper gold
certificates were much lighter and more practical. An interesting feature of
representative full-bodied money is that the actual item exchanged during
purchases (paper in the case of gold certificates) has negligible value as a
commodity.
Fiat Money. The use of representative full-bodied money accustomed

people to using paper money. The next innovation was the issuance of
unbacked money in the form of token coins and unbacked (fiat) paper money.
Fiat money has no value as a commodity and does not represent a claim to
any physical commodity. It is the type of money used in the United States
today. The United States went off the gold standard during the Great De¬
pression, and Federal Reserve notes (dollar bills) have been a fiat currency
ever since.
Token coins—the pennies, nickels, dimes, and quarters in your pocket—
resemble commodity money coins but contain less valuable metals such as
copper, nickel, or zinc instead of gold or silver. The physical value of these
coins as commodities is less than the value of what you can purchase with
them. In other words, if you went into the business of melting down quarters
and selling the resultant metal, you would end up in the poorhouse.
Fiat money also takes the form of unbacked paper money, usually fairly
small, rectangular pieces of paper with fancy engraving. Unlike representa¬
tive full-bodied money, however, fiat money is not backed by gold or silver
held in a depository. As a commodity, fiat money is worth only the paper it
is printed on. But fiat money is valuable as a medium of exchange, because
it can be used to purchase compact discs and other items. U.S. Federal
Reserve notes—the dollar bills in your pocket—are fiat money. Nothing is
held in a depository to back these paper dollars.
A final category of unbacked money is unbacked money issued by banks
and other financial institutions. The best-known example is checkable de¬
posits such as your checking account at your bank. You write checks and
businesses accept them as payment, even though the checks are backed by
no commodity. Instead, they are backed either by paper money held in the
bank vault or, more likely, by nothing tangible. We will say more about the
development of checks as money later in this chapter. As you will see, banks
can literally create money by creating checking account balances—balances
that are not even fully backed by holdings of paper dollars.
Does anything back fiat money? To answer this, consider gold coins.
You can always melt gold coins for their gold content. If the coins are full
bodied, the gold will equal their value as money. Even representative full-
bodied money, such as a gold certificate, is backed by the gold held in a
depository—gold that the owner of a gold certificate can obtain in exchange
for the certificate.
Fiat money, on the other hand, cannot be melted to yield an equivalent
value of metal, nor can it be exchanged for a commodity held in a depository.
Instead, fiat money is backed only by its general acceptance by society as

the medium of exchange. Without this acceptance, no one would value U.S.
dollar bills or accept them in exchange for goods and services. You accept
these Federal Reserve notes (or checks payable in Federal Reserve notes) as
payment for your work only because you are confident that your landlord
and your supermarket will accept these same notes in exchange for rent and
groceries. If all other people in society decided that Federal Reserve notes
were no longer acceptable as money, you would not accept them as payment
for your labor services.
Summary Determine whether the following episodes involve full-bodied money, rep¬
Exercise 13 j resentative full-bodied money, or fiat money: (a) During World War II,
prisoners of war used cigarettes as the medium of exchange and unit of
account for chocolate candy, cookies, and other items, (b) Yap islanders,
tired of hauling around stone wheels as the medium of exchange, decided to
use verbal acknowledgments of the ownership of stone wheels as the medium
of exchange, (c) The U.S. government decreed that its paper bills are legal
tender for all debts, both public and private.
I Answer: (a) In the POW camps, cigarettes were full-bodied commodity

money, (b) Verbal acknowledgments of ownership of stone wheels are rep¬
resentative full-bodied money, (c) The paper bills are fiat money, since they
are valuable as a medium of exchange only by government decree.
Physical Properties of Money

Our discussion of the roles of money provides numerous reasons money—
be it animal pelts, gold coins, or paper—tends to emerge in even the most
primitive societies. The numerous commodities that have served as money
throughout history, as well as the fiat money currently used in the United
States and other industrialized countries, all have certain desirable physical
properties. What makes a commodity a good choice to serve as money?
Ideally, money should be portable, divisible, durable, and of recognizable
value.
Portability. For ease of use in transactions, money should be portable.

The easier it is to carry around, the more effective it is as a medium of
exchange. Thus, commodity money should generally be a substance that is
valuable in small quantities. This explains why early humankind quickly
turned to gold and silver as forms of money. Even small quantities of these
rare metals are valuable enough to be used for large purchases. In contrast,
lead makes a poor choice as a medium of exchange, since large purchases
would require large amounts of lead.
Money 13
Divisibility. To permit transactions of various sizes, money should be

made of a commodity that is divisible into smaller units to facilitate making
“change.” Gold and silver also serve well in this respect, since small coins
can be minted to facilitate small transactions. In contrast, diamonds, which
are very portable (due to their high value per unit of weight) are not easily
divisible, making them a poor choice for a medium of exchange.
Money should be durable; that is, it should not wear out in

use and should not depreciate quickly when not in use. Gold and silver also
meet this criterion. Eggs, obviously, would serve poorly as a medium of
exchange.
Money should have easily recognizable value;

that is, it should be easy for people engaged in exchange to agree to the
value of the good used as money. Part of the motivation for coining gold
was to provide this property. In the days of gold coins, the coins were
stamped with a face value equal to the value in weight of the gold they
contained. In addition, each coin was stamped with the seal of the govern¬
ment or the “face” of the king as a guarantee of the coin’s authenticity and
weight. This practice made it difficult for unscrupulous individuals to snip
off portions of gold or manufacture “counterfeit” money. It also reduced
the uncertainty regarding whether a particular coin was indeed worth the
value stated by the individual wishing to exchange it, thus increasing the
acceptability of gold coins as payment.
A desire for recognizable value is also behind efforts to foil counter¬
feiting of fiat currency. Modern paper currency has complicated engraving,
rare papers and inks, and sometimes embedded metal strips. All of these
attributes make it easier to recognize genuine currency and more difficult to
produce counterfeit bills.
Box 1.3 describes a commodity that possessed all four physical prop¬
erties of money: cigarettes in a World War II POW camp.
Summary Do modern-day checks satisfy the desired properties of money?

Exercise 1.4
Obviously checks are easy to carry around (portable), can be
written in various amounts (divisible), and are reasonably durable. However,
checks are not always of recognizable value. If a stranger offers you a
$30,000 check for your car, chances are you will not hand over the keys.
The problem is that you do not recognize the true value of the check—it
may bounce, in which case you are out one car. Recent advances in electronic
check verification, however, are making it easier for some check recipients
(primarily businesses) to verify checks quickly.
Inside Money Box 1.3
Cigarettes as Money
in a Prisoner of War Camp
During World War II, Allied prisoners in a Ger¬ in terms of the number of cigarettes needed to
man POW camp developed a monetary system buy the goods. Even nonsmoking prisoners were
in which cigarettes served as the medium of willing to sell chocolates in exchange for ciga¬
exchange. Cigarettes played a dual role in this rettes, which they could then use to purchase
POW camp: They could be used as a commodity items of value from a third party.
(smoked) or to purchase other goods. The Because cigarettes were useful as money,
choice of cigarettes as the medium of exchange few cigarettes were actually smoked. While the
stemmed from their attributes. They were divisi¬ camp received a fairly steady supply of new cig¬
ble, from carton to pack to individual cigarette. arettes, a stock of unsmoked cigarettes circu¬
They were also durable, portable, and of easily lated as money. Smokers would have enjoyed
recognizable value. seeing all of the money go up in smoke, but
The prisoners received fixed weekly rations they did not because cigarettes were valued as
from the Red Cross, which included various the medium of exchange. The use of a com¬
foodstuffs in addition to cigarettes. They could modity money (like cigarettes) solves the double
trade the food among themselves to reallocate coincidence of wants problem, but it reduces
the rations according to their preferences for the consumption of the good used as money.
various items, but this practice required a
double coincidence of wants. When cigarettes
emerged as the medium of exchange, prices of Source: R. A. Radford, "The Economic Organization of a
chocolates and other goods began to be stated P.O.W. Camp," Economica (November 1945), 189-201.
The Evolution of Banks

Soon after society recognized the benefits of using money as a medium of
exchange, it recognized the need for a safe place to store it. This “safe
place” ultimately evolved into the banks of today—financial institutions that
accept deposits and make loans. The remaining chapters of this book will
explain in more detail the operation of modern banks and other financial
institutions such as savings and loan associations, credit unions, and mutual
savings banks. For now, bear in mind that all of these institutions are similar
in that they accept deposits and make loans. The remainder of this chapter
offers some background on how these institutions originated and evolved
into today’s banking system.
Banks as Depositories
Banking existed in Babylonia and in ancient Greece and Rome. During the
Middle Ages, gold and silver (known as specie, from the Latin for “in-
kind”) served as a full-bodied medium of exchange. People naturally wanted
to store their specie in a location safe from theft, fire, and other hazards.
Such a location required some sort of safe or strongbox, and the town
goldsmith was one place that invariably offered such accommodations. Other
wealthy individuals or businesses also had strongboxes for their own valu¬
ables and leased space in them to others, but goldsmiths of necessity had
strongboxes to store the gold they used in their trade.
Demand Deposits
A convenient way to illustrate the operation of banks as depositories is
depicted in Table 1.1, which is called a T-account. In a T-account, the bank’s
assets are listed on the left side and its liabilities are listed on the right side.
In this T-account, individuals have deposited $75 in gold, which are held in
reserve in the bank’s vault. This represents the bank’s assets. Notice that the
gold on deposit is not a “gift” to the bank; the depositors are free to
withdraw it whenever they choose. The obligation to give back $75 in gold
to depositors at their demand represents a liability to the bank. For this
reason, even today checkable deposits at banks are called demand deposits.
Demand deposits account for the $75 in liabilities on the right-hand side of
the T-account in Table 1.1.
At first, goldsmiths merely provided safekeeping for specie deposits and
charged depositors a fee for this service. When depositors wanted the use of
their gold or silver deposits, they went to the goldsmith’s place of business
and asked for their specie. However, it became apparent that it was not really
necessary to make a trip to the goldsmith to obtain specie for purchasing
goods. Instead, a depositor could use a written order to pay for goods. In
effect, a person with deposits at a goldsmith would trade this written order
for goods and services.
The written order to pay would state that the depositor was authorizing
the goldsmith to pay a specified amount of specie to the person named on
Table 1.1
A Bank Holding Demand Deposits
Assets Liabilities
Reserves (gold in vault) $75 Demand Deposits $75

the order. These orders to pay were the precursor of today’s familiar checks.
Consider the check you probably wrote to pay for your books at the begin¬
ning of the semester. On that check you indicated who was to be paid (XYZ
Bookstore), the amount to be paid, and the date, and you signed the check
to verify that it was you who wrote it. The bookstore was then free to present
the check to your bank, and your bank gave the bookstore the indicated
funds from your checking account. Except for the facts that goldsmiths held
gold deposits and written orders to pay indicated that gold was to be paid,
the early checking account arrangements worked just like today’s checking
accounts.
The innovation of the written order to pay made it easier to buy goods.
A shopper no longer had to visit the goldsmith to obtain gold before shop¬
ping; she or he could write checks for purchases once inside a store. Shoppers
were also less likely to be robbed, since they did not have to carry gold on
shopping trips.
The original written orders to pay were along the lines of “I, Jane K.
Depositor, order Gary Goldsmith to pay Sam J. Shoemaker X amount of
gold.” As we noted, these written orders to pay were very convenient, and
Sam J. Shoemaker actually had several options when he received such a
check. He could turn in the check to Gary Goldsmith and receive the indi¬
cated gold from Jane’s account; he could turn in the check to Gary and ask
him to take the gold out of Jane’s account and put it in his own account; or
he could endorse the back of the check and pass it on in a subsequent
exchange, perhaps with Theresa Tanner. In the latter case, Theresa Tanner
had the same options Sam did. Notice, however, that if Sam wanted to
simply endorse the check and use it to pay Theresa, the value of the purchase
had to be the same as the amount specified on the check from Jane to Sam.
Otherwise, Sam or Theresa would have had to “make change” from the
amount stated on the check. This inconvenience limited the number of times
any check would change hands before being presented at the bank for
redemption.
Box 1.4 shows the recent growth in currency and checking account
balances in the United States.
Bank Notes
Another innovation was the bank note, a document written by the early
banks (goldsmiths) promising to pay a sum of specie to the bearer on de¬
mand. A bank note stated something like “I, Gary Goldsmith, promise to
pay the bearer X amount of gold upon demand.” Bank notes were issued in
convenient denominations and could be used to buy goods and services.
Thus, when Jane K. Depositor deposited gold with the goldsmith, she could
elect to exchange some of the gold for bank notes. She then used these bank
notes as money, and anyone who received them could exchange them for
gold at Gary Goldsmith’s or use them to buy something.
Inside Money
Where Are All the Dollars?
The figure below plots currency and checking rency balances is individuals who participate in
account balances per capita in the United States the "underground economy," especially those
between 1980 and 1992. Notice that in 1992, who buy and sell illegal drugs. Currency, unlike
the average checking account contained almost checks, cannot be traced and is thus the pre¬
$3,000 per person and the average amount of ferred medium of exchange for people involved
currency held per person was nearly $1,000. in illegal activities.
The large amount of currency held per person is Another group holding U.S. currency is resi¬
something of a puzzle: How many families of dents of foreign countries. The U.S. government
four do you know who keep $4,000 ($1,000 for does not keep records of its currency held out¬
each person) in their home or pockets? side the United States, but the Federal Reserve
Most households hold a minimal amount of System estimates that as much as two-thirds of
currency, but businesses hold larger amounts to U.S. currency is held abroad.
use in making change. However, businesses and
households alone cannot account for the large
amount of currency per person in the United serve System, Federal Reserve Bulletin, various issues and
States. One more likely group holding large cur¬ Citibase electronic database.
Dollar Value of Currency and Checkable Deposits per Capita in the United States, 1980 - 1992
0
'80 '81 ’82 ’83 ’84 '85 '86 '87 '88 '89 '90 '91 '92
Notice that bank notes, like modern-day checks and currency, serve as
money. The main difference is that bank notes and checks are supplied by
banks rather than by the government. Moreover, bank notes have one big
advantage over checks: They are issued in convenient denominations that
facilitate their use in trade and thus stay in circulation for a longer time than
checks do. A check written for a specific amount to a particular person is
more difficult to use in a second transaction, and therefore the payee usually
takes it to the bank for redemption.
Bank notes do have one disadvantage compared to checks. Since bank
notes pay the bearer on demand instead of indicating payment to a particular
person, they are much more prone to theft than checks are. Using stolen
bank notes to make purchases is as easy as using gold, whereas it is difficult
for Joe Thief to cash a check payable to Hank Tanner.
Today two types of bank notes are issued in the United States. The more
prevalent type is the fiat currency issued by the Federal Reserve System, the
central bank of the United States established by the U.S. government to
manage the nation’s money supply. In contrast to the early bank notes, which
were issued by private banks and backed by gold, Federal Reserve notes are
not backed by gold or any other commodity. The second type of bank note
commonly used today, although comprising less than 1 percent of all money
used, is traveler’s checks. These bank notes are issued by major banks and
are redeemable for Federal Reserve notes on demand.
The effect of issuing bank notes can be visualized using the bank’s T-
account. Suppose the bank we examined in Table 1.1 issues $25 in new bank
notes in exchange for $25 in gold. This increases the reserves of the bank
(gold in the vault) from $75 to $100, which is reflected on the asset side of
the T-account in Table 1.2. Since the bank notes are redeemable for gold,
they reflect a $25 liability for the bank. This accounts for the additional entry
on the liability side of the T-account.
Table 1.2
A Bank Issuing Bank Notes
Assets Liabilities
Reserves (gold in vault) $100 Bank Notes $25
Demand Deposits $75
Fractional Reserve Banking

Initially goldsmiths functioned solely as depositories and issuers of bank
notes in exchange for specie, honoring requests for redemption of both bank
notes and demand deposits. In this function, the banks held specie in their
vaults to cover all possible redemptions of bank notes and demand deposits.
This practice is called 100 percent reserve banking, since the bank holds
reserves (in this case, specie in the vault) equal to the total value of outstand¬
ing bank notes and demand deposits. Reserves, being 100 percent of deposit
and note liabilities, fully cover the hypothetical situation in which all deposits
and bank notes are withdrawn on the same day.
Under 100 percent reserves, banks earn profits by charging a fee to store
deposits and to trade specie for bank notes. However, banks soon realized
that most of the specie in their vaults was never withdrawn; it sat idle while
demand deposits and bank notes circulated as money. These early banks
realized that they could in fact loan out some of their reserves and earn
additional profits on loans. This approach led to fractional reserve banking,
in which bank reserves equal only a fraction of outstanding demand deposits
and bank notes. With fractional reserve banking, banks held some deposits
as reserves in the vault but loaned out the remainder to people in need of
funds. In fact, a bank would actually lend in the form of bank notes, issuing
more such notes than there was gold in the vault. The bank would charge
interest on this loan and hence had a second source of revenue in addition
to the storage charges assessed on deposits.
To see the impact of fractional reserve banking, suppose the bank in
Table 1.2 decides to keep only 20 percent of its bank note deposits in reserve
instead of 100 percent. Thus, for every $100 in liabilities, the bank decides
to keep $20 in reserve and loan out the remaining $80 in gold. The bank’s
rationale for doing this is simple. Since it is unlikely that all depositors will
withdraw their gold on the same day, much of its previous reserves sat idle
in the vault. By loaning out a fraction of these reserves—keeping enough to
cover expected withdrawals but loaning out the rest at a profitable interest
rate—the bank can profit. The bank in Table 1.2 had $100 in liabilities, so
it keeps 20 percent ($20) in reserve and issues loans of $80. This leads to
the T-account in Table 1.3, where the bank has assets consisting of $20 in
reserves and $80 in loans. Its liabilities are still $25 in bank notes and $75
in demand deposits; the bank simply converted one form of an asset (re¬
serves) into a different asset (loans).
Table 1.3
A Bank Issuing Loans
Assets Liabilities
Loans $80 Demand Deposits $75

Note what happens to the available supply of money due to banking

activities. Suppose the public has $100 of gold. With no banking, the money
supply is $100 in the form of gold that circulates as the medium of exchange.
When banks issue demand deposits and bank notes but keep 100 percent in
reserves, as in Tables 1.1 and 1.2, the money supply in the hands of the
public is still only $100. The only difference is that with demand deposits
and bank notes, the gold that used to circulate as the medium of exchange
is placed in a bank vault and paper circulates as the medium of exchange.
In contrast, consider what happens with fractional reserve banking, the
situation in Table 1.3. Here the money supply in the hands of the public
consists of $25 in bank notes, $75 in demand deposits, and $80 in gold that
was placed back in the hands of the public when the bank issued the loan.
The total money supply is now $180. Of this, $25 is in the form of bank
notes, $75 is in the form of demand deposits, and $80 is in the form of gold.
By engaging in fractional reserve banking, the bank has created $80 in
additional money—seemingly out of nothing—by loaning out a fraction of
deposits and bank notes!
The story does not end here, however. The borrower who has obtained
$80 in gold from the local bank is not likely to have borrowed those funds
just to keep them idle. Instead, she will probably spend them. Once she
has, what will the new holder of the $80 in gold do with it? If he holds
on to the gold, the money supply will stay at $180. However, if he deposits
the $80 in the bank, the bank’s gold reserves will increase by $80 to $100.
Its demand deposits will also increase by $80 to $155, leading to the T-
account in Table 1.4.
Table 1.4
A Bank Gets New Deposits
Assets Liabilities

The bank is willing to lend out 80 percent of the value of its liabilities,
keeping only 20 percent in reserve. The new demand deposits at the bank
allow it to make additional loans. In particular, 80 percent of its current
liabilities of $180 is $144, but the bank has issued only $80 in loans. Thus,
the bank can increase its loans by $64 and still have reserves equal to 20
percent of its outstanding liabilities. Assuming the bank makes this loan, it
ends up with the T-account in Table 1.5.
Table 1.5
A Bank Issues New Loans
Assets Liabilities
Notice in Table 1.5 that the bank has gold reserves of $36 and loans of
$144 for total assets of $180. What is the value of the money supply in the
hands of the public? There are $25 in bank notes and $155 in demand
deposits, and the bank just made a $64 loan of gold that some individual
holds, making a $244 money supply. (Note that we do not count the original
loan of $80, since that loan is now included in the demand deposits in Tables
1.4 and 1.5.) Thus, with just two loans, the bank has transformed a money
supply consisting of $100 in gold into a $244 money supply!
This is still not the end of the story. The person who obtained the last
loan for $64 in gold is almost sure to spend it, and the person receiving the
$64 in payment for goods and services is likely to deposit it in the bank.
Compared with Table 1.5, demand deposits would increase by $64 to $219,
and reserves would likewise increase by $64 to $100 of gold. Then, of
course, the bank would have $244 of liabilities and can loan out up to 80
percent of those liabilities, or $195.20. With $144 of loans already made,
the bank can lend $51.20 while still meeting its self-imposed 20 percent
reserve rule. In fact, the bank can keep making loans, as long as the gold
eventually gets redeposited in the bank after the borrower spends it, until
the T-account looks like Table 1.6.
In Table 1.6, the bank has $500 of liabilities and, under its 80 percent
rule, has made loans of $400 while maintaining gold in the vault equal to
20 percent of liabilities, or $100. At this point, the bank is “fully loaned
Tablel.6
A "Fully Loaned-Out" Bank
Assets Liabilities

out’ ’; it can no longer make a loan without violating its 20 percent reserve
rule and thus can no longer create money. However, notice what the money
supply has become under fractional reserve banking. Money in the hands of
the public is $500 ($25 in bank notes and $475 in demand deposits). Notice
that all of the original $100 of gold is again in the bank vault, just as it was
under 100 percent reserves. However, with 20 percent fractional reserve
banking, the $100 of gold in the vault “creates” a money supply of $500.
Under fractional reserve banking, if all deposits and bank notes were
presented for redemption in a single day, the bank would be unable to honor
its pledge of redemption. Indeed, the bank in Table 1.6 is obligated to redeem
“paper” for $500 in gold but has only $100 in gold in the vault. If this
happened, the bank would not be bankrupt in the sense that it had a negative
net worth. Instead, it would be illiquid—unable to honor its obligations.
Note the difference between illiquidity and bankruptcy. Illiquidity is a
situation in which a bank cannot redeem its deposits on demand. In bank¬
ruptcy, the bank’s assets—the value of its reserves and the loans it has
made—are less than its liabilities—the value of the deposits it has accepted.
A bank can be illiquid but not bankrupt, and a bank can be bankrupt without
being illiquid.
If the bank were unable to honor its obligation to redeem demand de¬
posits and bank notes “on demand,” it would have to stop redeeming
deposits and notes. When a bank faces a long line of depositors and note
holders demanding redemption, we say that a “run on the bank” has oc¬
curred. Under a fractional reserve banking system, such an event is very
dangerous even to otherwise healthy banks, because no bank in such a system
can actually honor all its obligations to pay on demand if they are all
redeemed at the same time.
The U.S. banking system today is in fact a variation of the fractional
reserve banking system just described. Modern banks issue loans and hold
checkable deposits (demand deposits) that far exceed their actual reserves.
Figure 1.1 compares the level of reserves financial institutions hold and the
level of checkable deposits depositors could withdraw on demand. The total
reserves in January 1993 were $56.01 billion, which is considerably less
than the $738.8 billion in checkable deposits.
In later chapters, we will say much more about fractional reserve bank¬
ing, money creation by banks, and government regulation of fractional re¬
serve banks. Now you should have a basic understanding of how banks
evolved from depository institutions into fractional reserve banks and how
fractional reserve banking increases the amount of money used as a medium
of exchange.
Banks as Financial Intermediaries

Financial markets link savers and borrowers. This link is accomplished
through either direct finance or indirect finance. Direct finance occurs when
Source: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, various
issues and Citibase electronic database.
savers lend funds directly to borrowers. For example, if you ask a classmate
to loan you money to buy a new car and she agrees, you are engaging in
direct finance. No third party matched you (the borrower) with your class¬
mate (the lender). In fact, however, direct finance most often involves the
assistance of a third party, such as a broker, who brings buyer and seller
together. When IBM issues new bonds, it uses the services of various spe¬
cialists in the financial markets. The exact procedure need not concern us
here; simply note that the ultimate buyers of the IBM bonds will usually
engage the services of a broker to arrange the sale. This is one way financial
markets facilitate direct finance.
In contrast, indirect finance involves a particular type of middleman—
a third party who stands between the borrower and the lender. This middle¬
man is called a financial intermediary, and his or her role is to accumulate
funds from various savers and lend those funds to borrowers. Banks are
financial intermediaries because they accept deposits and lend those funds
to borrowers. Credit unions and savings and loan associations also are fi¬
nancial intermediaries. Less obviously, life insurance companies and pen-
sions accumulate savings and lend those funds to borrowers, thus performing
the role of financial intermediaries.
In all these cases, the financial intermediary borrows funds from savers
and lends them to borrowers. If you obtain a car loan from a bank, the bank
is actually loaning you money that other people have on deposit there. The
depositors receive interest from the bank; you pay the bank interest on the
loan. The bank is in the middle, earning a profit on the difference (margin)
between the rate you pay for the loan and the rate it pays depositors.
There is an important distinction between the function of a financial
intermediary and that of a broker. A broker brings together a buyer and a
seller of a financial instrument such as a bond, but does not personally issue
a bond or otherwise lend to the borrower; instead, the broker facilitates the
transaction without personally creating a financial instrument. In contrast, a
financial intermediary accumulates savers’ funds and lends them to borrow¬
ers by making a loan from itself. Thus, a bank makes a loan to a borrower
instead of merely introducing the borrower to one or more depositors and
brokering a loan between the two.
Thus, banks and other financial intermediaries are a special kind of
middleman. They profit on the difference between the interest rate they pay
on deposits and the interest rate they collect on loans. Their function is to
repackage depositors’ savings into loans to borrowers in the same way a
retailer might purchase produce in bulk from produce dealers and repackage
it for sale at the supermarket produce counter.
Figure 1.2 indicates the flow of funds between lenders and borrowers
under direct and indirect finance. With direct finance, the final lender (here
Ross Perot) provides funds directly to the borrower (AT&T). In this case,
the role of financial markets is to facilitate the transaction, and Ross Perot
will probably rely on a broker to carry it out. With indirect finance, Ross
Perot would deposit money in a financial intermediary such as NationsBank,
which in turn would make a loan to AT&T. Note that in this case, the
financial intermediary is both a borrower (from depositors) and a lender.
Why don’t depositors skip intermediaries and engage in direct finance?
After all, since a bank charges a higher rate on a loan than it pays depositors,
it would seem that a depositor could do better by making a direct loan to a
borrower or a borrower could get a lower rate by borrowing directly from a
saver. The answer to this question lies in the role intermediaries play in an
economy. Notice that we could also ask this question about product markets:
Why don’t consumers buy directly from factories instead of from retailing
outlets? The answer, of course, is that the benefits derived from using inter¬
mediaries exceed the extra cost of dealing with them.
What benefits do depositors (savers) and borrowers obtain from banks
that make it worth the extra cost? First, banks match up savers who want to
lend funds for short time periods with borrowers who want to borrow funds
for long time periods. Another way to say this is that financial intermediaries
are in a position to borrow short and lend long. Banks, for example, effec-
tively borrow funds from savers when they accept savers’ deposits. These
deposits are often redeemable on demand, or at least on relatively short
notice. Funds in checking accounts, for example, are redeemable on demand.
In contrast, banks often loan funds for fixed terms, and these terms may vary
from months (car loans) to years (home mortgages). The bank performs the
valuable function of converting funds that savers are willing to lend for only
short periods of time into funds that the bank itself is willing to lend to
borrowers for longer periods. Of course, a bank could get in severe trouble
if a depositor demanded redemption of her or his demand deposits while the
deposited funds were loaned out in a 15-year loan. If unable to come up
with the funds to redeem the demand deposits, the bank would be illiquid.
Banks are able to avoid illiquidity while borrowing short and lending
long by using several business practices. First, a bank seeks a widely diver¬
sified set of depositors so that no one depositor is likely to cause a liquidity
problem by withdrawing funds on short notice. Also, with a large number
of depositors, the odds are that on most days the dollars added to some
accounts and withdrawn from others will roughly cancel each other out,
leaving total deposits fairly stable. Second, a bank keeps a small amount of
deposits as reserves against sudden withdrawals and, more important, estab¬
lishes several lines of credit with other banks and with government regulators
(e.g., the Federal Reserve System). These lines of credit can be used to meet
unexpectedly large withdrawals of deposits. Third, a bank makes long-term
loans to only a small fraction of its loan clients. For example, banks make
a very small proportion of housing loans relative to their total loan portfolios.
Finally, a bank “lends” a portion of its funds by purchasing government
bonds, which are relatively easy to convert into currency. In this way, a
portion of the bank’s loan portfolio is kept in highly liquid assets, which can
be sold to obtain funds to meet unexpected withdrawals.
The second valuable function banks perform as intermediaries in finan¬
cial markets is pooling many small deposits to make relatively large loans
to borrowers. If you have only $1,000, you cannot make a $100,000 loan to
someone seeking a mortgage. But with a bank serving as an intermediary,
you and other individuals can deposit your small sums in a bank to earn
interest. The bank, in turn, can pool the numerous small deposits into a
sizable quantity of funds to make the $100,000 mortgage.
The third function banks serve is helping depositors reduce risk. Small
depositors, and even relatively large depositors, often find it difficult to
minimize the risk they face in lending funds directly to borrowers. If Natalie
lends her entire life’s savings to a single borrower who defaults, she loses
everything. While this may be an unlikely occurrence, it is not impossible.
An alternative for Natalie is to lend out her life savings in smaller amounts
to a large number of borrowers so that if one borrower defaults, all is not
lost. This alternative is called diversifying: The saver lends his or her funds
to a diverse group of borrowers so that all is not lost if one or even several
borrowers default. It is difficult to engage in diversification without a finan¬
cial intermediary; borrowers typically want large sums of money, and small
depositors usually cannot supply large sums to a diverse group of borrowers.
Thus, banks provide a way for savers, especially small savers, to diversify.
Each depositor bears only a small risk, since default by one or even several
borrowers will have relatively little impact on the bank or its depositors.
Finally, financial intermediaries economize on transactions costs relative
to those that would occur with direct finance. We have seen that transactions
costs are the costs borne in making an exchange, such as the time and legal
costs associated with a mortgage contract. One major transactions cost in
making loans is the cost of checking the creditworthiness of a potential
borrower. If individual savers attempt to diversify without using a financial
intermediary such as a bank, each will make small-denomination loans to a
number of borrowers. In so doing, each saver will want to personally verify
the creditworthiness of the borrower, and a duplication of effort will result.
For example, suppose a borrower wants to borrow $10,000 from each of 10
individual savers. Each saver will want to check the borrower’s creditwor¬
thiness. Thus, 10 separate credit checks will be done before the borrower
obtains $100,000 in loans. In contrast, a bank can borrow $10,000 from each
of the 10 savers, pool their funds, and make a single $100,000 loan to the
CONCL USION 27
borrower. This will require only a single credit check, saving the time and
effort involved in the nine extra credit checks that would be necessary under
direct finance.
Another transactions cost associated with lending is the cost of moni¬
toring the borrower. This involves ensuring that payments are current and,
more important, that the borrower stays in relatively good financial health.
Again, 10 savers engaged in direct finance would each have to bear these
costs, while a bank that pooled these 10 savers’ funds would have to bear
these costs only once. Moreover, banks specialize in making loans, and this
specialization leads to greater knowledge and efficiency in conducting credit
checks and monitoring loans. Obviously the average individual saver cannot
perform these tasks as efficiently as a bank can.
Summary How did bank loans evolve?

Exercise 7.5
Answer: As society began using commodity money as a medium of
exchange, a need for a safe place to deposit money developed. Early banks
were depositories in which individuals paid a fee for the privilege of depos¬
iting commodity money such as gold. Due to the transactions costs associated
with having to go to the bank to obtain gold to use in transactions, demand
deposits (checking accounts) and bank notes emerged. As these forms of
“paper money” became widely accepted as the medium of exchange, banks
recognized they had much more gold in their vaults than they needed to
honor demands for gold by depositors. Consequently, they began loaning
out a fraction of their deposits to earn additional profits in the form of interest
payments. Today, banks function as middlemen, or intermediaries, in finan¬
cial markets by (1) borrowing short and lending long, (2) pooling small
deposits into large loans, (3) diversifying risk, and (4) economizing on
ii transactions costs relative to engaging in direct finance.
Conclusion
This chapter examined what money and banks are and the economic reasons
for their existence. Money is not synonymous with income or wealth; rather,
it is anything that is widely and generally accepted as the medium of
exchange. Money is the most liquid of all types of assets; that is, it can easily
be converted into other assets or commodities. In addition to serving as a
medium of exchange, money serves as a unit of account, a store of value,
and a standard of deferred payment.
While various types of money have been used as a medium of exchange
in the history of humankind, societies learned very quickly that money
should consist of something durable, divisible, portable, and easily recog¬

nizable in value. The first type of money civilization used was full bodied
commodity money—a commodity with nonmonetary value that circulated
as the medium of exchange. Representative full-bodied money (paper money
that is “backed” by some commodity) was a predecessor to the current fiat
money (unbacked paper money) used today. Individuals accept fiat money
for payment only because they believe other individuals and businesses will
in turn accept it from them as payment for goods and services.
The emergence of money as a medium of exchange led to the need for
banks as depositories. Demand deposits (checks) and bank notes developed
out of a need to reduce the transactions costs of having to be physically
present to withdraw money from depositories before making purchases. This
development led to fractional reserve banking, wherein banks hold a fraction
of deposits in reserve and make loans to depositors in need of funds. Frac¬
tional reserve banking exists in the United States today and provides a means
by which banks actually “create money” and function as financial inter¬
mediaries.
KEY TERMS
money full-bodied money

medium of exchange representative full-bodied money
Federal Reserve note gold certificate
currency fiat money
liquidity specie
barter demand deposit
double coincidence of wants bank note
transactions cost fractional reserve banking
unit of account direct finance
store of value indirect finance
standard of deferred payment financial intermediary
commodity money
Questions and Problems
1. Look up the definition of money in your with an economist’s definition? Why or

dictionary. Does this definition agree why not?
Questions and Problems 29
2. On the island of Maka, the government 7. What would happen to your bank if all
has been unable to control the counterfeit¬ depositors showed up at 9:00 a.m. next
ing of its bills. Consequently, individuals Thursday and demanded payment for their
and merchants rely almost exclusively on deposits?
barter as a means of exchange. Why do
you think counterfeiting would lead indi¬ 8. Your brother-in-law wants to borrow
viduals to shun money and turn to barter? $1,000. Relationship aside, why might
you prefer to deposit the $ 1,000 in a sav¬
3. Discuss the advantages and disadvantages ings and loan and have your brother-in-
of using the following commodities as law borrow the funds there instead of
money. loaning him the $1,000 directly?
(a) . Bricks 9. In the example of fractional reserve bank¬

(b) . Wine ing presented in the chapter, the bank de¬
(c) . Com cided to keep only 20 percent of its lia¬
(d) . Pearls bilities in reserve. What would the final
(e) . Platinum T-account look like if the bank decided to
(f) . Uranium keep only 10 percent of its liabilities in
reserve?
4. Suppose you live in Llano, Texas, a small
10. Explain why money reduces transactions
town with a population of 2,500. Rank the
costs. Are there any other advantages to a
following assets from most liquid to least
monetary economy? Are there any disad¬
liquid, and explain why you ranked them
vantages?
as you did.
(a) . One hundred shares of IBM stock 11. Explain how the banking system reduces
(b) . A house in Llano, Texas transactions costs. Are there any other ad¬
(c) . A used computer vantages to the banking system? Are there
(d) . A used car any disadvantages?
(e) . One British pound (the medium of
12. What is the difference between full-bodied
exchange in Britain)
money and representative full-bodied
(f) . A roll of quarters
money? Between representative full-
bodied money and fiat money?
5. How would your answers to Problem 4
change if you lived in London, England? 13. Is one dollar or one dollar’s worth of gold
more valuable? Explain carefully.
6. In the early days of banking, goldsmiths
promised to pay a specified quantity of 14. “U.S. dollars are backed by the gold in
gold to anyone presenting a bank note or Fort Knox, Kentucky.” Is this statement
an endorsed check at their place of busi¬ true or false? Explain.
ness. What does your bank promise to pay
15. Can something serve as money without
you if you present the following?
serving as the unit of account? (Hint: How
(a) . An endorsed check many checks does it take to buy a shirt? A
(b) . A Federal Reserve note suit? A house?)
Selections for further reading
Aschheim, J., and G. S. Tavlas. “Doctrinal Founda¬ Towey, R. E. “Money Creation and the Theory of the
tions of Monetary Economics: Review Essay.” Banking Firm.” Journal of Finance, 29 (March
Journal of Monetary Economics, 28 (December 1974), 57-72.
1991), 501-510. Webb, S. C. “The Deficient Treatment of Money in
Buehler, J. E. “The Specific Role of Interest in Finan¬ Basic Undergraduate Texts: A Comment.” Journal
cial and Economic Analysis under Inflation: Dis¬ of Money, Credit, and Banking, 4 (February 1972),
cussion.” American Journal of Agricultural Eco¬ 109-112.
nomics, 67 (May 1985), 396-397. Wells, D. R., and L. S. Scruggs. “Historical Insights
Davis, S. I. Managing Change in the Excellent Banks. into the Deregulation of Money and Banking.”
New York: St. Martin’s Press, 1989, xi, 163. Cato Journal, 5 (Winter 1986), 899-910.
Fenstermaker, J. V., and J. E. Filer. “The U.S. Em¬ White, L. H., and G. A. Selgin. “The Evolution of a
bargo Act of 1807: Its Impact on New England Free Banking System.” Competition and Cur¬
Money, Banking, and Economic Activity.” Eco¬ rency: Essays on Free Banking and Money. Cato
nomic Inquiry, 28 (January 1990), 163-184. Institute Book. New York and London: New York
Ng, K. “Free Banking Laws and Barriers to Entry in University Press, 1989, 218-242.
Banking, 1838-60.” Journal of Economic History, Wicker, E. “A Reconsideration of the Causes of the
48 (December 1988), 877-889. Banking Panic of 1930.” Journal of Economic
Palasek, K. “The Case for a Laissez-Faire Monetary History, 40 (September 1980), 571-583.
System.” Cato Journal, 9 (Fall 1989), 399-403. Wulwick, N. “The Radcliffe Central Bankers.” Jour¬
Rockoff, H. “The ‘Wizard of Oz’ as a Monetary Al¬ nal of Economic Studies, 14 (1987), 36-50.
legory.” Journal of Political Economy, 98 (August
1990), 739-760.
CHAPTER
Financial Markets, Financial

Institutions and Instruments,
and Money
n Chapter 1, we saw how and why money and banking evolved and
looked briefly at the roles banks play as intermediaries in financial markets.
In this chapter, we turn to the modem world of financial markets and insti¬
tutions. Understanding this world helps on a practical level, such as when
saving money for a new car, a house, or even retirement. More important,
it is fundamental to understanding the workings of the economy and the
concepts covered in the remainder of this book.
We distinguish among three basic types of markets: debt, equity, and
financial service markets. We describe the financial institutions that partici¬
pate in these markets and the different kinds of assets these institutions hold.
We examine both commercial banks and nonbank depository institutions
such as savings and loan associations, as well as nondepository institutions
such as insurance companies and pension funds. As we will see, recent
deregulation of the banking industry has blurred the distinction between
banks and nonbank institutions. Finally, we look at the official definitions
of the money supply used by the U.S. Federal Reserve System. As you read
this chapter, keep in mind the functions and properties of money discussed
in Chapter 1.
Financial Markets
Economists define a market as an institution or arrangement that facilitates

the purchase and sale of goods, services, and other things. A financial
market is an institution or arrangement that facilitates the exchange of
financial assets, including deposits and loans, corporate stocks and bonds,
government bonds, and more exotic instruments such as options and futures
contracts. This market may or may not have a precise physical location. For
instance, the New York Stock Exchange (NYSE) is physically located on
31
32 Chapter 2 Financial Markets, Institutions and Instruments, and Money
Wall Street in New York City, whereas the over-the-counter (OTC) market
for stocks, called the National Association of Securities Dealers (NASD),
consists of brokers disbursed throughout the country who track prices via
computer and telecommunication lines. NASD is best known for the news¬
paper quotes of stock prices it generates, called NASDAQ (National Asso¬
ciation of Securities Dealers Automatic Quotation System).
There are several ways to classify financial markets. One obvious method
is to classify markets by the type of asset traded, for example, short-term or
long-term assets. (We will discuss these two classifications when we look at
financial instruments later in this chapter.) The broadest way to categorize
markets is to distinguish between primary and secondary markets. In a
primary market, new issues of financial assets are bought and sold. For
instance, if AT&T issues a new share of stock, (stock that has never been
owned by an investor before) the stock is sold in the primary market for
new issues of corporate stock. In contrast, existing financial assets are bought
and sold in a secondary market. You already know of a famous secondary
market for stocks: the New York Stock Exchange. If you wish to sell a share
of your AT&T stock, you sell it in the secondary market. The person who
buys your share of stock will purchase it in the secondary market.
The primary market for U.S. savings bonds is very widespread. These
instruments can be purchased from commercial banks nationwide, as well
as through payroll deductions. There is no central clearinghouse for such
purchases, nor is there a secondary market for U.S. savings bonds. You can
redeem the bonds, but you cannot sell them to a third party.
In contrast, U.S. Treasury bonds, which the U.S. government issues to
finance the federal debt, have active primary and secondary markets. The
primary market for these bonds is an auction held by the U.S. Treasury at
which banks, securities dealers, and other institutional investors bid for new
issues. The secondary market for these bonds is an over-the-counter market,
much like the NASD, that has no precise physical location.
The existence of a secondary market for a financial asset enhances the
asset’s liquidity. For example, suppose you purchase a share of IBM stock
in the primary market but later decide to sell it for cash to use as a down
payment on a house. Since IBM stock is actively traded in the secondary
market, it is relatively easy to sell. You simply call a brokerage firm such
as Merrill Lynch or Charles Schwab and tell the broker you wish to sell.
The broker, in turn, locates a buyer and sells your stock for you. The entire
process can be carried out in seconds.
Because a secondary market makes it easier for buyers and sellers to
find each other, it lowers the transactions costs of buying or selling an asset.
In the absence of a secondary market for a stock, you would have to per¬
sonally locate someone willing to purchase the stock from you. Not only
could this take considerable time; you would be unlikely to locate a buyer
willing to pay the highest price for your stock. Moreover, if you thought it
would be difficult to sell a financial asset such as IBM stock, you would be
Financial Markets 33
less likely to buy it in the first place; doing so would tie up your wealth in
an illiquid asset that you could not easily convert into cash. A secondary
market that allows trades between buyers and sellers of existing shares
enhances the liquidity of corporate stock. This induces investors to own
stock and makes it easier for firms to acquire funds in the primary stock
market.
In addition to distinguishing between primary and secondary financial
markets, we can classify financial markets according to the nature of the
instruments traded. This classification consists of debt, equity, and financial
service markets. A brief description of the nature and role of each type of
financial market follows.
Debt Markets
The financial market most familiar to students and professors alike is the
debt market. In the debt market, lenders provide funds to borrowers for
some specified period of time. In return for the funds, the borrower agrees
to pay the lender the original amount of the loan (called the principal), plus
some specified amount of interest. (We will examine interest in more detail
in Chapter 3.) Individuals use debt markets to borrow funds to finance
purchases such as new cars and houses. Corporate borrowers use them to
obtain working capital and new equipment. Federal, state, and local govern¬
ments use debt markets to acquire funds to finance various public expendi¬
tures. New funds (the issue of a new bond or a new-car loan, for instance)
occur in the primary debt market. When an individual or a financial insti¬
tution buys or sells an existing loan—such as when Sallie Mae buys your
student loan—the transaction takes place in the secondary debt market.
Assets in the debt market are further classified as short-term if the under¬
lying obligation when issued is one year or less, intermediate-term if the
obligation when issued is between one and ten years, and long-term if the
obligation is more than ten years in length.
Equity Markets
Ownership of tangible assets (such as houses or shares of stock) are bought
and sold in an equity market. New houses and new issues of stock are sold
in primary markets; existing houses and existing shares of stock are traded
in secondary markets. An example of a secondary equity market for shares
of stock is the American Stock Exchange.
Financial Service Markets

Individuals and firms use financial service markets to purchase services
that enhance the workings of debt and equity markets. For instance, com¬
mercial banks provide depositors not only with interest on deposits but also
with a host of services such as check processing, safety deposit boxes, or
ATM transactions. As we saw in Chapter 1, banks use funds on deposit to

participate in the debt market (by issuing loans). But banks really do more
than serve as an intermediary; they also provide “convenience,” a valuable
service that consumers are often willing to pay fees to enjoy. No secondary
market exists for financial services, since people do not sell “used” financial
services to third parties.
Another financial service is brokerage services. Brokers are intermedi¬
aries who compete for the right to help people buy or sell something of
value. Real estate brokers help individuals buy or sell houses; stockbrokers
help individuals buy or sell assets such as stocks and bonds. As interme¬
diaries, brokers receive a fee for performing the service of matching buy¬
ers and sellers of assets. Dealers differ from brokers in that dealers actually
buy and sell securities from their portfolio, and do not just match up buyers
and sellers.
Finally, financial service markets provide individuals and firms with a
means of insuring against various forms of loss, from loss of life to the loss
of a valuable painting. These services too are performed for a fee; the
proceeds are used not only to pay out insurance claims but also to purchase
financial assets in debt and equity markets.
Summary Is the market for used cars a primary or a secondary market? How are
Exercise 2.1 financial brokers similar to car dealers? How do they differ?
Answer: The market for an existing or “used” asset is the secondary

market. Thus, the used-car market is a secondary market. When cars are sold
in the secondary market, General Motors, Ford, and Chrysler do not obtain
funds. Of course, if you buy a used car from a dealer, the dealer will obtain
funds. But a used-car dealer acts as an intermediary in the market, earning
a profit on the margin between what the dealer paid an individual for the
used car and what it sells the car for. Of course, the same is true when a
financial asset such as a stock or bond sells on the secondary market. Note
that if a used car is purchased by a car dealer, it is still a used car, and as
such the car dealer provides a brokerage or intermediary service: buying the
car from you to sell to another purchaser. The car itself is still a “used car,”
and the market for used cars remains a secondary market.
More generally, securities dealers and brokers, like car dealers, can deal
in both primary and secondary markets at the same time. Thus, a stockbroker
sometimes helps with the sale of a new issue of stock while also offering
her or his services to investors wishing to buy or sell existing shares of stock
in the secondary market. The main distinction between car dealers and
securities brokers is that a car dealer actually purchases the used car and
resells it later, whereas a broker never actually owns the stock but instead
facilitates a trade between buyer and seller.
Financial Institutions
Depository institutions (e.g., banks) and nondepository institutions (e.g.,

insurance companies) both serve financial markets. In this section, we look
at the three basic markets in which these institutions participate, the services
they offer and how they provide them, where they obtain funds, and the
assets they hold.
Depository institutions
Depository institutions accept deposits from individuals and firms and use
these funds to participate in the debt market, making loans or purchasing
other debt instruments such as Treasury bills. As we saw in Chapter 1, the
deposit market is a special type of loan market in which depositors “loan”
money to depository institutions, which in turn use the funds to purchase
other financial assets. The major types of depository institution are com¬
mercial banks, savings and loan associations, mutual savings banks, and
credit unions.
Commercial Banks. Commercial banks are the largest and most im¬
portant depository institutions. The roughly 12,000 commercial banks in the
United States have the largest and most diverse collection of assets of all
depository institutions. Their main source of funds is demand deposits (i.e.,
checking account deposits) and various types of savings deposits (including
time deposits and certificates of deposit). As Figure 2.1 reveals, the major
use of funds by commercial banks is making loans. In 1992, loans made up
61.5 percent of commercial banks’ assets; of these, 24.7 percent were real
estate loans and 36.8 percent were other loans such as loans to businesses
and automobile loans. The remaining 38.5 percent of commercial banks’
assets includes securities (primarily federal government bonds), vault cash,
and deposits at Federal Reserve banks.
Savings and Loan Associations. Savings and loan associations

(S&Ls) were originally designed as mutual associations, (i.e., owned by
depositors) to convert funds from savings accounts into mortgage loans. The
purpose was to ensure a market for financing housing loans. Over the past
20 years a very active market for mortgages has developed, diminishing the
need for regulations forcing S&Ls to concentrate on housing loans. Largely
for this reason, during the 1980s government regulation of the types of assets
savings and loans can hold was weakened, and today the distinction between
S&Ls and commercial banks is minimal. However, Figure 2.2 reveals that
as recently as 1992, nearly 74 percent of the assets held by S&Ls were
mortgages. Thus, S&Ls continue to hold a less diversified set of assets than
commercial banks do.
This figure shows Figure 2.1

loans and securities as Loans and Securities of Commercial Banks
a percentage of the
assets of commercial
banks as of September Other Assets
1992. Real estate $1,379 Billion
loans made up almost (38.5%)
25 percent of assets,
while other loans
comprised nearly 37
percent. Assets other
than loans were about
38.5 percent of total
assets and included
such things as securi¬
ties, vault cash and
deposits at Federal Re¬
serve banks.
Real Estate Loans

Non-Real Estate Loans
$883 Billion
$1,316 Billion
(24.7%)
(36.8%)
Source: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, January
1993, Table 1.25.
Mutual Savings Banks. Mutual savings banks are much like savings
and loans, but are owned cooperatively by members with a common interest,
such as company employees, union members, or congregation members.
Originally they accepted deposits and made mortgage loans. As with savings
and loans, recent deregulation allows the approximately 500 mutual savings
banks in the United States to issue checkable deposits and engage in financial
activities roughly on par with other depository institutions.
Credit Unions. Credit unions are organized as cooperative depository

institutions, much like mutual savings banks. Depositors are credited with
purchasing shares in the cooperative, which they own and operate. Like
savings and loans, credit unions were originally restricted by law to accepting
savings deposits and making consumer loans. Recent regulatory changes
allow them to accept checkable deposits and make a broader array of loans.
Nondepository institutions
In contrast to depository institutions, nondepository institutions do not accept
checkable deposits. With one exception that will be noted shortly, you cannot
simply write a “check” to withdraw funds from a nondepository institution.
Nondepository institutions serve various functions in financial markets,

ranging from financial intermediation to selling insurance against risk. We
discuss the principal types of nondepository institutions next.
Mutual Funds. Mutual funds sell shares to investors, and invest the
proceeds in a wide choice of assets. Owners of shares receive pro rata shares
of the earnings from these assets, minus management and other fees assessed
by the fund. Some mutual funds, called money market mutual funds, invest
in short-term, safe assets such as U.S. Treasury bills and large bank certifi¬
cates of deposit. Largely for historical reasons, money market mutual funds
are not considered depository institutions even though shareholders are often
allowed to write checks on their accounts. Unlike depository institutions,
money market mutual funds do not promise that the price of a share will
stay constant, but in fact, the price of each share tends not to fluctuate over
time like prices of stock mutual funds do. Fund managers work to keep share
prices constant, and owners of these “shares” receive a portion of the
earnings derived from the assets bought with the money received for the
shares. These funds enjoyed widespread popularity in the late 1970s and
This figure shows Figure 2.2

loans and securities of Loans and Securities of Savings and Loan Associations
S&Ls insured by SAIF
(Saving Association In¬
surance Fund) as of Other Loans
August 1992. Mort¬ $50,634 Million
gage loans and mort¬ (5.9%)
gage-backed assets
Other Assets
made up about 74
$173,070 Million
percent of assets,
(20.2%)
other loans about 6
percent, and other se¬
curities about 20 per¬
cent.
Mortgage Loans and

Mortgage-Backed Securities
$634,464 Million
(73.9%)
Source: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, January
1993, Table 1.37.
Changes in the Relative Size of Financial

Institution Assets, 1976-1990
The accompanying figure shows the percentage percent, respectively. Money market mutual
of assets held by various classes of financial in¬ funds had a paltry .2 percent of total assets.
stitutions in 1976 and 1990. In March 1976, to¬ Compare that to the situation in October
tal assets held by life insurance companies, 1990, not quite 15 years later. In that year total
money market mutual funds, commercial banks, assets of these four types of institution totaled
and institutions insured by Federal Savings and $4,281 billion, an increase of 174 percent. This
Loan Insurance Corporation, or FSLIC (i.e., sav¬ translates into a yearly increase of 7.2 percent
ings and loans), were $1,561 billion. Of this for 14V2 years.
amount, commercial banks had the lion's Note, however, that while total holdings
share—almost 59 percent—while FSLIC-insured grew, the distribution of these assets among the
institutions and life insurance companies divided four types of institution also changed, some-
up most of the remainder—about 22 and 19 Continued on p. 39
1976 1990
Money Market Funds

Life Insurance
Money Market Funds $341 Billion Commercial Banks
Companies
$3 Billion $1,563 Billion
$299 Billion
(0.2%)
(36.5%)
(19.1%)
Life Insurance
Companies
$1,303 Billion FSLIC Insured Institutions
FSLIC Insured
(30.4%) $1,074 Billion
Institutions
Commercial Banks (25.1%)
$340 Billion
$919 Billion
(21.8%)
(58.9%)
Continued from p. 38 ond, savings and loan associations began offer¬

times dramatically. For instance, by 1990 the ing share draft accounts (essentially interest-
share of assets held by commercial banks had bearing checkable accounts), which attracted
fallen to 36.5 percent, while the share of assets depositors away from commercial banks. It is in¬
held by the other three categories had ex¬ teresting that even after the troubles the savings
panded. FSLIC-insured institutions' share had ex¬ and loan industry experienced in the 1980s, the
panded from 22 to 25 percent, life insurance share of assets held by S&Ls increased relative to
companies' share from 19 to more than 30 per¬ commercial banks.
cent, and money market mutual funds' share Life insurance companies also gained mar¬
from 0.2 to 8 percent. ket share due to the innovation of offering a
How can we explain these movements of broad range of alternative accounts that allowed
funds among institutions? During the 1970s, savers to take advantage of high interest rates
two factors led depositors to seek alternatives to through life insurance policies. Another factor in
commercial banks. First, inflation rates were the growth of life insurance assets was demo¬
generally increasing. Second, checking accounts graphics. The baby boom generation was aging
at commercial banks (demand deposit accounts) and entering the market for life insurance in
were prohibited from paying interest. The in¬ large numbers. This change in demographic
creased inflation rates were reflected in higher structure increased the market share of life in¬
interest rates on financial instruments that were surance companies relative to other financial
substitutes for deposits at banks. Depositors re¬ institutions.
sponded in two ways. First, money market mu¬
tual funds offered attractive interest rates and at
Source: Board of Governors of the Federal Reserve Sys¬
least some checking privileges, and many depos¬ tem, Federal Reserve Bulletin and Citibase electronic data¬
itors moved their deposits to these funds. Sec¬ base.
early 1980s, when they grew at a rate of 37 percent per year. Box 2.1 gives
some reasons for this growth.
Insurance Companies. Insurance companies protect individuals

against risk. Life insurance companies accept regular payments from indi¬
viduals in exchange for contracted payments in the event of the insureds’
death. By insuring a large pool of individuals, life insurance companies can
consult actuarial tables and predict very accurately what percentage of the
insured individuals will die each year. Because of this, life insurance com¬
panies hold long-term assets, including long-term bonds. They also hold
substantial quantities of commercial real estate.
Figure 2.3 shows the disposition of assets of life insurance companies.
Loans make up only about 24 percent of assets, and the majority, roughly
20 percent, are mortgage and other real estate loans. The bulk of life insur¬
ance company assets, around 76 percent, are not in loans. This is in sharp
contrast to both commercial banks and savings and loans. Furthermore, the
assets insurance companies hold are purchased with insurance premiums
Source: Board of Governors of the Federal Reserve System, Federal Reser\>e Bulletin, various
rather than deposits. Thus, it is not possible to write a check against an

insurance policy.
Other insurance companies, called fire and casualty insurance compa¬
nies, insure against loss from fire, theft, and accident. If you own a car or a
house, you probably have purchased this type of insurance. Insurance claims
on these policies are somewhat less easy to predict. More important, the
duration of the liability (e.g., the “life expectancy” of your car) is lower
than for life insurance, so these companies usually invest in more liquid,
shorter-term assets.
Private and government (including federal, state, and

local) pension funds provide retirement income to employees covered by
the pension plan. Funds are collected by regular contributions from employ¬
ees, usually via payroll deduction. Since the funds flowing in are not demand
deposits, you cannot write a check against your balance in a pension fund.
Like life insurance companies, these institutions can accurately predict pay¬
outs and hence can hold long-term assets. They hold portfolios consisting
mostly of stocks and bonds. The returns on these assets are paid out to
participating individuals when they reach retirement age.
Brokerage Firms. As we already noted, brokerage firms serve the

valuable function of linking buyers and sellers of financial assets. In this
regard they function as intermediaries, earning a fee for each transaction
they create. Unlike garden-variety nondepository institutions, however, mod¬
ern brokerage firms such as Merrill Lynch and Charles Schwab also compete
with depository institutions in the deposit market, where they attract depos¬
itors with money market mutual funds. Nonetheless, brokerage firms are not
formally considered depository institutions because their main function is to
serve as brokers in the secondary debt and equity markets.
Summary Would you think commercial banks’ or life insurance companies’ assets are
Exercise 2.2 more liquid? Why?
Answer: Commercial banks are depository institutions, whereas life in¬

surance companies are not. A depositor at a commercial bank can withdraw
funds on demand by writing a check, whereas the owner of a life insurance
policy cannot write a check against the policy. For this reason, banks tend
to hold more liquid assets than life insurance companies do.
Financial Instruments
If you follow the financial news, you have heard of T-bills, commercial
paper, and federal funds. In this section, we will see what these and other
types of financial instruments are and how they are used to link borrowers
and savers in financial markets. Our discussion will use a classification
scheme that distinguishes between the market in which short-term financial
instruments are traded (the money market) and the market for longer-term
instruments (the capital market). We begin by introducing money mar¬
ket instruments, which are used in the market for short-term funds. We
then look at capital market instruments, which link long-term borrowers
and lenders.
Money Market instruments

The most liquid, short-term debt obligations are traded in the money market.
As we will see at the end of this chapter, the instruments traded in this
market are included in one or more official definitions of the money stock.
Figure 2.4 indicates the size of various money market instruments in 1976
and 1992 and also shows the relative growth in the various instruments over
this period. For example, money market mutual funds increased from $2.4
billion in 1976 to more than $360 billion in 1992. In this section, we examine
each money market instrument shown in Figure 2.4 in detail.
42 Chapter 2 Financial Markets , Institutions and Instruments, and Money
Commercial Paper. Commercial paper is a form of direct short-term

finance by large, creditworthy companies. If a company such as AT&T needs
immediate funds, it can sell commercial paper (a debt instrument) to another
corporation or financial institution. Commercial paper is a promise to pay

back a higher specified amount at a designated time in the immediate
future—say, 30 days. By issuing commercial paper, a corporation avoids the
process of applying for a loan and instead engages in direct finance. To
engage in direct finance effectively, the issuing company must be large and
creditworthy enough to find someone willing to accept its commercial paper,
which is sold with the aid of brokers. As Figure 2.4 shows, the use of
commercial paper grew from $50 billion in 1976 to $339 billion in 1992,
an increase of about 12.7 percent per year. The growing use of commercial
paper has increased the competitive pressure on banks, which are finding
some of their potential loan customers turning to the commercial paper
market.
Negotiable Bank Certificates of Deposit. Certificates of de¬

posit (CDs) are debt instruments sold by banks and other depository insti¬
tutions. A CD pays the depositor a specified amount of interest during the
term of the certificate, plus the purchase price of the CD at maturity. For
example, a $1,000, one-year CD paying 5 percent interest would pay $1,000
plus $50 interest at the end of one year (the term of the CD). Today nego¬
tiable CDs are sold in large denominations (over $100,000) and can be resold
in the secondary market. This makes negotiable CDs highly liquid. The
original purchaser need not hold the CD to maturity or pay “a substantial
penalty for early withdrawal” if he or she needs to liquidate the CD. Instead,
the person can sell the CD in the secondary market at a price that will depend
on the market interest rate in effect when it is sold. As Figure 2.4 shows,
the use of CDs grew from $119 billion in 1976 to $416 billion in 1992.
Treasury Bills. U.S. Treasury bills, or T-bills, are short-term debt

instruments used by the federal government to obtain funds. They are issued
in 3-, 6-, and 12-month maturities. These instruments do not pay regular
interest payments, but instead are sold at a discount. This means they are
sold for an amount that is less than what the government promises to pay at
maturity, and the difference between the purchase price and the face value
is the return from buying the T-bill. For instance, if you purchased a one-
year Treasury bill in July 1994 for $9,766, in July 1995 (when it matures)
the government would pay you $10,000.
There is a very active secondary market for U.S. Treasury bills. If you
purchased the above Treasury bill but decided to liquidate it for cash before
maturity, a simple phone call to a broker would allow you to sell it to another
investor. Furthermore, the likelihood that the U.S. government will default
on its obligation (not pay) is very low. These two factors make U.S. Treasury
bills one of the most liquid of all financial instruments.
It is important to distinguish Treasury bills from other U.S. government
securities such as Treasury bonds. T-bills mature in less than 1 year, T-notes
mature in 1 to 10 years, and T-bonds mature in more than 10 years. One
other practical difference exists among bills, notes, and bonds. U.S. Treasury
bills, unlike T-notes and T-bonds, pay no explicit interest. Instead, interest
is earned implicitly, that is, based on the difference between the purchase
price and the face value. For example, some Treasury bills have a $10,000
face value, and this is the amount the U.S. government promises to pay you
when the bill matures. You would pay less than the face value to purchase
this T-bill, perhaps $9,500. The difference between the price you pay and
the face value, $500, is the implicit interest you earn from lending the
government your money. Unlike with other bonds, this interest is not paid
monthly or quarterly but instead is received at maturity. Hence these Treas¬
ury bills are called zero coupon instruments. As Figure 2.4 indicates, Treas¬
ury bills and other short-term Treasury securities increased substantially in
volume over the 1976-1992 period, from $73 billion to $317 billion.
Repurchase Agreements. A repurchase agreement (or simply

repo) is an agreement by two parties in which the borrower sells and agrees
to buy back a financial instrument such as a government bond, note, or
T-bill. Suppose a bank needs short-term cash today. The bank can sell some
Treasury bills to a firm such as IBM with the agreement that the bank will
repurchase the T-bills in 30 days at a higher price. In effect, this repurchase
agreement is a short-term loan in which the Treasury bills serve as collateral.
As Figure 2.4 shows, repurchase agreements accounted for $71 billion of
liquid assets in the United States in 1992.
Eurodollars. Eurodollars are U.S. dollars deposited in banks located

in other countries. Foreign banks and offshore branches of U.S. banks hold
dollar deposits to service firms engaged in international trade, as well as for
other purposes. U.S. banks sometimes borrow Eurodollars when they need
short-term funds. The growth of the Eurodollar market in the 1970s was
spurred by the relative lack of regulations on these funds, including the
absence of regulations requiring banks to hold reserves against Eurodollar
loans. Figure 2.4 reveals that Eurodollars increased from $14 billion in 1976
to $56 billion in 1992.
Banker's Acceptances. A banker’s acceptance is a letter of credit

(a bank’s promise to pay on a specific date) that has been stamped as
“accepted” (guaranteed) by another bank. You might think of a banker’s
acceptance as analogous to a post-dated “check” or bank draft. If the party
issuing the check has insufficient funds in the account to cover the draft
when it is payable, the bank that stamped the draft is obligated to pay the
amount of the “check” to the party who holds it. Obviously a banker’s
acceptance is more valuable as a medium of exchange than a standard check,
since a bank guarantees that the banker’s acceptance will be honored. The
party issuing a banker’s acceptance pays a fee to the bank for its guarantee.
Banker’s acceptances are particularly valuable in international transac¬
tions, since it is extremely costly for a firm located in, say, France to recover
a bad check from a party located in Egypt. There is a relatively small

secondary market for banker’s acceptances, which essentially operates as a
scaled-down version of the market for Treasury bills. Figure 2.4 shows that
banker’s acceptances accounted for only $23 billion of the outstanding
money market instruments in 1992.
Federal Funds. Suppose it is 2:50 p.m. and a bank needs $10 million
by 3:00 p.m. to meet the reserve requirements set by the Federal Reserve.
Obviously it is too late to attract additional deposits, and the bank must find
a quick way to acquire the funds. One way the bank can obtain such funds
on short notice is to acquire federal funds. Federal funds are simply short¬
term (usually overnight) loans between banks. The funds are loaned at an
interest rate known as the federal funds rate. They are called federal funds
because they are held in deposit at the Federal Reserve rather than because
they are loaned by the federal government. In fact, the transfer of federal
funds is merely a bookkeeping transfer from the ledger of a bank with an
excess of reserves to the ledger of a bank with deficient reserves. Box 2.2
gives a sampling of money market interest rates as published in the Wall
Street Journal.
Capital Market Instruments

In contrast to money market instruments, which have maturities of one year
or less, capital market instruments have maturities of more than one year.
The principal capital market instruments are described next.
Corporate Stock. A share of corporate stock is an equity instrument

that represents ownership of a share of the assets and earnings of a corpo¬
ration. When a corporation like AT&T needs long-term funds, it can sell
shares of stock to individuals or other investors. AT&T uses the funds
received to purchase assets and run the company; in return, the shareholder
owns a share of these assets and the earnings they generate for AT&T. The
profits earned by a corporation and paid to shareholders are known as divi¬
dends. Unlike interest payments, dividends can vary with the health of the
company.
It is important to emphasize that the only time a corporation receives
money from stock is the time at which it issues the stock—the primary
market transaction. When the company decides to issue stock, it offers the
shares to underwriters, investment banks that guarantee the firm a certain
price for the issue. Then the investment banker (or bankers, if the issue is
large) sells the stock to individual investors, with the assistance of brokers,
at what they hope is a higher price than the guaranteed price. Effectively the
underwriters provide insurance to the company issuing the new stock and
bear the risk associated with the low price investors pay for the stock.
Once a new issue is in the hands of individual investors, the stock can
be sold and purchased by another investor (with the aid of a broker) in a
vm
Inside Money
Money Market Rates from
the Wall Street Journal
The Wall Street Journal publishes a variety of

money market interest rates each day. The first MONEY RATES
item listed is usually the prime interest rate—the Monday, April 12,1993
The key U.S. and foreign annual Interest rates below are
interest rate banks charge on loans to credit a guide to general levels but don't always represent actual
transactions.
worthy corporate borrowers. Market analysts of¬ PRIME RATE: 6%. The base rate on corporate loans
posted by at least 75% of the natlon's30 largest banks.
ten use changes in the prime interest rate as a
FEDERAL FUNDS: 3'/«% high, 3% low, 3% near closing
barometer of other interest rates. bid, 3 1/16% offered. Reserves traded among commercial
banks for overnight use In amounts of Si million or more.
Source: Prebon Yamane (U.S.A.) Inc.
DISCOUNT RATE: 3%. The charge on loans to depository
Source: The Wall Street Journal, April 12, 1993. Re¬ Institutions by the Federal Reserve Banks.
CALL MONEY: 5%. The charge on loans to brokers on
printed by permission of the Wall Street Journal, © 1993 stock exchange collateral. Source: Telerate Systems Inc.
Dow Jones & Company, Inc. All Rights Reserved World¬ COMMERCIAL PAPER placed directly by General Elec¬
tric Capital Corp.:3.07% 30 to 119 days; 3.08% 120 to 149 days;
wide. 3.10% 150 to 179 days; 3.13% 180 to 239 days; 3.15% 240 to 270
days.
COMMERCIAL PAPER: High-grade unsecured notes sold
through dealers by major corporations: 3.15% 30 days; 3.15%
60 days; 3.16% 90 days.
CERTIFICATES OF DEPOSIT: 2.61% one month; 2.64%
two months; 2.68% three months; 2.78% six months; 2.94% one
year. Average of top rates paid by malor New York banks on
primary new Issues of negotiable C.D.s, usually on amounts of
$1 million and more. The minimum unit Is $100,000. Typical
rates In the secondary market: 3.02% one month; 3.04% three
months; 3.12% six months.
BANKERS ACCEPTANCES: 3% 30 days; 3% 60 days; 3%
90 days; 3.02% 120 days; 3.04% 150 days; 3.05% 180 days.
Negotiable, bank-hacked business credit instruments typically
financing an Import order.
FOREIGN PRIME RATES: Canada 6%; Germany 9%;
Japan 4%; Switzerland 7.50%; Britain 6%. These rate Indica¬
tions aren't directly comparable; lending practices vary
widely by location.
TREASURY BILLS: Results of the Monday, April 12, 1993,
auction of short-term U.S. government bills, sold at a discount
from face value In units of $10,000 to $1 million: 2.89%, 13
weeks; 3%, 26 weeks.
FEDERAL HOME LOAN MORTGAGE CORP. (Freddie
Mac): Posted yields on 30-year mortgage commitments. De¬
livery within 30 days 7.07%, 60 days 7.18%, standard conven¬
tional fixed-rate mortgages; 4.125%, 2% rate capped one-year
adiustable rate mortgages. Source: Telerate Systems Inc.
FEDERAL NATIONAL MORTGAGE ASSOCIATION
(Fannie Mae): Posted yields on 30 year mortgage commit¬
ments (priced at par) for delivery within 30 days 7.16%, 60 days
7.26%, standard conventional fixed rate-mortgages; 4.95%, 6/2
rate capped one-year adiustable rate mortgages. Source:
Telerate Systems Inc.
MERRILL LYNCH READY ASSETS TRUST:2.76%. An¬
nualized average rate of return after expenses for the past 30
days; not a forecast of future returns.
secondary stock market such as the New York Stock Exchange or the Amer¬
ican Stock Exchange. Notice that the funds transferred in these secondary
markets pass between individual buyers and sellers of the stock rather than
to the corporation. Individuals own the majority of stock in the United States,
and pension funds, insurance companies, and mutual funds own the re¬
mainder. *
Corporate Bonds. A corporate bond is a debt instrument issued by

a corporation that states the firm will make specified interest payments
(typically twice each year) and a principal amount or “face value” (usually
$1,000) at maturity (say, 30 years). The original purchaser of a bond buys
this promise from the firm for an up-front amount, known as the price of
the bond. Unlike stockholders, bondholders own no share of the profits;
rather, they are entitled only to the interest payments and the face value due
on maturity. Obviously the firm’s “promise” is valuable to the purchaser
of the bond only if the firm does not go bankrupt. For this reason, only
corporations with strong credit ratings tend to issue bonds. If a company
fails to meet its payment obligations, it is said to be in default.
Corporate bonds, like corporate stock, provide funds to the issuing firm
when sold in the primary market. Like stocks, new bond issues are under¬
written by investment banks, which sell the bonds to individual investors.
When bonds are bought and sold in the secondary bond market (for example,
the New York Bond Exchange), money changes hands among individual
investors, and no funds flow to the corporation that issued the bond. Box
2.3 describes the most common types of corporate bonds.
Mortgages. A mortgage is a debt instrument used to finance the pur¬

chase of a home or other form of real estate when the underlying real estate
serves as collateral for the loan. If the borrower defaults, the lender receives
title to the real estate as payment of the debt. Several terms common to
mortgage market transactions are useful not only to students of money and
banking but to anyone who plans to use a mortgage to finance a house.
The two major types of mortgage instruments today are fixed-rate and
adjustable-rate mortgages. Each type of mortgage specifies a term (the length
of the mortgage), a down payment (usually expressed as the fraction of the
house value the buyer must finance personally), and points (the fraction of
the loan value that must be paid up front as prepaid interest). A fixed-rate
mortgage specifies an interest rate that is fixed during the term of the loan,
whereas the rate on an adjustable-rate mortgage (ARM) can change (usually
every one or three years). An adjustable-rate mortgage also stipulates a
margin that reflects the premium above some index of interest rates (usually
one-year U.S. Treasury bills) that will be used to adjust the interest rate at
specified times during the term. An adjustable-rate mortgage also stipulates
a cap, which is the maximum amount by which the rate can change at any
adjustment point, and a ceiling and floor, or the maximum and minimum
interest rate that will be charged during the life of the mortgage.
To illustrate these two mortgage instruments, suppose you wish to fi¬
nance a $100,000 house. A 30-year fixed-rate mortgage at an interest rate
Inside Money
A Lexicography for Bonds
Bonds take many forms. Definitions for some of A callable bond can be called, or redeemed,
the most common bonds are provided below. by the issuer after a specified period. Borrowers
prefer callable bonds if they are concerned that
I
A debenture is backed not by any specific
interest rates will fall in the future. If this occurs,
collateral but by "the full faith and credit" of
the borrower can call the bond, paying off the
the selling firm or institution. For example, if you
principal early, and then issue new bonds at the
take out a mortgage loan, the property you pur¬
lower interest rate. This allows the borrower to
chase is the collateral; thus, this loan is not like
refinance his or her bond debt, just as a home-
a debenture. However, a signature loan—a loan
owner might refinance a mortgage when the in¬
that you promise to pay on your good word—is
I
terest rate falls.
analogous to a debenture.
A bearer bond can be redeemed by who¬
A subordinated bond is a bond whose claim
ever holds it. It is payable "to the bearer."
on the assets of the issuer—the borrower—is
Bearer bonds are similar to U.S. Federal Reserve
subordinate to a claim by another bond. The lat¬
notes—dollar bills—which are payable "to the
ter is called a senior bond, since its claim must
bearer" on demand.
be paid before the claim of the subordinated
bond. If the issuing firm goes bankrupt, the sen¬ A registered bond is registered to the
ior claims are paid before the subordinated owner; that is, the owner's name and address
claims. are known to the issuer, and the bond is re¬
deemable by the owner.
A junk bond has a relatively high risk of de¬
fault. Junk bonds are riskier than most other
bonds and thus pay a higher interest rate.
of 9 percent, a 5 percent down payment, and 1 point works as follows. First,

you are required to pay 5 percent of the purchase price of the house as a
down payment, which amounts to $5,000. The rest, $95,000, is the amount
to be financed at an interest rate of 9 percent, with monthly payments for 30
years (360 payments in all). In addition, you pay 1 point, or 1 percent of the
loan amount, in the form of prepaid interest; in this case, 1 point amounts
to $950. The bank loans you $95,000 in return for your promise to abide by
the terms of the agreement.
In contrast, suppose you obtain a 30-year adjustable-rate mortgage for
the $100,000 house. The down payment and points work just as in the case
of a fixed-rate mortgage. However, the interest rate will vary during the
course of the loan. Suppose the initial interest rate is 7 percent but varies
each year, and the ARM stipulates a 3 percent margin, a 2-percent-per-year
cap, a 13 percent lifetime ceiling, and a 0 percent floor. Next year, if the
interest rate index is 4 percent, your interest rate for that year will remain at
7 percent (your interest rate next year is the index plus the margin, in this
case 4% + 3% = 7%). But if the index increases to 5 percent, your interest
rate next year will be 8 percent (the index plus the margin, in this case 5%
+ 3% = 8%). If the index drops to 3 percent, your interest rate will drop
to 6 percent (3% + 3% = 6%). Notice, however, that if the index increases
to 10 percent, your interest rate next year will increase to only 9 percent,
since the cap on the amount by which your rate can change in any one year
is 2 percent above the previous year’s rate (which was 7 percent). Over the
life of the mortgage, your rate will never rise above the ceiling rate of 13
percent or fall below 0 percent, regardless of how high or low the index is.
Various types of financial institutions issue mortgages, and there is an
active secondary market in which mortgages are bought and sold. If you
obtain a mortgage, it is unlikely that the original lender will hold the mort¬
gage for the life of the loan; in fact, your mortgage can change hands
numerous times during its life if the holders need to liquidate the loan for
immediate cash.
Numerous government and private agencies participate in the mortgage
market by insuring mortgages; that is, they guarantee that lenders will be
paid back in the event borrowers default. The Federal Housing Administra¬
tion (FHA) is a federal agency that insures mortgages for individuals, while
the Veterans Administration (VA) insures mortgages for veterans. The Gov¬
ernment National Mortgage Association (GNMA, or Ginnie Mae) is a gov¬
ernment organization that buys mortgages issued to low-income borrowers
and also guarantees mortgage-backed securities issued by private lenders.
The two major private companies that buy mortgages are the Federal Na¬
tional Mortgage Association (FNMA, or Fannie Mae) and the Federal Home
Loan Mortgage Corporation (FHLMC, or Freddie Mac). These two private
companies operate with an implicit government guarantee and also guarantee
mortgage-backed securities issued by private lenders.
Parts a and b of Figure 2.5 show the mortgage debt held by various
institutions in 1971 and 1991, respectively. In 1971 the total mortgage debt
held by these institutions was $524 billion, and of this amount the largest
portion, 45 percent, was held by savings institutions. Commercial banks, life
insurance companies, and other lenders each had about a 15 percent share
of the mortgage market, and federal agencies held less than 9 percent of
mortgage debt. The picture has changed dramatically in recent years. By
1991, the dollar value of mortgage debt held by all institutions had grown
to $3,919.5 billion, an increase of 648 percent! This translates into a growth
rate of about 10.6 percent per year for 21 years. In addition, notice the
change in the distribution of mortgage debt holding. Savings institutions
Figure 2.5
Mortgage Debt Held by Institutions, 1971 and 1991
This figure shows the dollar value (in billions) of mortgage debt held by major financial institutions in
1971 and 1991. The institutions include savings institutions (savings and loan associations and mutual
savings banks), commercial banks, life insurance companies, federal and related agencies (including the
Government National Mortgage Association, the Federal Housing Administration, the Veterans Adminis¬
tration, the Farmers Home Administration, and U.S.-sponsored institutions such as the Federal National
Mortgage Association, the Federal Home Loan Mortgage Corporation, Federal Land Banks, and mort¬
gage pass-through securities issued or guaranteed by these agencies), and others (which includes individ¬
uals and private mortgage pools). In 1971 savings institutions were the major holders of mortgages,
followed by commercial banks, life insurance companies, and other investors (including individuals). Fed¬
eral agencies held less than 9 percent of this debt. By 1991 the largest holders of mortgages were
federal agencies, which accounted for almost 37 percent of mortgage debt. Commercial banks were
second with 22 percent, followed by savings institutions and other investors. Life insurance companies
had slipped to less than 7 percent.
(a) 1971 (b) 1991
Federal Agencies
$46.4 Billion Federal Agencies
Commercial Banks
(8.9%) Commercial Banks $1,439.3 Billion
$82.5 Billion Life Insurance
Companies $876.3 Billion (36.7%)
(15.7%)
$75.5 Billion (22.4%)
(14.4%)
Other
$83.6 Billion
(15.9%)
Life Insurance
Savings Institutions
Companies
$705.4 Billion
Other $265.3 Billion
(18%)
Savings Institutions (6.8%)
$633.2 Billion
$236.2 Billion
(16.2%)
(45%)
Source: Economic Report of the President, (Washington D.C.: U.S. Government Printing Office, 1993).
held 45 percent of mortgage debt in 1971 but only 18 percent in 1991. The
share held by life insurance companies fell from more than 14 percent to
under 7 percent, while the share held by commercial banks increased from
less than 16 percent to more than 22 percent. The big gainer was federal
agencies, with an increase in share from less than 9 percent in 1971 to nearly
37 percent by 1991. Government or government-sponsored agencies now

lend well over $1 of every $3 in mortgage loans, and the trend appears to
be continuing.
Government Securities. How does the federal government finance

its debt, which is more than $3 trillion? To pay off the debt today, the
government would have to send a bill to every adult and child in the United
States for roughly $14,000. Since few families of four could afford to cough
up the required $56,000, the government issues more government securities.
Government securities are debt instruments issued by the U.S. Treasury
and include such instruments as Treasury bonds. The funds obtained from
the sale of these bonds are used to refinance the current federal debt as well
as current federal deficits (the amount Congress spends this year in excess
of tax revenues). We will look at government finance many times in this
book, including Chapters 16 and 17, and Chapter 21 is devoted entirely to
this topic. For now, we concentrate on the financial instruments. Because
there is an active secondary market for government securities and the instru¬
ments are backed by the full faith and credit of the U.S. government, gov¬
ernment securities are the most liquid of all capital market instruments.
Consumer and Commercial Loans. Consumer loans are loans

obtained by individuals for intermediate-term purchases such as car pur¬
chases, as well as merchandise bought with credit cards. Commercial loans
are essentially credit lines issued to businesses. There is a less active sec¬
ondary market for consumer and commercial loans, making them the least
liquid of all capital market instruments. However, there has been a growing
movement to securitize (convert to marketable securities) some consumer
debt.
Figure 2.6 shows the major categories of consumer credit in 1992.
Automobile loans constituted the largest component of consumer credit
(about 35 percent), followed closely by revolving credit (about 33 percent),
which includes personal bank lines of credit and credit issued by credit card
companies.
Municipal Bonds. State and local governments issue municipal bonds

to obtain long-term funds for such things as highways and schools. The
interest payments holders of these bonds receive are exempt from federal
income tax (although some state and local governments do collect income
tax on municipal bond interest earnings). This makes municipal bonds an
attractive investment for lenders in high-income tax brackets. Box 2.4 com¬
pares municipals with other types of bonds.
For example, a person in a 36 percent tax bracket earning 10 percent
taxable interest gets to keep only 6.4 percent after tax. A tax-free municipal
bond paying 6.5 percent or more would be an attractive investment in this
case. A secondary market exists for municipal bonds issued by large cities
This figure shows out¬ Figure 2.6

standing consumer Consumer Credit Outstanding
credit in February
1992 broken down by
various categories.
These figures are for
all nonmortgage loans
to consumers from fi¬
nancial institutions.
Automobile loans
made up about 35
percent of consumer
credit, followed by re¬
volving credit (such as
credit cards) with
around 33 percent
and other credit com¬
prising about 30 per¬
cent. Mobile home
credit lending made
up less than 3 percent
of consumer credit
outstanding.
and states, but it is less active than the corporate and federal government
bond markets. The bonds issued by smaller municipalities do not trade on a
secondary market, which makes it very difficult for a holder to liquidate a
bond prior to maturity.
Internationa/ Financial instruments

Tremendous growth has occurred in international financial instruments in
recent years. For the most part, the financial instruments traded in interna¬
tional markets function like the instruments issued in the United States. The
only difference is that the unit of account for these instruments is the local
currency of the country in which they are issued. For instance, bonds issued
in Britain are denominated in British pounds, while bonds issued in Germany
are issued in German marks. Both are foreign bonds to U.S. residents.
Eurobonds are an important exception. Eurobonds are bonds denominated
in a currency other than that of the country of origin. For example, a bond
issued in Germany but denominated (paying interest and its face value) in
U.S. dollars is a Eurobond.
The recent surge in activity in world stock and bond markets has broad¬
ened the possibilities for investors and borrowers alike. A borrower no longer
has to obtain funds from financial intermediaries in his or her own country;
similarly, a lender need not lend funds only to borrowers in its own country.
Inside Money
Are All Bonds Created Equal?
Bonds differ with respect to their face amounts bonds issued by different borrowers involve dif¬
(the amounts paid when the bonds mature), ferent risks. Corporate bonds are riskier than
maturities (the number of years before the face bonds issued by the federal government. Private
amounts are paid), and their tax status (whether rating companies, such as Moody's or Standard
interest income earned on the bonds is taxed by and Poor's, provide investors with a rating of
federal, state, or local government). Some bonds the underlying risk associated with particular
are callable; that is, the borrower may pay the bonds. The table below summarizes features of
face amount prior to maturity. In addition, some of the more common types of bonds.
Summary of Features of Various Bonds

U.S. Treasury U.S. Agency
Corporate Bonds and U.S. Treasury Bonds (FNMA, Municipal
Bonds Notes Bills GNMA, etc.) Bonds
Face amount $1,000 $1,000 $10,000 Varies $5,000
Maturity Varies Bonds: over 10 1 year or less Varies Varies
years
Notes: between
1 and 10 years
Tax status Taxable Exempt from Exempt from Some fully Exempt from
state and local state and local taxable, others federal taxes and
taxes taxes exempt from sometimes from
state and local state and local
taxes taxes
Callable Depends; often Not usually No No Depends
such provisions
included
Risk rated by Yes No; no risk of No; no risk of Sometimes Yes
private rating default default
companies?
Because of time differences across the globe (when it is 4 a.m. in New York,
it is 9 a.m. in London), international markets allow borrowers and lenders
to make financial transactions at virtually any time of day. This fact has
greatly enhanced the liquidity of financial assets that trade on exchanges
around the world.
Summary For each of the following transactions, determine (1) the type of financial
Exercise 2.3 instrument and (2) whether the trade occurred through direct finance, a
primary market transaction, or a secondary market transaction, (a) Exxon
sells 1 million new shares of its stock through a broker, (b) Martha calls her
broker and tells him to sell 100 shares of Exxon, (c) To increase its reserves,
First City Bank contacts Second City Bank to borrow $1 million overnight.
Second City Bank agrees, but the borrowed funds never leave the vaults of
the Federal Reserve.
Answer: (a) The financial instrument is Exxon corporate stock and is sold
in the primary market. Exxon receives cash when it issues stock in the
primary market, (b) The financial instrument is Exxon corporate stock and
is traded in the secondary market. Exxon receives no funds in this transaction,
(c) The financial instrument is federal funds and was exchanged through
direct finance.
Official Definitions of the Money Stock
In Chapter 1 we defined money, discussed various forms money has taken

throughout history, and listed the desirable properties of money. Now that
you have a basic understanding of various financial instruments, we focus
on the definitions of the money stock used today by the Federal Reserve
System. These measures of the money stock—the total amount of money
in the economy—are the official definitions used by the U.S. government
and its agencies.
Official definitions of the money stock are derived in two ways, which
sometimes conflict. The first approach stresses the medium of exchange
function of money and defines the money stock as only those assets that are
reasonably regarded as media of exchange. The second approach stresses
the liquidity of money and defines the money stock as the assets that are
most liquid, or most easily converted into purchasing power. Besides cur¬
rency and checkable deposits, the most liquid assets are those that are readily
convertible into money. For assets that are not themselves the media of
exchange, this requires a well-developed secondary market or some other
guarantee of quick convertibiity to money (such as the guarantee on savings
accounts). A highly liquid marketable asset not only must have a secondary
market; that market must handle a fairly high volume of transactions so that
owners of the assets are fairly certain of their ability to sell them at the going
market price. Moreover, the assets should be standardized so that the char¬
acteristics of assets offered for sale are readily ascertainable by purchasers
not physically present at the market (i.e., those making purchases by phone).
The Federal Reserve System defines a series of monetary aggregates—
ways of adding up the money in the economy into a measure of the money
Official Definitions of the Money Stock 55
stock. The money aggregate based most closely on transactions considera¬

tions is known as Ml. As we discussed in Box 1.1 of the previous chapter,
Ml contains assets that serve as the media of exchange. Other monetary
aggregates are M2, M3, and L. These are based on liquidity considerations
and include various other assets considered to be more or less readily con-
vertable into media of exchange.
mi ;
The monetary aggregate Ml is defined as Currency -4- Demand deposits +
Traveler’s checks + Other checkable deposits. Notice that Ml contains only
those assets that are readily accepted in exchange for goods and services.
Thus, Ml is the most narrowly defined monetary aggregate.
M2
The monetary aggregate M2 is defined as Ml + Savings deposits + Small
time deposits + Money market deposit accounts + Money market mutual
shares (not owned by institutions) + Overnight repurchase agreements +
Overnight Eurodollars (specifically U.S. residents’ holdings of overnight
offshore deposits in U.S. branches). M2 includes all assets in Ml and adds
savings and time deposits, money market deposit accounts, and money mar¬
ket mutual funds that can all be readily converted into purchasing power.
Thus, while you cannot trade your savings passbook for a new pair of
designer jeans, you can trade your savings account for currency or a checking
account at your bank and then purchase the jeans. Similarly, a money market
account has restrictions on the number and dollar amounts of checks you
can write to withdraw funds, but you can deposit withdrawn funds in large
dollar denominations into your checking account, from which you can pur¬
chase goods and services in any denomination. The last two items in the
definition of M2—overnight repurchase agreements and Eurodollars—in¬
volve items not typically held by individuals, but they are included because
they can be converted into M1 overnight.
M3
The monetary aggregate M3 is defined as M2 + Large-denomination time
deposits + Money market mutual shares (institutional) + Term repurchase
agreements + Term Eurodollars. Thus, M3 includes everything in M2 and
adds assets that are less liquid. For example, term repurchase agreements
and term Eurodollars cannot be converted into Ml until the term, which is
longer than overnight, expires. Likewise, large time deposits and institutional
money market mutual shares are less easily converted into M1 than the assets
included in M2 but not in M3.
L
The monetary aggregate L consists of M3 + Short-term Treasury securities
+ Commercial paper + U.S. savings bonds + Banker’s acceptances. The
broadest monetary aggregate defined here, L consists of M3 plus highly
liquid assets that the Federal Reserve System does not consider money, such
as U.S. savings bonds—hence the name L instead of Ml, M2, or M3.
Recent attempts to provide alternative definitions of the money stock
stress that there is no reason to simply add up the quantity of dollars held
in various assets and name that sum M. Instead, some economists have
argued that various assets should be weighted by their liquidity and then
added together to obtain a monetary aggregate. While this approach to
defining monetary aggregates is attractive on theoretical grounds, it has not
led to a universally accepted alternative definition of the money stock.
Figure 2.7 shows trends in Ml, M2, M3, and L for the period 1959
through early 1993. Notice that Ml is less than M2, which is less than M3
and L. This is, of course, due to the fact that the assets contained in Ml are
also in M2, and so on for M3 and L. Furthermore, M2, M3, and L increased
much more over this period than Ml. The primary reason for this was the
growing importance of financial instruments such as commercial paper,
money market mutual funds, and the like, which are not included in Ml.
A natural question is which of the official definitions of the Ms is best.
The honest answer is that economists don’t really know. Most economists
think either Ml or M2 is the best definition of the four official Federal
Reserve System definitions given here, but even here there is disagreement.
In the 1970s, Ml appeared to be the best definition from a government policy
standpoint; changes in M1 seemed to be linked to changes in unemployment
or output. However, in the early 1980s this relationship broke down, and
many economists looked to M2 as the more reliable policy variable. In the
1990s, this relationship too looks tenuous. In the next chapter, we will
examine this issue in greater detail as we look at the relation between the
money stock and what policymakers call the price level and gross domestic
product.
Summary What is the primary difference in the monetary aggregates Ml, M2, M3,
Exercise 2.4 and L?
Answer: Ml consists of the most liquid assets, M2 consists of Ml plus

the next most liquid assets, M3 consists of M2 plus the next most liquid
assets, and L consists of M3 plus the next most liquid assets. Thus, Ml is
the narrowest of the four monetary aggregates (it does not include some
assets that are included in M2, M3, and L). L is the broadest monetary
aggregate (it includes everything in the other aggregates, plus some addi-
CONCL USION 57
tional assets). Moreover, the assets included in Ml tend to serve as media

of exchange, whereas many of the assets included in M2, M3, and L do not,
although they can be easily converted into various forms of Ml.
Conclusion
In this chapter, we overviewed money and banking in today’s economy and

saw how financial markets operate. We distinguished between the market
for new issues of financial assets (the primary market) and the market for
existing assets (the secondary market). Financial markets can be further
distinguished by the characteristics of what is traded on the market (debt,

equity, or financial services). The principal participants in financial markets
are depository institutions (e.g., commercial banks and savings and loans)
and nondepository institutions (e.g., insurance companies and pension
funds). We noted that the recent deregulation of the banking industry has
blurred the distinction between banks and other depository institutions, and
today savings and loans, credit unions, and money market funds all perform
many of the functions traditionally filled by commercial banks.
We also examined money market and capital market instruments. Money
market instruments include assets that mature in one year or less, such as
Treasury bills, commercial paper, and repurchase agreements. Capital market
instruments have maturities exceeding one year and include mortgages, cor¬
porate stocks and bonds, and long-term government securities such as Treas¬
ury bonds.
Finally, we saw how the most liquid financial instruments are included
in official definitions of the U.S. money stock. Ml is the most narrowly
defined monetary aggregate and includes currency, demand deposits, trav¬
eler’s checks, and other checkable deposits. M2 is a slightly broader measure;
it consists of Ml plus various types of savings accounts, as well as overnight
repurchase agreements and Eurodollars. The broadest measure of the money
stock, known as L, includes a host of instruments like savings bonds and
short-term Treasury securities. In the next chapter, we will see how changes
in the money stock can affect factors like inflation and output.
KEY TERMS
financial market certificate of deposit (CD)

primary market Treasury bill (T-bill)
secondary market repurchase agreement (repo)
debt market Eurodollars
short-term asset banker’s acceptance
intermediate-term asset federal funds
long-term asset capital market instrument
equity market corporate stock
financial service market corporate bond
depository institution mortgage
commercial bank government security
savings and loan association (S&L) consumer loan
mutual savings bank commercial loan
KEY TERMS continued
credit union municipal bond

money market mutual fund Eurobond
nondepository institution money stock
insurance company monetary aggregates
pension fund Ml
brokerage firm M2
money market M3
commercial paper L
1. How are financial intermediaries similar to come more important sources of funds for
supermarkets? How do they differ? banks? Why or why not?
2. Critically evaluate and interpret the fol¬ 7. Explain how a banker’s acceptance differs
lowing statement: “Since corporate bond¬ from an ordinary bank draft.
holders do not own a share of the under¬
8. “Federal funds are dollars loaned to
lying company, they don’t care whether
banks by the Federal Reserve at the fed¬
the company earns profits or makes
eral funds rate.” Is this statement true or
losses.”
false? Explain.
3. Critically evaluate and interpret the fol¬
9. If a corporation needs short-term funds,
lowing statement: “Since firms do not re¬
does it obtain the funds in the money mar¬
ceive money when their stock is sold in
ket or in the capital market?
the secondary stock market, they would
benefit from the elimination of that mar¬ 10. Explain the terms of an 8 percent, 30-year
ket.” fixed-rate mortgage that requires 10 per¬
cent down and 2 points to a person wish¬
4. How do depository financial institutions
ing to buy a $200,000 house.
differ from nondepository institutions?
11. A 30-year ARM with an initial interest
5. Rank the following financial instruments
rate of 8 percent, a 2 percent margin, a 15
from most liquid to least liquid.
percent ceiling, a 7 percent floor, and a 3-
(a) . A $1 bill
percent-per-year cap is available. The cur¬
(b) . A share of IBM stock
rent interest rate index is 6 percent.
(c) . A municipal bond issued by a small
(a) . Explain these terms to a person wish¬
town in Alaska
ing to purchase a $150,000 house.
(d) . A U.S. Treasury bill
(b) . If the interest rate index increases to
6. As we move toward a more global econ¬ 9 percent next year, what interest rate will
omy, would you expect Eurodollars to be¬ be paid on the mortgage?
12. Why are municipal bonds attractive to in¬ lion in Eurodollars, and $75 million in
dividuals and corporations with high in¬ Treasury securities. Determine the money
comes or profits? stock as measured by
(a) . Ml
13. Which of the monetary aggregates con¬ (b) . M2
tains only those financial instruments that (c) . M3
serve as media of exchange? (d) . L
14. Suppose the economy has $100 million in 15. Why are credit cards not included in any
currency, $50 million in demand deposits, of the definitions of the money stock pre¬
$150 million in savings deposits, $25 mil¬ sented in this chapter?
Selections for further Reading

Hutchinson, P. M., J. R. Ostas, and J. D. Reed. “A Morrissey, T. F., and O. K. Gregory. “A Study of
Survey and Comparison of Redlining Influences in FHLB Advances and Their Limit under Efficient
Urban Mortgage Lending Markets.” American Intermediation.” Journal of Economics and Busi¬
Real Estate and Urban Economics Association ness, 27 (Winter 1975), 122-130.
Journal, 5 (Winter 1977), 463-472. Ostas, J. R. “The Federal Home Loan Bank System:
Knodell, J. “Open Market Operations: Evolution and Cause or Cure for Disintermediation?” Journal
Significance.” Journal of Economic Issues, 21 of Monetary Economics, 8 (September 1981),
(June 1987), 691-699. 231-246.
Kopecky, K. J. “Nonmember Banks and Empirical Van Fenstermaker, J., J. E. Filer, and R. S. Herren.
Measures of the Variability of Reserves and “Money Statistics of New England, 1785-1837.”
Money: A Theoretical Appraisal.” Journal of Fi¬ Journal of Economic History, 44 (June 1984),
nance, 33 (March 1978), 311-318. 441-453.
CHAPTER
3 Money, Inflation, and Interest
n 1970, the United States produced about $1 trillion worth of final

goods and services—products like cars, homes, and textbooks. This year,
more than $6 trillion worth of final goods and services will be produced.
Does this mean six times more goods and services are available today than
in 1970? Or does it mean you must pay six times more for goods and services
than people paid in 1970? To address these questions, this chapter introduces
what economists call economic growth and inflation. We will learn that
inflation is a situation of generally rising prices. During the past quarter-
century, much of the increase in the dollar value of goods and services
produced was due to inflation—the fact that things cost about four times
more today than in 1970. However, economic growth—the presence of more
things to consume today than in 1970—also contributed to the increased
dollar value of goods and services. In the first part of this chapter, we examine
the major causes of inflation.
In the second part of the chapter, we overview interest rates and see how
inflation affects the interest rate you pay on loans such as car loans and
mortgages. We conclude with a look at how income taxes affect interest
rates. Since the material in this chapter will play a key role in our economic
analysis of financial markets and banks in the next three chapters, we en¬
courage you to master the material presented in this chapter before contin¬
uing your study of money and banking.
Distinguishing Between Inflation and Economic Growth
When your grandparents talk about paying a nickel for a loaf of bread or a
quarter for a ticket to see Gone With the Wind, they are telling you prices
that existed in the past. Today you might pay $6 or more for a movie ticket
or $1.50 for a loaf of bread. A portion of the increase in prices of these items
reflects inflation: a general rise in the level of prices in the economy. In
contrast, when your parents tell you they didn’t own a car or a TV when
they got married, they are telling you about the real quantity of goods
available in the past. The increased quantity and variety of things available
to consume today reflects economic growth, and is the result of technological
advances and increases in the pool of productive resources (e.g., the size of
the work force).
61
62 Chapter3 Money, Inflation, and Interest
The Price Level and Real Output

In this section, we look at how economists distinguish between inflation and
economic growth. We begin by defining terms you may have heard on the
evening news or read in the newspaper: the price level and real output. These
definitions are not ends in themselves; rather, they are used at the end of
this section, where we examine the relationship between money and inflation.
Measuring the Price Level. The price level is a measure of the

average prices of goods and services in the economy. It serves as a gauge
of the general purchasing power of money. The consumer price index (CPI)
is the measure of the price level most familiar to Americans—and for good
reason. The percentage change in the CPI is published each month in news¬
papers throughout the nation. Many union contracts have cost-of-living ad¬
justments that are indexed to the CPI; that is, the contracts specify that wages
will rise when the CPI increases. Social security benefits are also indexed to
the CPI, and each year social security recipients find that their retirement
benefits increase by the same percentage as the increase in the CPI.
The Bureau of Labor Statistics (BLS) calculates the CPI, which measures
the cost of the “basket” of goods and services purchased by the average
urban household. The items in the basket are determined from a survey
conducted in the base year, which was 1987 at the time of this writing. Thus,
the CPI reports how much more or less expensive the fixed base year basket
of goods and services would be in different years. If the CPI is higher today
than in 1987, it costs more to buy the basket of goods and services today
than it did in 1987. A lower CPI means it costs less to buy the basket today
than it did in 1987.
In addition to the CPI, economists use several other measures of the
price level to track prices in the United States. These include the producer
price index (PPI) and the implicit price deflator for gross domestic product.
The primary difference between these alternative measures of the price level
is the composition of the basket of goods and services used to measure price
changes. Historically all these measures of the price level have provided
similar measures of price movements, so we will not concern ourselves with
the minor technical distinctions among them. All these measures share a
common feature: Their values are normalized to equal 100 in the base year—
the year corresponding to the basket of goods and services that is being
priced over time. If the price level was 100 in 1987 and increased to 103 in
1988, the price of the average good in the United States increased by 3
percent between 1987 and 1988.
All three indices follow this custom of stating price levels with a base
year value of 100. Thus, you will see something like “1987 = 100” to
indicate that 1987 is the base year.
Measuring Nominal and Real Output. Nominal output refers to

the current dollar value of the final goods and services produced in the
economy. Gross domestic product (GDP), the most commonly used measure
of nominal output, is the total dollar value of all final goods and services
produced in the economy in one year. Recall from your principles course
that GDP measures the current dollar value of final output; that is, it measures
only the output of goods and services sold to final consumers of the products.
Intermediate sales of, say, paper products from a sawmill to a book publisher
such as Houghton Mifflin or of a book from Houghton Mifflin to a college
bookstore are not incuded in GDP. The sale of a book by the bookstore to
you, the final consumer, is included in GDP. Moreover, GDP includes only
new output, not sales of used goods. If you sell your textbook as a used
book at the end of the semester, your transaction is not included in GDP.
Since nominal output is the current dollar value of all final sales of newly
produced goods and services in the United States, it is the summation of the
quantities of final goods and services multiplied by the products’ prices. We
can represent this mathematically by using P to denote the price level and
letting Y represent real output. Real output is a measure of the physical
quantity of goods and services available to the final consumers of the items.
In this case, the dollar (or nominal) value of the real goods and services in
the economy is simply the price level times the real quantity of goods:
Nominal output = P X Y.
Notice that if the price level (P) doubled but the level of real output (Y)
remained the same, nominal output would double even though no additional
goods and services are available. For this reason, when the price level
increases, nominal output will increase even if real output remains constant.
To avoid confounding growth and inflation, economists frequently use real
output to measure the output of the economy. Real output is obtained by
dividing nominal output by the price level:
Nominal output
P
In the United States, economists use GDP to measure nominal output

and the implicit price deflator to measure the price level associated with
GDP. If the base year is 1987, the price level effectively converts nominal
values of output today into the dollar values that would prevail today if
prices remained at their 1987 levels. In this case, real output is called real
GDP and is obtained by dividing GDP in a given year by the price level in
that year. For example, at the end of 1992, GDP was $6,061 trillion and the
reported GDP deflator was 121.72. Consequently, real GDP in 1992 was
GDP 1992 6.061

Real GDP1992 = -- = - = 4.979 trillion 1987 dollars.
Pl992 1.2 1 72
(We divided by 1.2172 instead of by 121.72 because the reported GDP

deflator is the actual price level times 100.) Our calculation reveals that had
prices in 1992 remained at their 1987 levels, the value of final goods and
64 Chapter 3 Money, Inflation, and Interest
services produced in 1992 would have been $4,979 trillion—considerably

less than the $6,061 trillion nominal value of output (GDP) in 1992.
Figure 3.1 provides historical trends in the nominal value of output
(GDP), real output (real GDP), and the price level (GDP deflator) in the
United States. Due to differences in the magnitude of the numbers, GDP and
real GDP are measured on the scale on the right, while the GDP deflator is
measured on the left. Notice that in 1987 (the base year), real GDP and GDP
coincided, since the price level was 100 in the base year. Also notice that
GDP was $6 trillion in 1992, which is considerably greater than the $1
trillion in final goods and services produced in 1970.
Why was GDP higher in 1992 than in 1970? Figure 3.1 reveals two
reasons. First, real GDP was higher in 1992 than in 1970, which tended to
increase GDP (remember that GDP = P X Real GDP). Second, the price
level as measured by the GDP deflator increased from 34.5 in 1970 to 121.7
by the end of 1992. This reveals that average prices increased by 252.8
percent over this period.
Inflation and Economic Growth

Now that you have a basic understanding of the price level and real output,
we will show you how to use these measures to calculate the rate of inflation
and the rate of economic growth.
Source: U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Busi¬
ness, various issues, and Citibase electronic database.
The inflation rate is the rate of change in the price level.

Inflation rates are stated as a percentage change on an annual basis. For
instance, if the price level is Pt in year t and Pt-\ in year t — 1, the inflation
rate (tt) between years t and r — 1 is defined as
For example, the price level as measured by the implicit price deflator was
117.8 in 1991 and increased to 120.8 in 1992. Thus, the inflation rate between
1991 and 1992 was
P1992 ~ P\99\ 120.8 — 117.8

tr = - = - = .025,
Pl9 9i 117.8
or 2.5 percent. In other words, average prices in the United States increased
by 2.5 percent between 1991 and 1992.
We emphasize that inflation is a persistent, general rise in the average
prices of all goods. Literally millions of goods can be purchased in our
economy. If the price of only one good increases by 5 percent, that increase
does not reflect inflation; rather, it is an increase in the price of that single
commodity. But if the average prices of all goods in the economy increase
each year by, say, 5 percent, then we say the inflation rate is 5 percent.
Box 3.1 describes the fall in inflation rates in five major economies
between 1981 and 1991.
Economic Growth. Economic growth is the rate of change in real

output. The economic growth rate is usually stated as a percentage change
on an annual basis. If real output was Yt in year t and Yt_ j in year t — 1,
the economic growth rate between years t and r — 1 is defined as
SY -
Yt -
y
U-,
It~ 1
For example, real output (real GDP) was $4796.7 billion in 1991 and in¬
creased to $4873.7 billion in 1992. Thus, the economic growth rate between
1991 and 1992 was
Y 1992 1991 4873.7 - 4796.7

gY = .016,
1991 4796.7
or 1.6 percent. In other words, between 1991 and 1992, the real quantity of
final goods and services produced in the United States increased by 1.6
percent.
The Relationship Between inflation and Economic Growth.

We can use a very simple formula to relate the rates of economic growth,
growth in nominal output, and inflation. The growth rate of GDP equals the
growth rate of real GDP plus the growth rate of prices (the inflation rate).
inflation Rates Around the World
The United States experienced rapid inflation 1981 than in 1991. In 1981, inflation rates in
during the 1970s and early 1980s. In 1981, for the U.K. and Canada rivaled those in the United
instance, the U.S. inflation rate was 11.24 per¬ States. But by 1991, these countries had experi¬
cent. This means that in the United States goods enced a dramatic decline in inflation, just as the
and services cost more than 11 percent more at United States had. Japan and Germany experi¬
the end of the year than at the beginning. enced lower inflation in 1981 than the other
Workers who did not receive wage increases of three major Western nations. Still, their inflation
at least 11 percent found themselves unable to rates were lower in 1991 than in 1981.
afford the same standard of living they enjoyed The moral of the story is that some experi¬
at the beginning of 1981. By 1991, the U.S. in¬ ences of the U.S. economy are also the experi¬
flation rate had fallen to about 4 percent. ences in other nations. We truly live in a global
The accompanying figure reveals this experi¬ economy.
ence was not purely an American phenomenon.
All five of these major countries—the United Source: U.S. Department of Commerce, Bureau of the
States, Japan, Germany, the United Kingdom, Census, Summary of U.S. Export-Import Merchandise
Trade-FT900, various issues, and Citibase electronic data¬
and Canada—experienced higher inflation in
base.
Thus, using g to represent growth rates, we can calculate the growth rate of
real GDP as
great GDP gnominal GDP gprice level?
where the subscripts refer to what is growing.1 This formula reveals that the
growth rate in real GDP (the rate of economic growth) equals the growth
rate in nominal GDP (called the nominal growth rate) minus the growth in
the price level (inflation). For instance, if prices increase by 5 percent per
year and nominal GDP increases by 7 percent per year, real output increases
by only 2 percent per year, since 7% — 5% = 2%.
Figure 3.2 plots the growth rates of GDP, real GDP, and the GDP deflator
for the period 1980-1992. As we showed earlier, the arithmetic difference
between the growth rate of nominal GDP and the growth rate in the price
level equals the rate of economic growth in each of these years. Notice the
volatile movement in these growth rates over this period. For instance, in
1980 the inflation rate (as measured by the growth rate in the GDP deflator)
was around 10 percent. By the end of 1992, the inflation rate was less than
3 percent. This reveals that inflation can vary considerably from year to year.
In the remainder of this section, we will examine why inflation tends to vary
over time. To do so, we must first look at the major causes of inflation.
Causes of Inflation
What causes inflation? As we see below, anything that causes the growth
rate of the money supply to increase, the growth rate of velocity to increase,
or the growth rate of output to decrease causes inflation.
Growth in the Money stock. Imagine that the room in which you
are sitting is the economy and you are the sole consumer in that economy.
Suppose one liter of soda is the only good in the room and the total money
stock (the money in your pocket) is $1. Clearly, then, the maximum price
you could pay for the soda is $1. Since the money in your pocket is fiat
money, you cannot consume it; all you can consume is the one liter of soda.
Now suppose that by some miracle, another dollar appears in your
pocket. The total money stock is now $2, but only one liter of soda is still
available. You would now be willing to pay a price of $2 for the liter of
soda (since you can’t drink the $2 in fiat money). Since the money stock
doubled, the price of soda doubled, and thus nominal output doubled as well.
But since there is still only one liter of soda in the economy, real output is
constant and you are no better off than before the extra dollar appeared. The
increase in the money stock led to an increase in prices and nominal output
1 This formula is an approximation. The exact formula is greal c.dp = (gnominai gdp gprice levelVt i
T <?price level)-
V
Source: U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Busi¬
ness, various issues, and Citibase electronic database.
but no increase in the real amount of goods available in the economy. In

this sense, increases in the money stock lead to a higher price level.
You may wonder how such a simple story could explain something as
complicated as inflation in the U.S. economy. Figure 3.3, which graphs two
indices of the money stock (Ml and M2) and an index of the price level for
the period 1959-1993, provides an answer. The figure suggests that periods
of rising prices tend to be associated with periods of increases in the money
stock. Such a relationship between growth in the money stock and inflation
is precisely what we would expect based on our hypothetical example.
Velocity and Economic Growth. Our one-soda economy is clearly

an oversimplification. Yet the same logic explains the relationship between
the money stock and the price level in an economy with many individuals
and goods and services. The only additional complication is that with many
people and goods and services, we have to take into account the velocity of
money, which measures the number of times the average dollar changes
hands in the economy in one year. Why does velocity matter? With many
people and goods but only a single dollar in money, something remarkable
happens: People can make more than $1 in purchases. You can use the $1
to buy something, the person you pay can in turn use the same $1 to buy
something from someone else, and so on. If velocity is 2, the average dollar
bill changes hands two times; $1 in money leads to $2 worth of purchases.
If velocity is 10, $1 in money leads to 10 transactions, or $10 worth of
purchases.
More generally, if the money stock is M and velocity is V, the total
dollar value of transactions in the economy is MV. Similarly, if P is the price
level and Y is real output, the dollar value of this output—nominal output—
is P X Y. Since the dollar value of transactions equals the dollar value of
the goods and services in the economy, it follows that
MV = PY.
This fundamental relationship between the dollar value of exchanges (trans¬

actions) and nominal output is known as the equation of exchange.
The equation of exchange can be used to obtain a detailed picture of the
causes of inflation. In particular, it implies that
AM AT AP AT
+ — + ,
M V P Y
This graph plots the Figure 33

measures of the Relationship Between Money and the Price Level, 1959-1993
money stock known
as Ml and M2 and
the price level from
1959 through 1993.
The money stock is
graphed on the right-
hand scale in billions
of dollars. The price
level—the GDP defla¬
tor—is graphed on the
left-hand scale and is
set to equal 100 in
1987, the base year.
Source: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, Table
1.21, various issues, U.S. Department of Commerce, Bureau of Economic Analysis, Survey of
Current Business, various issues, and Citibase electronic database.
where A means “a change in.” In words, A M/M is the percentage change

in (or growth rate of) the money stock, AV/V is the percentage change in (or
growth rate of) velocity, A P/P is the percentage change in prices (or the
inflation rate), and AY/Y is the percentage change in (or growth rate of) real
output.2
This admittedly complicated relation is easier to visualize if we let gM
represent the growth rate in the money stock, gv the growth rate in velocity,
gY the growth rate in real output, and tt the inflation rate and rewrite the
above equation as
gM + gv = 77 + §Y-
This equation simply says that the growth rate in the money stock (gM) plus
the growth rate in velocity (gv) equals the inflation rate (it) plus the growth
rate in real output (gy). We can rearrange this formula to obtain an expression
for the inflation rate in terms of the growth rates of money, velocity, and
real output:
Tt - 8m + gv — gY-
The inflation rate thus equals the growth rate of the money stock, plus the
growth rate of velocity, minus the growth of rate in real output. Changes in
any of the three growth rates on the right-hand side of this equation can lead
to changes in the inflation rate.
The so-called classical economists hypothesized that the growth rate of
velocity and the growth rate of output are relatively constant and unaffected
by changes in the growth rate of money. In this case, changes in the growth
rate of money would lead directly to changes in the inflation rate. This is
the simplest form of the famous quantity theory, which attributes rises in
the price level to changes in the quantity of money. With regard to growth
rates, changes in the growth rate of the money supply lead to changes in the
inflation rate.
More generally, suppose velocity is constant so that its growth rate is
zero. In this case, the inflation rate is simply the difference between the
growth rate of the money stock and the growth rate of real output:
^ — gM §Y-
If the money stock increases at 10 percent per year but real output increases
by only 3 percent annually, the inflation rate is 10% — 3% = 7%. This
fundamental relation reveals that when velocity is constant, (positive) infla-
2 This formula is also an approximation that holds for small percentage changes in M, V, P, and
Y. The exact formula is
AM AV /AM AV\ AP AY Iap at\

+ - + - X — = - + + — X —
M V \ M v P Y \ P Y
tion is caused by increases in the money stock that exceed the growth rate
in real output: Too much money is chasing too few goods and services.
Of course, velocity need not be constant. During periods when individ¬
uals have inflationary expectations—that is, they expect prices to rise over
time—velocity may grow. This is because inflationary expectations induce
individuals to convert money into goods more rapidly in an attempt to buy
things before prices increase. Since the inflation rate increases as the growth
rate in velocity increases, growing inflationary expectations can in them¬
selves fuel inflation.
We will say much more about these and other causes of inflation in the
second half of this book. For now note that, as Figure 3.4 shows, velocity
has not been constant in the United States. In fact, the velocity of Ml
increased fairly steadily until 1980, after which its movements became more
volatile. Currently the velocity of Ml is about 6.5, indicating that the average
dollar of Ml changes hands 6.5 times each year. The velocity of M2 is more
stable varying about a mean of around 1.7.
1.21, various issues, U.S. Department of Commerce, Bureau of Economic Analysis, Survey of
Current Business, various issues, and Citibase electronic database.
Summary In 1980 the price level was 71.73, Ml was $397.9 billion, and nominal
Exercise 3.1 output was $2,708 billion. In 1981 the price level increased to 78.88 and
Ml increased to $426.8 billion, (a) Assuming the base year of the price level
is 1987, what was real output in 1980? (b) What was the inflation rate
between 1980 and 1981? (c) What do you think led to this inflation?
(a) Since the base year of the given price level is 1987, real
output is measured in 1987 dollars. Specifically, real output in 1980, meas¬
ured in 1987 dollars, was
Nominal output1980 2,708

Real output1980 = 3,775.3 billion 1987 dollars.
Pi980 .7173
(b) The inflation rate between 1980 and 1981 was
78.88 - 71.73
tr = = .0997,
71.73
or about 10 percent, (c) During the same period, the money stock increased
at an annual rate of
426.8 - 397.9
Sm
sm — - 397.9 = -0726,
or about 7 percent. Since tt = gM + gv — gY, we know that 7 percent of

the 10 percent inflation was due to the 7 percent increase in the money stock.
The other 3 percent was due to an increase in the difference, gv — gY; the
growth rate of velocity exceeded the growth rate of real output by 3 percent.
This growth in velocity was probably due to growing inflationary expecta¬
tions, brought on by rapid increases in the price level during the late 1970s.
Interest Rates
Up to this point, we have examined the relationship between the growth

rates of money, output, and velocity and the inflation rate. We have seen
that for constant growth rates of output and velocity, an increase in the
growth rate of money will lead to an increase in the inflation rate. Now we
introduce interest rates, which play an integral role in financial markets.
There are many determinants of interest rates, among which is the rate of
inflation. In this chapter, we draw out the relationship between the inflation
rate and the interest rate. In Chapter 4, we will see how the market forces
of supply and demand ultimately determine interest rates, such as the interest
rate you pay on a car loan and the rate IBM pays when it issues a bond to
obtain funds. Moreover, we will learn in later chapters that changes in the
Interest Ra tes 73
money stock can affect interest rates. Before we tackle these more compli¬
cated issues, we need a basic overview of interest rates.
In the first part of this section, we look at what interest is and examples
of how interest is paid on a few common types of loans. Next, we examine
the important principles of present value, future value, and the time value
of money. Then we use these concepts to see how inflation affects the “real”
interest received by those owning interest-bearing financial instruments.
What Is Interest?
When a bank or other financial institution lends you money, it requires you
to repay the funds lent (the principal), plus an additional payment called
interest. The timing of the payment of interest and principal are negotiable
and vary from one loan to another. However, virtually all loans (except
perhaps loans from close friends or relatives) require some payment in
addition to the principal.
The interest rate on a loan is the ratio of the interest payments on the
loan to the principal amount (the amount borrowed). In other words, it is
the cost of borrowing funds as a percentage of the amount borrowed. If you
borrow $500 to buy a fancy stereo and pay the money back next year plus
$50 interest, you have paid an interest rate of 10 percent, calculated as
$50/$500 = .10 or 10%. This example oversimplifies, however. To compare
the interest rate on loans with different maturities (such as 1-year and 30-
year loans), we need to calculate the interest rate for a given time span. For
this reason, interest rates are stated as interest rates per year (the so-called
“annual rate of interest” you have probably seen in TV ads or on a sign at
your bank). The interest rate is the amount over and above the principal
amount that is paid in a given year, expressed as a percentage of the principal
amount. Also, interest can be indirect, as Box 3.2 shows.
When you consider interest charges on a loan, you need to realize that
interest is charged because funds available for use today are more valuable
than funds available for use in the future. This is because people prefer to
satisfy their desires today to waiting until later. (Would you prefer to receive
a $1,000 gift today or a $1,000 gift in 30 years?) Borrowers want funds for
use today and promise to repay those funds in the future. But a lender requires
more than that: A lender requires interest from the borrower as compensation
for the lost use of the funds for the length of the loan. In essence, interest is
the price of borrowing money. The interest rate expresses this price as the
percentage of the principal amount that the borrower pays to use someone
else’s money. A brief description of how interest payments are structured
on some common types of loan contracts follows.
A Simple Loan. A simple loan allows the borrower the use of a given
amount of funds (the principal) for a specified period of time. In exchange,
Points and Mortgage Interest Rates
A telephone call to a mortgage company on gage interest rate involves more points. A bor¬
April 12, 1993, allowed us to obtain the follow¬ rower can avoid paying points by paying a 7%
ing interest rates on a 30-year, conventional percent interest rate. In this case, the monthly
fixed-rate mortgage (you might call a local payments on a 30-year, $100,000 loan are
lender today to see what the current rates are in $707.79. If the borrower pays 2 points ($2,000
your area): in this case), the interest rate is only 7% percent,
Rate Points and the monthly mortgage payments drop to
7% 0 $690.68.
7% 2 In effect, a home buyer shopping for a
Notice that the interest rate varies depending on mortgage must decide whether to pay points
how many points are paid on the loan. Recall now in exchange for lowering the mortgage
from Chapter 2 that points are the percentage payments by $17.11 each month over the life of
of the loan amount that must be paid when the the loan (360 months). If the borrower plans to
loan is obtained and is a form of prepaid inter¬ stay in the house for 30 years and cannot invest
est. On a $100,000 loan, V2 points amount to the $2,000 at as high an interest rate, it is prob¬
$500 and 2 points equals $2,000. Other things ably best to pay the points. In contrast, if the
equal, borrowers prefer mortgages that involve borrower can invest the $2,000 at a higher rate
fewer points. of interest or plans to sell the house in two or
The catch, which the above table reveals, is three years, it is best to pay the higher interest
that other things are not equal. The lower mort¬ rate in lieu of the points.
the borrower must repay the principal on a specified date, at which point he
or she also makes the full interest payment. For example, First National
might lend you $1,000 and require you to pay back $1,050 (the principal
plus a $50 interest payment) in six months.
A Fixed-Payment Loan. With a fixed-payment loan, a borrower ob¬

tains funds in return for a series of fixed payments (usually monthly) that
include both principal and interest. The interest rate is usually fixed for the
length of the loan, and the regular payments are of equal amounts. Each
fixed payment includes two parts: an interest payment and a payment to
Interest Ra tes 75
reduce the principal. Thus, with a fixed-payment loan, the outstanding bal¬
ance (principal) declines during the life of the loan. The loan is paid in full
when the principal balance declines to zero. Most auto loans are of this type,
as are fixed-rate home mortgage loans. Box 3.3 illustrates how the payments
on these loans will vary with the length of the loan, using 15 and 30 year
mortgage loans as an example.
A Coupon Bond. A coupon bond specifies (1) a face value, (2) a term,
and (3) the coupon rate, which is the percentage of the face value that will
be paid annually as interest when the holder redeems coupons attached to
the bond at specified points in time (say, every six months). These coupons
gave coupon bonds their name. Upon maturity, the bondholder redeems the
bond for the face value. If the bond is sold to the lender at face value (or
par), the face value is the principal and the coupons represent the interest on
the bond. For example, suppose your bank purchases a 10 percent coupon
bond at the face value of $10,000. In this case, the principal is $10,000—
the price at which the bond is sold to the bank. Annual interest receipts are
$1,000, or 10 percent of the principal amount. Sometimes coupon bonds sell
at other than face value, as we will see in Chapter 8.
A Zero Coupon Bond. As the term implies, the owner of a zero

coupon bond receives no regular interest payments, since the bond has no
coupons attached. Instead, the bond is sold at a discount from its face or
redemption value. The change in value from its price at issue to its redemp¬
tion value is implicit interest that the purchaser receives on the bond. For
example, a zero coupon bond might have a redemption value of $ 1,000 one
year from today. A lender might purchase this bond for $900, meaning the
interest on the loan is effectively $100. The implicit interest rate is then
$100/$900, or 11.11 percent.
The Time Value of Money

Now that you have a basic understanding of what interest is and how it is
paid on several types of loans, we look at two basic tools used to evaluate
the consequences of interest: present and future value analysis. Mastering
future value analysis will simplify present value analysis, so we encourage
you to make sure you fully understand the material in the next section before
you read on.
Future Value Analysis. Imagine that you put $1,000 in the bank today
and the bank agrees to pay you an interest rate of 5 percent per year. Future
value analysis provides a way to determine how much you will have in the
future.
15-Year Versus 30-Year Mortgages
Suppose you borrow $100,000 at 7 percent in¬ would cost you only $898.83 per month to pay
terest to buy your dream house. Your banker in¬ off the loan in 15 years—only $233.53 more per
I forms you that your monthly payments will be

$665.30 for the next 30 years. How much
would you have to pay each month if you
month than with the 30-year payment schedule.
How can this be?
To answer this question, suppose you take
wanted to pay off the loan in 15 years instead out a 30-year loan for $100,000 on January 1,
of 30? Most borrowers incorrectly believe the 1995. The first year your payment schedule,
monthly payment required to pay off the loan in which consists of interest and principal, will look
15 years would be $ 1,330.60—twice the as follows:
amount on the 30-year mortgage. In reality, it
Date Payment Interest Principal Balance After
01/01/95 665.30 583.33 81.97 99,918.03

02/01/95 665.30 582.86 82.44 99,835.59
03/01/95 665.30 582.37 82.93 99,752.66
04/01/95 665.30 581.89 83.41 99,669.25
05/01/95 665.30 581.40 83.90 99,585.35
06/01/95 665.30 580.91 84.39 99,500.96
07/01/95 665.30 580.42 84.88 99,416.08
08/01/95 665.30 579.93 85.37 99,330.71
09/01/95 665.30 579.43 85.87 99,244.84
10/01/95 665.30 578.93 86.37 99,158.47
11/01/95 665.30 578.42 86.88 99,071.59
12/01/95 665.30 577.92 87.38 $98,984.21
Total for 1995 $7,983.60 $6,967.81 $1,015.79
Continued on p. 77
After one year, the bank will pay you interest of 5 percent on the $ 1,000,
which amounts to
.05 X $1,000 = $50.
Thus, at the end of one year, you will have your principal of $1,000 plus
$50 in interest for a total of $1,050. Notice that this amount can also be
calculated as follows:
(1 + .05) X $1,000 = $1,050;

Interest Ra tes 77
Continued from p. 76
Thus, at the end of the first year, your In contrast, if you obtain a 15-year mort-
$7,983.60 in payments reduced your loan bal- gage, your first-year payments will look like this:
ance by only $1,015.79. You still owe the bank
$98,984.21.
Date Payment Interest Principal Balance After
01/01/95 898.83 583.33 315.50 99,684.50

02/01/95 898.83 581.49 317.34 99,367.16
03/01/95 898.83 579.64 319.19 99,047.97
04/01/95 898.83 577.78 321.05 98,726.92
05/01/95 898.83 575.91 322.92 98,404.00
06/01/95 898.83 574.02 324.81 98,079.19
07/01/95 898.83 572.13 326.70 97,752.49
08/01/95 898.83 570.22 328.61 97,423.88
09/01/95 898.83 568.31 330.52 97,093.36
10/01/95 898.83 566.38 332.45 96,760.91
11/01/95 898.83 564.44 334.39 96,426.52
12/01/95 898.83 562.49 336.34 $96,090.18
Total for 1995 $10,785.96 $6,876.14 $3,909.82
With a 15-year payment schedule, a much in the second year are even lower, allowing you
greater portion of each payment made goes to¬ to reduce your loan balance with each addi¬
ward reducing the balance of the loan, thereby tional payment even further compared to the
reducing the future interest you pay on the out¬ 30-year payment plan. The same is true for
standing balance. The net effect of this after every year through the end of year 15, at which
one year is that you have paid off $3,909.82 of point your balance is zero. The end result is that
the loan. Since your balance at the end of the by paying only $233.53 more each month, you
first year is $96,090.18, your interest payments pay off the $100,000 loan in half as much time.
the 1 in parentheses reflects the fact that you get your principal back, and
the .05 is the interest rate on the principal. Thus, we see that the future value
of $1,000 in one year when the interest rate is 5 percent is $1,050.
What is the future value of $1,000 in two years when the interest rate is
5 percent? As we just saw, at the end of the first year you will have
(1 + .05) X $1,000 = $1,050.
This amount will earn interest of 5 percent for an additional year. Thus, at
the end of two years you will have
(1 + .05) X $1,050 = $1,102.50.
Notice that interest income earned during the second year is more than the
amount earned during the first year. Why? During the second year, you earn
interest of 5 percent not only on the original $1,000 but also on the $50 in
interest you received during the first year. In the vernacular of financial
markets, compounding of interest has occurred. In other words, it is interest
earned in year 2 on your interest earnings from year 1. We can also express
the future value of this investment with the formula
(1 + .05) X (1 + .05) X $1,000 - $1,102.50,
or, more simply,
(1 + .05)2 X $1,000 = $1,102.50.
The fact that (1 + .05) is raised to the power of 2 reflects the fact that the
original amount is invested for two years.
This basic principle for calculating the future value of an investment
leads to the following general formula. If the interest rate is i and the present
amount to be invested is PV, the future value (FV) of that amount in T
years is
FV = PV X (1 + if
Summary Suppose you invest $2,000 today at an interest rate of 10 percent. How much
Exercise 3.2 will you have in 10 years?
To calculate the future value, set PV = $2,000, T = 10, and

i = .1 in the future value formula:
FV = PV X (1 + if = $2,000 X (l.l)10 = $2,000 X 2.5937423 =

$5,187.48.
Thus, in 10 years your $2,000 will have grown into $5,187.48.
Present Value Analysis. Future value analysis tells us how much a

given sum invested at a specified interest rate will be worth at a future time.
For example, future value analysis allows us to calculate how much money
you would have at age 65 if you invested $1,000 today at a 5 percent interest
rate. We might, however, be interested in knowing what the promise of a
given sum of money in the future is worth to us today; that is, what is the
value today of $100,000 when you turn 65? The ability to answer such a
question is of immediate importance in pricing bonds, which entail the
borrower’s promise to pay a certain sum in the future. We have seen, for
instance, that a zero coupon bond promises to pay the face value of the bond
at a future date and has no explicit interest payments. How do you value
such a bond? What is such a promise worth to you today? To answer these
questions, we use present value analysis.
Simply put, the present value is the value today of funds available at
some future date. Thus, $1,000 available today has a present value of $1,000.
Interest Ra tes 79
What is the present value of an amount received not today but at some date
in the future? Finding the answer is easy. All we have to do is rewrite our
equation for future value so that PV is on the left-hand side. Recall that the
future value formula is
FV — PV x (1 + if
If we solve this equation for PV, we obtain the formula for the present
value (PV) of an amount received T years in the future when the interest
rate is i:
FV
PV
(1 + if
To see how to use this formula, let us calculate the present value of
$1,050 available one year from today when the interest rate is 5 percent.
According to the formula (here FV = $1,050, T = 1, and i = .05), the
present value is
FV _ $1,050
PV $1,000.
(1 + if ~ 1.05
The answer should not surprise you. In the previous section, we saw that the
future value of $1,000 invested for one year at 5 percent is $1,050. Thus, if
the interest rate is 5 percent, you would need to invest $ 1,000 today to have
$1,050 a year from now. In effect, present value is future value calculated
in reverse.
Notice two things about the present value formula. First, as long as the
interest rate is positive, $1 in the future is worth less than $1 today. This
illustrates the time value of money: By waiting to receive money in the
future, you forgo the use of the money (the interest you could have earned
if you received the money today, for instance). Second, the present value of
a given sum received in the future is inversely related to the interest rate.
As the interest rate rises, the present value falls; conversely, the lower the
interest rate, the greater the present value of a given future amount.
Summary Recall that T-bills are zero coupon bonds issued by the federal government
Exercise 3.3 that are sold at a discount. Suppose a T-bill with a maturity of one year and
a face value of $10,000 is offered for sale at a Treasury bill auction. The
interest rate you could earn on an alternative one-year “safe” asset, such as
a certificate of deposit, is 8 percent. What would be a reasonable price for
this T-bill?
Answer: To answer this question, consider what you are purchasing. You
are trading a purchase price (the present value, not yet determined) for a
promise that the U.S. government will pay you $10,000 in one year. This
promise is almost risk free, since it is extremely unlikely that the government
will be unable to pay you this sum in one year. (For instance, as a last resort
the government could always simply print paper money to pay you.) Since
the interest rate on an alternative investment is 8 percent, the maximum price
you should be willing to pay for the T-bill is the amount that, after earning
8 percent for one year, will equal $10,000. (This is exactly the definition of
present value!) Thus, the highest price you will be willing to pay for the
T-bill is the present value of $10,000 received one year in the future when
the interest rate is 8 percent. Algebraically, the present value (PV) of the
T-bill is
$10,000
PV $9,259.26,
1.08
which is the highest price you would bid for the T-bill. If you bid lower and
succeeded, your return on the T-bill would exceed 8 percent.
How Inflation and Taxes Affect Interest Rates
In the first part of this section, we look in more detail at how inflation affects
interest rates and how to ensure that the interest rate you receive on an
investment is high enough to protect your purchasing power from inflation.
In the second part, we examine how to determine the aftertax interest rate
received on investments and how current tax policy also affects the interest
rate paid by borrowers who use funds to make purchases such as homes.
The Real Interest Rate

Suppose you make a $100, one-year loan to someone at a 5 percent interest
rate. One year from now, you will have $105; this is the future value of the
$100 loan. Before you conclude you will be able to purchase more one year
from now than today due to the extra $5 you earned as interest, you had
better think again. When you receive the future amount ($105), the prices
you see in stores will be not today’s prices but the prices one year from
today. If the inflation rate is 7 percent, prices in one year will be 7 percent
higher than they are today. The extra 5 percent you earned in interest will
be more than offset by the 7 percent increase in prices.
To see why this is so, suppose that today you can buy a watch for $100.
If the inflation rate is 7 percent, the same watch will cost $107 in one year.
If you do not make the $100 loan, you can use the $100 to buy the watch
at its current price of $100. But if you loan out the money at 5 percent
How Inflation and Taxes Affect Interest Rates 81
interest, the $105 you will have in one year will be $2 short of what you
need to buy the watch. In real terms, you will lose $2 in purchasing power
(2 percent of the initial amount loaned) if you make the loan.
This example reveals that the presence of inflation can affect the real
value of the money lenders receive in the future. A prudent investor should
consider the real interest rate when making a loan. The nominal interest rate
is the interest rate stated in the loan contract. The real interest rate is the
nominal interest rate minus the expected rate of inflation. More formally, if
i is the nominal interest rate and is the expected rate of inflation, the real
interest rate, r, is defined as3
r — i — TTe.
In the preceding example, the inflation rate is 7 percent and the nominal
interest rate is 5 percent. According to this formula, the real interest rate is
-2% = 5% - 7%.
The fact that the real interest rate is negative means that by loaning out
money at 5 percent when prices are expected to rise by 7 percent, you will
actually lose 2 percent of your purchasing power. This is precisely what you
lost in the preceding example (the $2 loss in puchasing power as a fraction
of the original $100 loan is 2 percent).
Of course, investors are not very likely to loan out money if they think
they will earn a negative real interest rate. To induce lenders to give up the
use of funds, borrowers must compensate them such that they receive a
greater real value in the future. Notice that we can rewrite the equation for
the real interest rate to see exactly what the nominal interest rate would have
to be to ensure that the lender earned a given real interest rate:
i — r + TTe.
This is called the Fisher equation, after American economist Irving Fisher.
It says that the nominal or stated interest rate on a loan (/) equals the real
interest rate (r) plus the expected rate of inflation (tf). In effect, lenders will
charge an inflation premium—an amount over and above the real interest
rate to compensate for the decline in purchasing power due to inflation. This
inflation premium will equal the expected inflation rate during the period of
the loan.
For instance, suppose a lender expects inflation to be 4 percent and
wants to earn a 3 percent real rate of interest; that is, she wants the amount
3 Technically, the real interest rate is
i —
r = -.
1 + TT<J
The formulas in the text are widely used approximations.

paid back to have 3 percent more purchasing power than the money loaned
out. In this case, the interest rate the lender must charge on the loan is
7% = 3% + 4%.
By charging 4 percent above her desired real interest rate, the lender obtains
additional funds that offset her expectation of inflation. Notice that the higher
the lender’s expected inflation rate, the higher the nominal interest rate must
be for her to receive a given real interest rate. Remember, however, that the
lender does not know what the inflation rate will be. She has to form an
educated guess—an expectation. As inflationary expectations change over
time, nominal interest rates will also change.
One final note about real interest rates is that calculations and discussions
of the real interest rate should make it clear whether the real interest rate
being considered is an ex ante real interest rate or an ex post real interest
rate. The ex ante real interest rate is the real interest rate we have been
discussing: the nominal interest rate minus the expected inflation rate. It is
called ex ante to signify that it is “before the fact,” meaning it is calculated
from expected inflation rates and hence is itself an expectation of the real
return from a given nominal interest rate. The ex post real interest rate is
the nominal interest rate minus the actual inflation rate. Ex post means “after
the fact,” and the ex post real interest rate is calculated using actual inflation
rates, information that would be unavailable to an investor at the time he or
she would make an investment decision.
Sometimes ex post real interest rates are used in discussions of the real
interest rate, because the data on actual inflation rates are easier to get than
data on expected inflation rates. For example, Figure 3.5 graphs U.S. inflation
rates, nominal interest rates, and ex post real interest rates for the period
1960-1992. Note the link between nominal interest rates and the inflation
rate, as well as the unusually high real interest rates of the early and middle
1980s. Notice too that between 1974 and 1980, the real interest rate was
negative. During those years, lenders actually lost real purchasing power by
loaning out money.
Are real interest rates the same in all countries? The answer is no, as
shown in Box 3.4.
Anticipated Inflation. Anticipated inflation is inflation that partic¬

ipants in financial markets expect and thus take into account when making
their plans. If a lender expects the inflation rate to be 4 percent and it turns
out to be exactly 4 percent during the term of the loan, the inflation premium
charged on the loan exactly offsets the inflation. In real terms, the lender
receives a real return exactly equal to what led him to lend the money in the
first place. The borrower pays back the loan with dollars that are worth
exactly what she expected them to be worth.
How Inflation and Taxes Affect Interest Rates S3
1.35, various issues, U.S. Department of Labor, Bureau of Labor Statistics, The Consumer Price
Index, various issues, and Citibase electronic database.
Unanticipated inflation. Unanticipated inflation, on the other

hand, is inflation over and above what participants in financial markets
expected at the time funds were issued. Unanticpated inflation has a different
impact on borrowers and lenders, as described next.
Suppose a lender expects the inflation rate to be 4 percent and thus adds
an inflation premium of 4 percent to the real interest rate. And, suppose the
actual inflation rate turns out to be 10 percent instead of the anticipated 4
percent. The lender expected prices to rise by 4 percent and charged an
International Banking V
Real Interest Rates Around the World
How do real interest rates compare among 5.9 percent rates in both Germany and the
major Western industrialized nations? The an¬ United Kingdom during that year. Second, no¬
swer depends on the point in time at which the tice how much real interest rates changed
comparisons are made. This point is illustrated in within each country between 1981 and 1991. In
the accompanying diagram, which compares Canada, for instance, the real interest rate was
real interest rates in the United States, Japan, 7.08 percent in 1981 but fell to 3.5 percent by
Germany, the United Kingdom, and Canada in 1991. Thus, real interest rates vary not only
1981 and 1991. across countries but over time as well.
This diagram reveals several interesting pat¬
terns. First, notice that real interest rates varied
considerably across countries. In 1991, for in¬ Source: OCED, various issues, U.S. Department of Com¬
stance, the real interest rate in the United States merce, Bureau of Economic Analysis, Survey of Current
was 1.33 percent, considerably lower than the Business, various issues, and Citibase electronic database.
inflation premium to cover these expectations. Unfortunately for the lender,

prices increased by 6 percent more than he anticipated, and he thus ends up
with 6 percent less in purchasing power than he expected when he made the
loan.
Is unanticipated inflation “bad”? It depends on whether you are a bor¬
rower or a lender. Clearly unanticipated inflation hurts lenders, since they
are paid back with money that is less valuable than anticipated. Borrowers,
on the other hand, are better off. They get to pay back the loan with money
that is less valuable than they thought it would be.
How Inflation and Taxes Affect Interest Rates 85
Summary In 1991, the prime interest rate was 8.46 percent. Between December 31,
Exercise 3.4 1990, and December 31, 1991, the price level increased from 112.87 to
116.98. What was the real prime interest rate in 1991?
Answer: The inflation rate in 1991 was
P1991 P 1990 116.98 - 112.87

TT]99i — = .0364,
1990 112.87
or 3.64 percent. Assuming this inflation was anticipated, the real prime
interest rate in 1991 was r = i — ire = 8.46% — 3.64% = 4.82%.
The Aftertax Interest Rate

In the United States, interest income is subject to personal income tax. This
means lenders must pay a fraction of interest earnings to the government in
the form of income taxes. (We noted in the previous chapter that some
exceptions exist. One exception is interest income from municipal bonds,
which is exempt from federal income taxation.) Similarly, some forms of
interest payments—mortgage interest payments, for instance—are tax de¬
ductible. Interest payments that are tax deductible reduce the taxes that a
borrower would otherwise pay.
How do taxes affect the “effective” aftertax interest rate borrowers pay
and lenders receive? Suppose your marginal tax rate is 30 percent; that is,
you are in the 30 percent tax bracket and must pay 30 percent of each
additional dollar you earn to the government as income tax. What will you
earn if you make a $100, one-year loan at an interest rate of 10 percent?
(For now, we keep things simple and ignore inflation.)
One year from today, you will receive $110 from the borrower. Of this
amount, $100 is principal and $10 is interest income subject to the income
tax at the 30 percent rate. You pay the government 30 percent of the $10,
which is $3, and keep the remaining $7. In effect, you get to keep only the
fraction (1 — .3) = .7, or 70 percent, of the interest income; the other 30
percent goes to the government. After taxes, you end up with $7 in interest
on the $100 loan. In other words, your aftertax rate of interest is 7 percent,
which is (1 — .3) X 10%.
We can state this finding much more generally. Suppose the marginal
tax rate is t and the pretax interest rate is i. Then the aftertax interest rate
the lender earns is
ir = (1 — t) X i.
Summary Your client is in the 38 percent tax bracket and thus pays $38 of every
Exercise 3.5 additional $100 she earns to the federal government. She has $10,000 to
invest for one year and wants your advice in choosing between a corporate
bond paying 9 percent interest and a municipal bond paying 6 percent

interest. Interest on the municipal bond is exempt from taxation, whereas
that on the corporate bond is not. Which bond should she purchase?
I Answer: The aftertax interest rate is
ij = (1 - t) X i,
so the corporate bond pays an aftertax interest rate of (1 — .38) X (.09) =

.0558, or 5.58%. The municipal bond pays 6 percent before and after taxes.
Therefore, your client should buy the municipal bond, since it yields a higher
aftertax return.
Historical Data on Interest Rates
To acquaint you with past trends in interest rates, Table 3.1 provides histor¬
ical data on various interest rates, including short-term Treasury securities
(three-month and six-month Treasury bills), longer-term government bonds
(three-year and ten-year bonds, converted to constant maturities), corporate
bonds (both Aaa and Baa, where Baa bonds are riskier), municipal bonds,
home mortgage rates, commercial paper and prime rates, the Federal Reserve
discount rate, and the federal funds rate. These rates can be used to compare
assets of different maturities and different assets with the same maturity, as
well as to get a feel for historical movements in U.S. interest rates. As you
look at the interest rates in Table 3.1, you should note two things (in the
next chapter, we discuss these observations in more detail).
First, just as a supermarket has many prices, financial markets have
many interest rates. We will look at the reason for this in the next chapter,
but the basic reasons are simple to understand. First, financial instruments
issued by different borrowers have different default risks. Riskier borrowers
must offer higher interest rates to compensate lenders for the additional risk
of default. Second, debt instruments differ with respect to their underlying
liquidity. More liquid instruments tend to have lower interest rates than do
less liquid instruments. Again, the higher interest rates paid on less liquid
financial assets compensate lenders for the fact that they cannot as easily
liquidate the assets prior to maturity. This makes less liquid assets attractive
to investors seeking higher returns. Finally, different financial instruments
have different terms to maturity; the interest rate on a 1-year loan generally
does not equal the interest rate on a 30-year loan. Chapter 10 is entirely
devoted to explaining this phenomenon, which is known as the term structure
of interest rates.
The second thing to note about interest rates is that they vary over time.
This is because inflation premiums, tax rates, the risk of default, and market
conditions in debt markets all change over time, and changes in each of
Historical Data on Interest Rates 87
Table 3.1
Interest Rates on Various Financial Instruments, 1963-1992
U.S. Treasury Securities

Corporate High-
Bonds Grade Discount
New- Commer¬ Prime
(Moody’s) Munici- Rate, Fed¬
Bills Constant Home cial Rate
pal Federal eral
Year (new issues) Maturities Mort¬ Paper, Charged
Bonds Reserve Funds
gage 6 by
Aaa Baa (Stand¬ Yields Months Banks
Bank of Rate
3- 10- ard & New York
3-Month 6-Month
Year Year Poor’s)
1963 3.157 3.253 3.67 4.00 4.26 4.86 3.23 5.89 3.55 4.50 3.23 3.18
1964 3.549 3.686 4.03 4.19 4.40 4.83 3.22 5.83 3.97 4.50 3.55 3.50
1965 3.954 4.055 4.22 4.28 4.49 4.87 3.27 5.81 4.38 4.54 4.04 4.07
1966 4.881 5.082 5.23 4.92 5.13 5.67 3.82 6.25 5.55 5.63 4.50 5.11
1967 4.321 4.630 5.03 5.07 5.51 6.23 3.98 6.46 5.10 5.61 4.19 4.22
1968 5.339 5.470 5.68 5.65 6.18 6.94 4.51 6.97 5.90 6.30 5.16 5.66
1969 6.677 6.853 7.02 6.67 7.03 7.81 5.81 7.81 7.83 7.96 5.87 8.20
1970 6.458 6.562 7.29 7.35 8.04 9.11 6.51 8.45 7.71 7.91 5.95 7.18
1971 4.348 4.511 5.65 6.16 7.39 8.56 5.70 7.74 5.11 5.72 4.88 4.66
1972 4.071 4.466 5.72 6.21 7.21 8.16 5.27 7.60 4.73 5.25 4.50 4.43
1973 7.041 7.178 6.95 6.84 7.44 8.24 5.18 7.96 8.15 8.03 6.44 8.73
1974 7.886 7.926 7.82 7.56 8.57 9.50 6.09 8.92 9.84 10.81 7.83 10.50
1975 5.838 6.122 7.49 7.99 8.83 10.61 6.89 9.00 6.32 7.86 6.25 5.82
1976 4.989 5.266 6.77 7.61 8.43 9.75 6.49 9.00 5.34 6.84 5.50 5.04
1977 5.265 5.510 6.69 7.42 8.02 8.97 5.56 9.02 5.61 6.83 5.46 5.54
1978 7.221 7.572 8.29 8.41 8.73 9.49 5.90 9.56 7.99 9.06 7.46 7.93
1979 10.041 10.017 9.71 9.44 9.63 10.69 6.39 10.78 10.91 12.67 10.28 11.19
1980 11.506 11.374 11.55 11.46 11.94 13.67 8.51 12.66 12.29 15.27 11.77 13.36
1981 14.029 13.776 14.44 13.91 14.17 16.04 11.23 14.70 14.76 18.87 13.42 16.38
1982 10.686 11.084 12.92 13.00 13.79 16.11 11.57 15.14 11.89 14.86 11.02 12.26
1983 8.63 8.75 10.45 11.10 12.04 13.55 9.47 12.57 8.89 10.79 8.50 9.09
1984 9.58 9.80 11.89 12.44 12.71 14.19 10.15 12.38 10.16 12.04 8.80 10.23
1985 7.48 7.66 9.64 10.62 11.37 12.72 9.18 11.55 8.01 9.93 7.69 8.10
1986 5.98 6.03 7.06 7.68 9.02 10.39 7.38 10.17 6.39 8.33 6.33 6.81
1987 5.82 6.05 7.68 8.39 9.38 10.58 7.73 9.31 6.85 8.21 5.66 6.66
1988 6.69 6.92 8.26 8.85 9.71 10.83 7.76 9.19 7.68 9.32 6.20 7.57
1989 8.12 8.04 8.55 8.49 9.26 10.18 7.24 10.13 8.80 10.87 6.93 9.21
1990 7.51 7.47 8.26 8.55 9.32 10.36 7.25 10.05 7.95 10.01 6.98 8.10
1991 5.42 5.49 6.82 7.86 8.77 9.80 6.89 9.32 5.85 8.46 5.45 5.69
1992 3.45 3.57 5.30 7.01 8.14 8.98 6.41 — 3.80 6.25 3.25 3.52
Source: Economic Report of the President (Washington, D.C.: U.S. Government Printing Office, 1993), p. 428.
these factors affect interest rates. We briefly touched on the reasons inflation
and taxes affect interest rates in this chapter. In the next chapter, we will
take a closer look at these issues.
Conclusion
In this chapter, we focused on inflation and interest rates. We learned that

inflation is growth in the overall price level and is caused by changes in the
money stock and velocity that exceed changes in real output. We also saw
how to use publicly available data to calculate such variables as nominal
output, real output, and the rate of inflation.
Our study of interest rates provided examples of how interest is paid on
different types of financial instruments. We also examined the basics of
determining the present and future value of loans. We will use the basic
tools of present value analysis extensively in Chapter 8 to see how a host of
financial instruments are priced.
Finally, we saw how taxes and inflation affect the real purchasing power
of lenders and borrowers. We also looked at historical data on interest rates
for various financial assets; the data illustrate that interest rates vary consid¬
erably over time and across different financial instruments.
The material presented in this chapter will serve as our starting point in
the next chapter, where we develop economic models we can use to under¬
stand and predict changes in interest rates over time.
KEY TERMS
price level future value (FV)

nominal output present value (PV)
real output real interest rate
inflation rate inflation premium
economic growth nominal interest rate
velocity anticipated inflation
equation of exchange unanticipated inflation
inflationary expectations aftertax interest rate
1. If the price of a textbook increases by 7 2. The price level was 26.02 in 1960 and
percent, can you conclude that the infla- 26.25 in 1961. What was the inflation rate
tion rate is 7 percent? Explain. between 1960 and 1961?
3. Ml was $167.9 billion in 1965 and U.S. Treasury bill rate. Which is usually
$172.1 billion in 1966. Assuming velocity higher? Is this always the case? Can you
was constant during this period, how explain why one rate is usually higher
much would real output have had to grow than the other?
for no inflation to have occurred? Explain.
10. “If the expected inflation rate is 5 percent
4. A coupon bond with a face value of and a lender loans out money at a 4 per¬
$1,000 pays $80 at the end of each year cent interest rate, he or she will end up
for five years and then matures. The inter¬ with a 9 percent real return on the invest¬
est rate on similar one-year investments is ment.” Is this statement true or false? Ex¬
8 percent. What is the present value of plain.
this bond?
11. Does inflation “hurt” borrowers or lend¬
5. (a). Determine the present value of ers? Explain carefully.
$10,000 when the interest rate is 5 percent
and the money is to be received in (1) 1 12. Joe issues a loan at an interest rate of
year, (2) 5 years, and (3) 10 years. 9 percent. Determine the “effective”
(b). What do you conclude about the rela¬ interest rate Joe earns if
tionship between when money is received (a) . Joe is in the 28 percent tax bracket
and present value? and expects no inflation.
(b) . Joe does not pay taxes but expects
6. (a). Determine the present value of the inflation rate to be 6 percent.
$10,000 to be received in five years when (c) . Joe is in the 28 percent tax bracket
the interest rate is (1) zero percent, (2) 5 and expects the inflation rate to be 6 per¬
percent, and (3) 10 percent. cent.
(b). What do you conclude about the rela¬
tionship between the interest rate and the 13. Suppose you just learned that the govern¬
present value of a future amount? ment plans to print more money to finance
the federal budget deficit.
7. If the interest rate is 10 percent, how (a) . How would this action affect your es¬
much money would you have to put in the
timate of expected inflation? Why?
bank today to be a millionaire in 30
(b) . How would this action affect the in¬
years?
flation premium you would charge on a
8. “If you put $100 in a bank account that loan? Why?
draws 5 percent interest, in one year you (c) . How would this action affect the
will have $105 and in two years you will nominal interest rate you would charge on
have $110.” Is this statement true or a loan? Why?
false? Explain.
14. Briefly explain why interest rates vary (1)
9. Using Table 3.1, compare the six-month across different types of financial instru¬
commercial paper rate with the six-month ments and (2) over time.
Adams, R. D., and M. Moghaddam. “Searching for Hula, D. G. “The Phillips Curve and the Natural Rate
the Darby Effect in Tax Exempt and Taxable Inter¬ of Inflation.” Policy Sciences, 24 (November
est Rate Data.” Quarterly Journal of Business and 1991), 357-366.
Economics, 30 (Summer 1991), 48-63. Ibrahim, I. B., and R. M. Williams. “The Fisher Rela¬
Allen, S. D. “The Determinants of the Tax-Adjusted tionship under Different Monetary Standards: A
Real Interest Rate.” Journal of Macroeconomics, Note.” Journal of Money, Credit, and Banking, 10
14 (Winter 1992), 15-32. (August 1978), 363-370.
Amsler, C. E. “What Determines Expected Real In¬ Mark, N. C. “Some Evidence on the International In¬
terest Rates? Quarterly Review of Economics and equality of Real Interest Rates.” Journal of Inter¬
Business, 25 (Winter 1985), 59-67. national Money and Finance, 4 (June 1985), 189-
Bach, G. L., and J. B. Stephenson. “Inflation and the 208.
Redistribution of Wealth.” Review of Economics Marquis, M. H., and K. L. Reffett. “Real Interest
and Statistics, 56 (February 1974), 1-13. Rates and Endogenous Growth in a Monetary
Ball, L., and S. G. Cecchetti. “Inflation and Uncer¬ Economy.” Economics Letters, 37 (October 1991),
tainty at Short and Long Horizons.” Brookings pp. 105-109.
Papers on Economic Activity (1990), 215-245. Pelaez, R. F. “Interest Rates as Predictors of Inflation
Biswas, B., and P. J. Saunders. “Money and Price Revisited.” Southern Economic Journal, 55 (April
Level in India: An Empirical Analysis.” Indian 1989), 1025-1028.
Economic Journal, 38 (July-September 1990), Ratti, R. A. “Sectoral Employment Variability and
103-114. Unexpected Inflation.” Review of Economics and
Brunner, L. P., H. Beladi, and H. A. Zuberi. “Infla¬ Statistics, 61 (May 1985), 278-283.
tion and Indexation: An Empirical Approach.” Ring, R. J., Jr. “Variability of Inflation and Income
Southern Economic Journal, 52 (July 1985), 250- across Income Classes.” Social Science Quarterly,
264. 66 (March 1985), 203-209.
Cecchetti, Stephen G. “The Case of the Negative Swofford, J. L., and G. A. Whitney. “The Composi¬
Nominal Interest Rates: New Estimates of the tion and Construction of Monetary Aggregates.”
Term Structure of Interest Rates During the Great Economic Inquiry, 29 (October 1991), 752-761.
Depression.” Journal of Political Economy, 96 Thistle, P. D., R. W. McLeod, and B. L. Conrad. “In¬
(December 1988), 1111-1141. terest Rates and Bank Portfolio Adjustments.”
Gerdes, W. D. “Mr. Fisher and the Classics.” Ameri¬ Journal of Banking and Finance, 13 (March 1989),
can Economist, 30 (Spring 1986), 66-72. 151-161.
Graham, F. C. “The Fisher Hypothesis: A Critique of Varghese, K. T. “Wage Indexation, Inflation, and Un¬
Recent Results and Some New Evidence.” South¬ employment.” Journal of Economics and Busi¬
ern Economic Journal, 54 (April 1988), 961-968. ness, 32 (Fall 1979), pp. 51-55.
Grennes, T., and J. S. Lapp. “Neutrality of Inflation Wolf, H. A., and R. W. McEnally. “The Unemploy¬
in the Agricultural Sector.” Journal of Interna¬ ment-Inflation Trade-off, 1948-1968: A Multiple
tional Money and Finance, 5 (June 1986), 231- Phillips Curve Analysis.” Social Science Quar¬
243. terly, 51 (September 1970), 275-284.
PART TWO
The Microeconomics of Banking

and Financial Markets
4
Supply and Demand in Financial Markets
5
The Bank as a Firm: Loans
6
The Bank as a Firm: Deposits
7 I
The Banking Industry '
CHAPTER
Supply and Demand in

Financial Markets
ast week the Wall Street Journal reported that mortgage

interest rates increased by .25 percent. What economic
factors would have led to such a change? What impact will this change have
on your ability to borrow funds for a house or a new car? This chapter
presents the basic economic tool needed to answer these questions: supply
and demand analysis.
Recall from your principles course that economists use supply and de¬
mand to analyze how market forces determine the price and quantity of a
good or service. In the first part of this chapter, we review these basic tools
and see how they can be used to explain how market forces determine the
prices of ATM transactions, insurance policies, and other fee-based financial
services. In the second part, we develop a supply and demand apparatus that
is very useful for analyzing the market for loanable funds—the market that
determines the interest rate charged for such loans as mortgages and car
loans.
Basic Supply and Demand Analysis: A Brief Review with

Applications to Fee-Based Financial Services
If you took an introductory economics course, you probably learned how
the supply and demand for products like gasoline ultimately determine the
price you pay at the pump. Many financial services are priced like any other
commodity. Just as the price of gasoline is quoted in terms of the amount
you must pay for each gallon, the prices of fee-based financial services,
such as withdrawals from an automatic teller machine (ATM), are quoted in
terms of the amount you must pay for each withdrawal. Similarly, the price
of an automobile insurance policy is quoted in terms of the amount you must
pay to insure a given type of car. Accountants and bookkeepers price their
services based on the amount you must pay for each hour of their services.
Other examples of fee-based financial services are wire transfers, traveler’s
checks, credit reports, and the like.
Since prices of a fee-based financial service are quoted in terms of dollars
per unit of the service purchased, we can use basic supply and demand
92
Basic Supply and Demand Analysis 93
analysis to see how the prices for these services are determined. Accordingly,
the remainder of this section provides a brief review of basic principles of
supply and demand analysis and uses them to analyze markets for fee-based
financial services. We will use supply and demand analysis extensively
throughout this book, so make sure you master this material before reading
on.
Demand for Fee-Based Financial Services

Suppose you must determine whether or not to use an ATM to obtain cash
from your checking account. Your decision will depend on a number of
things, including the price charged for each withdrawal, your income, and
the price of alternatives such as writing a check or withdrawing cash directly
from the bank.1
The demand curve in Figure 4.1 shows the relationship between the
price of ATM transactions and the quantity demanded of ATM transactions,
holding other things constant. By the law of demand, an increase in the
price of a fee-based financial service results in a decline in its quantity
demanded. In Figure 4.1, we see that when the price of an ATM transaction
is $1, 15 million in transactions are made each month. When the price rises
to $2, consumers reduce the quantity of monthly ATM transactions to 10
million. The movement along a demand curve, such as the one from A to B
in Figure 4.1, denotes a change in quantity demanded. In other words, a
change in the price of a fee-based financial service leads to a change in the
quantity demanded for that service.
Remember that the position of the demand curve in Figure 4.1 assumes
that other things, such as income, prices of related goods, and consumer
tastes, are held constant. If these things changed, the demand curve would
shift to the right or the left depending on the nature of the change. A shift
in the entire demand curve, such as the shift from D] to D2 or D3 in Figure
4.2, denotes a change in demand for fee-based services.
Determinants of Demand for Fee-Based Financial Serv¬

ices. Economists call variables that affect the position of the demand curve
determinants of demand. Changes in one or more of these determinants—
including a change in the price of complementary or substitute goods, in¬
come, and tastes—shift the demand curve.
1 This example has several complications. Banks sometimes offer their customers the “free” use
of ATM machines located on their premises. Instead of paying a fee for each transaction, customers
are required to keep a certain minimum balance in their accounts at the bank. Banks gain in two
ways from this arrangement. First, they can hire fewer tellers, thus reducing labor costs. Second,
they can earn interest by lending out the minimum balance customers are required to hold. Note,
however, that even in these cases customers usually pay a fee for using ATM machines not located
at their own banks. Our analysis of ATM transactions can be thought of as using ATM machines
located away from the customer’s bank.
94 Chapter4 Supply and Demand in Financial Markets
Prices of Substitutes. One determinant of the demand for anv good »
or service, including fee-based financial services, is the prices of substitutes.

Goods or services are substitutes if, when the price of one increases, the
demand for the other increases. For instance, withdrawing cash from an
ATM is a substitute for writing a check. If your bank substantially increased
the fee it charges you for each check you write, you would likely increase
your use of ATMs. Figure 4.2 illustrates this increase in demand for ATMs
by the shift from D1 to D\ which reflects the fact that consumers now make
more ATM transactions than before at each price of an ATM withdraw al. A
decrease in the price of a substitute good or service shifts the demand curve
in the opposite direction, from D1 to D:.
Prices of Complements. Another determinant of the demand for a

financial service is the prices of complements. Complements are goods or
services that tend to be used jointly. An easy way to see why the prices of
complements affect the position of the demand curve for a given fee-based
financial service is to consider car insurance, which is an example of such a
service.2 For a fee—the insurance premium—the insurance company agrees
2 We analyze insurance more explicitly in Chapters 6. 7. and c), where we look at insurance as a
way to reduce risk.
Basic Supply and D e ma v d A \ a l rs ; 95
to accept some of the financial responsibility for any accidents or other

damages incurred b;. you or others. Car insurance and cars are complements:
•Then the price or car insurance rises, fewer individuals will bur car insur¬
ance c^e to the law of demand and in turn will desire fewer cars ("since
insurance is m~ah;. a prerequisite for purchasing—or at least driving—a
car Indeed, mar.;, teenagers find the;, can more easih afford a car than they
car. afford the insurance or. that car.
Since cars and car assurance are complements, an increase in the prices
of cars reduces the demand for car insurance. This would be depicted as a
shift to the left ir. the demand for car insurance. Similar!;.. a decrease in the
prices of cars w oeld lead to an increase in the demand for car insurance—a
shift to the right in the demand for car insurance.
Income. Income also influences the demand for fee-based financial serv¬
ices. Normal];, increases in income lead to increases in the demand for fee-
based financial services, since the extra income affords individuals a greater
opportunity to purchase these services. A decrease in income would lead to
a decrease :r the demand for financial services—a shift to the left in the
demand cur. e.
Tastes. Final A. a change in consumer tastes can change the position of

the demand cur e. For instance, if a change in tastes leads consumers to
Figure 4.2
Changes in Demand
increase their use of cash and shun credit cards, the demand for ATM
transactions will shift to the right. Conversely, if a taste change causes
consumers to decrease their use of cash, the demand for ATM transactions
will shift to the left.
Summary Homeowner’s insurance is a fee-based financial service. Illustrate graphically

Exercise 4.1 the impact of the following on the demand for homeowner’s insurance: (a)
a reduction in the price of homeowner’s insurance, (b) a reduction in income,
and (c) a reduction in the price of new houses.
Answer: (a) By the law of demand, a reduction in the price of home-

owner’ s insurance leads to an increase in the quantity demanded of home-
owner’s insurance. This corresponds to the movement from A to B in part
a. (b) A reduction in income would decrease the demand for homeowner’s
insurance. This corresponds to the movement from D1 to D2 in part b. (c)
Since new houses and homeowner’s insurance are complements, a reduction
in the price of new houses would increase the demand for homeowner’s
insurance. This corresponds to the shift from D1 to D2 in part c.
Price per Price per Price per

Insurance Insurance Insurance
Policy Policy Policy
Insurance Insurance Insurance

(a) (b) (c)
Supply of Fee-Based Financial Services

The supply curve for a fee-based financial service shows the quantity of
financial services suppliers are willing and able to provide at alternative
prices, holding everything else constant. Figure 4.3 depicts typical supply
curves for ATM transactions. The movement along a supply curve, such as
the one from A to B along S1, is called a change in quantity supplied. The
fact that the supply curve for fee-based financial services slopes upward
simply reflects the law of supply: Providers of financial services provide
more services when the price is high than when it is low.
A change in a variable other than the price of the financial service, such
as input prices or a change in technolgy, leads to a change in supply. This
corresponds to a shift in the entire supply curve, such as the shift in Figure
4.3 from Sl to either S2 or S3.
Determinants of the Supply of Fee-Based Financial Serv¬

ices. The determinants of supply are those factors that affect the position
of the supply curve. They include input prices and the level of technology.
Input Prices. A decrease in the price of an input (a decline in the price

of electricity, for instance) shifts the supply curve to the right because
suppliers of fee-based financial services are willing and able to provide more
output at each price due to the lower input price. In Figure 4.3, suppose Sl
is the initial supply curve for ATM transactions. If the price of electricity
falls, suppliers are willing to provide more ATM transactions at each price,
so the supply curve shifts to the right to S3. Conversely, an increase in the
The law of supply Figure 4.3

states that as the price Supply of a Fee-Based Financial Service
of a fee-based finan¬
cial service rises, the
quantity supplied of
the service increases.
This is represented by
the movement from A
to B along supply
curve 51. A change in
a determinant of sup¬
ply leads to a change
in the position of the
supply curve. An in¬
crease in supply is rep¬
resented by a shift in
the supply curve to
the right from 51 to
53. A decrease in sup¬
ply is reflected by a
shift in the supply
curve to the left from
51 to 52.
98 Chapter 4 Supply and Demand in Financial Markets
price of an input (for example, an increase in the price of labor) shifts the
supply curve to the left, such as the shift from S1 to S2 in Figure 4.3.
Technology. Changes in technology also affect the position of the sup¬

ply curve. Technological advances shift the supply curve to the right, since
more of the services will be supplied at each price. This is reflected in Figure
4.3 as the increase in supply from S1 to S3. Indeed, improvements in computer
technology have substantially increased the supply of ATMs, and banks
provide more ATM transactions at each price today than they did five years
ago.
Summary Many banks charge a fee for wire transfer services—the transfer of funds
Exercise 4.2 from one financial institution to another via telephone lines or other elec¬
tronic media, including wireless communications. Illustrate graphically the
impact on the supply of wire transfer services of (a) the innovation of
satellites that replace telephone wires and (b) an increase in the wage paid
to wire transfer operators.
Answer: Wire transfers are a fee-based financial service, since a charge

is levied for each transfer, (a) The innovation of satellites is an improvement
in technology, which leads to an increase in supply from S to in part a.
(b) An increase in the wage paid to wire transfer operators is an increase in
the price of an input and therefore decreases supply from S to S2 in part b.
Price per Wire Price per Wire

Transfer Transfer
(a) (b)
Equilibrium in the Market for Fee-Based Financial Services

Interactions among all buyers and sellers in the market determine the prices
of fee-based financial services. More precisely, the interaction of supply and
demand for a service determines its price.
Figure 4.4 shows hypothetical supply and demand curves for ATM
transactions. To see how the market price is determined, suppose the price
of ATM withdrawals is $1 per transaction. This price corresponds to point
B on the demand curve, where consumers wish to make 15 million trans¬
actions per month. Similarly, the price of $1 corresponds to point A on the
supply curve, so banks provide only 5 million ATM transactions per month
at this price. When the price is $1, quantity demanded exceeds quantity
supplied, creating a shortage. Shortages are associated with long lines; thus,
when the price is $1, consumers experience higher transactions costs due to
the inconvenience of having to wait in line to use an ATM.
Shortages put pressure on the price to rise. As the price rises from $1 to
$2 in Figure 4.4, banks increase their quantity supplied of ATM services to
accommodate 10 million transactions per month at this higher price. Simi¬
larly, as the price rises, consumers use ATMs less frequently. When the
price rises to $2, the quantity demanded is 10 million transactions per month.
At this price, banks provide just enough ATMs to accommodate all consum¬
ers willing and able to use them at that price; quantity demanded equals
quantity supplied. The market is in equilibrium, meaning there is no ten¬
dency for the price to change any further.
Why wouldn’t the price rise above $2 to, say, $3? At a price above the
equilibrium level, there would be a surplus of ATM transactions because
100 c hafter 4 Supply and Demand in Financial Markets
banks’ quantity supplied of ATM transactions would exceed the quantity

demanded by consumers. This underutilization of ATM machines would
ultimately put pressure on banks to lower the price of ATM withdrawals.
When the price falls to $2, the quantity demanded and the quantity supplied
are both 10 million, and the market is in equilibrium.
Thus, the interaction of supply and demand ultimately determines a
market price for the fee-based financial service—in this case, $2—such that
neither a shortage nor a surplus of the good exists. This price is called the
equilibrium price, and the corresponding quantity (10 million) is called the
equilibrium quantity for the market. Once this price and quantity are realized,
the market forces of supply and demand are balanced; there is no tendency
for prices to either rise or fall until something changes the position of the
demand or supply curve.
Changes in Equilibrium: Applications

to the Insurance Market
The final step in our review of supply and demand is to show how changes
in demand or supply can lead to changes in the prices of fee-based financial
services. While the techniques described can be used to analyze markets for
all types of fee-based financial services, we focus on the insurance market
to provide a foundation for the analysis of insurance in Chapters 6, 7, and 9.
Impact of an Increase in the Demand for Insurance. Imagine

that an increase in the U.S. birth rate increases the demand for life insurance.
How will this change affect the price of a typical life insurance policy?
Figure 4.5 shows the initial supply (5°) and demand (D°) curves for life
insurance. The initial equilibrium is at point A, where the equilibrium price
of the average policy is $300 per year. Assuming other things remain un¬
changed, the increase in demand from D° to Dl leads to a new equilibrium
at point B. As a consequence of the increase in the demand for life insurance,
the market price increases from $300 to $350, and the quantity of policies
sold increases from 500 million to 600 million.
The reason for the change in equilibrium is as you would expect. When
demand increases, there are too few insurance companies and agents to
process applications at the old price to satisfy the number of consumers
willing and able to buy policies at that price. This creates excess demand
that puts pressure on the price to rise. As the price rises, insurance companies
increase their quantity supplied (moving along S° from A to B) and consum¬
ers reduce their quantity demanded (moving along D] from C to B) until
ultimately enough policies are offered at the new price of $350 to exactly
equal quantity demanded.
Impact of a Decrease in the Supply of Insurance. What would

happen to the price of life insurance if rising costs in the insurance industry
An increase in the de¬ Figure 4=5

mand for life insur¬ Increased Demand in the Market for Life Insurance
ance from D° to D1
causes the equilibrium
price to rise from
$300 to $350 and the
equilibrium quantity of
life insurance policies
to increase from 500
million to 600 million.
caused the supply to decrease? Point A in Figure 4.6 shows the initial
equilibrium in the insurance market, where demand curve D° intersects the
market supply curve S°. The increase in the cost of providing insurance
decreases the supply of insurance from S0 to S1, resulting in a new market
equilibrium at point B. In this instance, the market price rises from $300 to
$400, and the equilibrium quantity of insurance policies decreases from 500
million to 300 million. At the initial price of $300, consumers wish to buy
more insurance than insurance companies provide at that price. The market
mechanism eliminates the resulting shortage by raising the equilibrium price,
in this case from $300 to $400 per policy.
Impact of a Simultaneous Increase in Demand and Decrease

in the Supply of Insurance. Finally, let us see what would happen
to the price of life insurance if the supply and demand changed at the same
time due to a simultaneous increase in the birth rate and costs in the insurance
industry. In Figure 4.7, the insurance market is initially in equilibrium at
point A, where demand curve D° intersects the market supply curve 5°.
Supply decreases from S° to S] due to the higher costs of providing insurance,
and demand increases from D° to D1 as a result of the higher birth rate. In
102 Chapter 4 Supply and Demand in Financial Markets
A decrease in the sup¬ Figure 4.6

ply of life insurance Decreased Supply in the Market for Life Insurance
from 5° to 51 causes
the equilibrium price
to rise from $300 to Price of
$400 and the equilib- Insurance c1
rium number of poli- Policy ($)
cies to fall from 500 \ / 5°
million to 300 million. \ / /
X / /
\ / /
400
V
300 -
o
O
300 500
Quantity of Insurance Policies (Millions)
this instance, a new equilibrium occurs at point B; the price increases from
P° to P\ and the quantity increases from Q° to Qx.
Given the magnitude of the increase in demand and the decrease in
supply in Figure 4.7, both the price and the quantity of insurance policies
increase. Notice, however, that if the supply curve had shifted much farther
to the left—say, to S—it would have intersected the new demand curve,
D\ at point C instead of B. In this case, the price (P2) would still be higher
than the initial equilibrium price of P°. However, the resulting quantity (Q2)
would be lower than at the initial equilibrium quantity of Q°, since at point
C fewer insurance policies are bought and sold than at point A. Thus, when
demand increases and supply decreases, the market price of fee-based finan¬
cial services rises, but the market quantity may rise or fall depending on the
relative magnitude of the shifts.
This example illustrates a general principle: Simultaneous changes in
the demand and supply of fee-based financial services will lead to some
ambiguity regarding whether price or quantity rises or falls. To determine
the precise impact of simultaneous changes in demand and supply, we must
take care that the conclusions we draw are not due to how far we have shifted
the curves.
Summary Graphically illustrate what would happen in the market for traveler’s checks
Exercise 4.3 if (a) due to a reduction in terrorist activity, families did more traveling; (b)
a bank card was introduced that rebated 1 percent of all charges to the
cardholder; and (c) the cost of issuing traveler’s checks increased.
When demand in¬ Figure 4.7

creases from D° to D1 Simultaneous Increase in Demand and Decrease in the Supply
and supply decreases of Life Insurance
by a small amount
(from S° to 51), the
equilibrium quantity
rises to Q1 and the
price rises to P\ If the
shift in supply is rela¬
tively large (a shift
from 5° to 52), the
falls to Q2 and the
price rises to P2. In
both cases, price rises
but the effect on the
depends on the mag¬
nitude of the decrease
in supply.
Answer: (a) This change in tastes for traveling would increase the demand
for traveler’s checks from D to D] in part a, resulting in an increase in both
the equilibrium price and the equilibrium quantity of traveler’s checks, (b)
Since bank cards are a substitute for traveler’s checks, a reduction in the
price of using a bank card would decrease the demand for traveler’s checks
from D to D2 in part b, reducing the equilibrium price and quantity of
traveler’s checks, (c) An increase in the cost of issuing traveler’s checks
would reduce the supply of traveler’s checks in part c from S to Sl. This
would increase the equilibrium price and reduce the equilibrium quantity of
traveler’s checks.
Quantity of Quantity of Quantity of

Traveler's Checks Traveler's Checks Traveler's Checks
(a) (b) (c)
The Market for Loanable Funds
In contrast to the prices of fee-based financial services, the price you pay
for a mortgage or a car loan is not stated in terms of a dollar amount; rather,
it is quoted as an interest rate. In shopping for a car, you generally wish to
buy from the dealer that asks you to pay the lowest dollar price. When
shopping for a car loan, you seek the lender offering the lowest interest
rate. Financial services with prices expressed in terms of an interest rate are
called loanable funds. You can think of the interest rate as the price of a
dollar’s worth of credit; that is, an interest rate of 10 percent means each
dollar’s worth of credit costs 10 cents. We can easily modify the supply and
demand apparatus reviewed in the previous section to see how interest rates
are determined in the market for loanable funds.
The Demand for Loanable Funds

Individuals borrow funds for such things as mortgages, car loans, personal
loans, and education; businesses borrow funds to obtain capital equipment
and working capital; and governments borrow funds to pay for government
services, public investment in infrastructure, education, and so on. These
three groups comprise the primary demanders of loanable funds in our
economy.
Figure 4.8 shows a typical demand curve for loanable funds. This curve
reflects the quantity of loanable funds that will be demanded at alternative
nominal interest rates, holding everything else that might affect a borrower’s
According to the law Figure 4.8

of demand, when the Demand for Loanable Funds
nominal interest rate
falls from 10 to 5 per¬
cent and other things
remain constant, the
quantity demanded of
loanable funds rises
from $300 million to
$600 million. Thus,
the demand for loana¬
ble funds is downward
sloping. The move¬
ment from point A to
point B represents a
change in quantity de¬
manded.
Loanable Funds (Millions $)

decision constant. When the interest rate on loanable funds is 10 percent,

the quantity demanded of loanable funds is $300 million. When the interest
rate falls to 5 percent, the quantity demanded of loanable funds increases to
$600 million. Thus, the demand curve for loanable funds reflects the law of
demand: The quantity demanded of loanable funds by borrowers increases
as the price of credit (the interest rate) falls.
The demand curve for loanable funds holds everything constant except
the interest rate on the particular type of loan. The movement along a demand
curve, such as the one from A to B in Figure 4.8, denotes a change in quantity
demanded. Whenever a variable other than the interest rate on a particular
type of loan changes, it shifts the entire position of the demand curve. This
shift in the entire demand curve, as from D1 to either to D2 or D3 in Figure
4.9, reflects a change in demand.
Determinants of the Demand for Loanable Funds. Table 4.1

summarizes the determinants of demand for loanable funds and shows the
variables that can affect the willingness of either households, firms, or gov¬
ernment to borrow funds. These variables are discussed next.
Interest on Alternative Sources of Funds. Recall that the de¬

mand for a good or service depends on the prices of substitutes. The demand
A change in a deter¬ Figure 4,9

minant of the demand Changes in the Demand for Loanable Funds
for loanable funds re¬
sults in a change in
demand. An increase Interest
in the demand for Rate (/)
loanable funds is rep¬
resented by the shift
from D1 to D3. A de¬
crease in demand is
represented by the
shift from D1 to D2.
Loanable Funds
Table 4.1
Determinants of Demand for Loanable Funds
The position of the demand curve for loanable funds depends on the interest rate on
alternative sources of funds, inflationary expectations, the tax deductibility of interest pay¬
ments, the taste for borrowing, the profitability of business projects, and the size of the
government budget deficit. The pluses in the right-hand column indicate that increases in
any of these variables lead to a rightward shift in the demand for loanable funds.
Effect of an Increase in the

Variable on the Demand For
Variable Loanable Funds
Interest rate on alternative sources of funds +
Inflationary expectations +
Tax deductibility of household interest +

payments
Taste for borrowing +
Profitability of business projects +
Size of government budget deficits +
for loanable funds is no exception. A decrease in the interest rate on alter¬

native sources of funds induces borrowers to substitute away from existing
sources of funds and toward the sources with the lower interest rates.
Suppose a bank decides to offer a 1.9 percent interest rate on a new-car
loan—a much lower rate than that offered at car dealerships. We would now
expect fewer borrowers to obtain loans from car dealers and instead substitute
toward the lower-priced funds at the bank. The reduction in interest rates
offered by the bank thus leads to a decrease in the demand for loans from
car dealers, since the two sources of car loans are substitutes.
This relation works the same way for firms, except that firms borrow to
fund projects rather than buy cars. Regardless, the willingness of firms to
acquire a given type of loanable funds still depends on the interest rate on
substitute sources of funds. For instance, when IBM seeks funds for working
capital, it can borrow funds in the bond market or directly from a bank. An
increase in the rate IBM must pay on bonds will increase IBM’s demand for
bank loans; it substitutes toward them due to the higher cost of obtaining
funds in the bond market.
Inflationary Expectations. Since borrowers receive current funds in

exchange for the promise of future interest and principal payments, a bor¬
rower’s inflationary expectations affect her or his perceived real cost of
repaying the loan. For a given nominal interest rate, a higher expected
inflation rate translates into a lower real cost of borrowing, since the borrower
expects to pay back the loan with dollars that have lower purchasing power
due to inflation. For a given nominal interest rate, borrowers will therefore
increase their current demand for loanable funds whenever the expected rate
of inflation increases.
Figure 4.9 on page 105 illustrates the impact of a change in inflationary
expectations on the demand for loanable funds. The initial demand curve,
Dl, represents the quantity of loanable funds demanded at alternative nom¬
inal interest rates, holding everything else (including the expected rate of
inflation) constant. An increase in the expected rate of inflation shifts the
demand curve to the right to D3, where a greater quantity of funds is de¬
manded at each interest rate. A decrease in inflationary expectations shifts
the demand curve to the left to D2, where fewer funds are desired at each
interest rate.
Box 4.1 compares actual and expected inflation rates from 1982 to 1992.
Tax Deductibility of Household Interest Payments. In the

United States and many other industrialized countries, governments allow
households to deduct certain types of interest payments from income before
computing their tax liability. For instance, in the United States, households
can deduct mortgage interest payments from income before computing fed¬
eral taxes. The demand for mortgage funds by households therefore depends
in part on the marginal tax rate imposed on households.
To see why this is so, suppose a consumer with $100,000 in income
borrows $50,000 at an interest rate of 10 percent. If the government taxes
the consumer’s income at 30 percent and interest payments are not tax
deductible, the consumer pays $100,000 X .3 = $30,000 in taxes. However,
if interest payments are tax deductible, the consumer deducts the interest
payments (in this case, $50,000 X .1 = $5,000) from income before com¬
puting taxes:
Income $100,000
Less interest $5,000
Taxable income $95,000
If interest is deductible, the consumer’s tax liability is only $95,000 X .3

= $28,500, which is $1,500 less than it is when interest is not tax deductible.
The tax deductibility of interest payments reduces a borrower’s tax liability
and thus encourages more borrowing.
More generally, we saw in Chapter 3 that when the nominal interest rate
is i and the marginal tax rate is t, the effective aftertax interest rate is
z‘T = (1 — t) X i.
Actual and Expected Rates of Inflation
How accurate are expectations of inflation? The tion rate. During this period, the "experts" were
accompanying figure graphs the actual and ex¬ making forecasts that on average were too
pected inflation rates in the United States from high, or biased upward. From 1987 onward,
1982 through 1992. These expected rates of in¬ there was a less strong bias in the expected in¬
flation are based on surveys of economists who flation rates, although in the 1991-1992 period,
regularly forecast changes in the price level. a tendency to overstate the actual inflation rate
There are several things to notice about this recurred.
graph. First, the actual and expected inflation
rates tend to move together over time. Second,
the expected inflation rate is much smoother Source: Citibase electronic database. For actual inflation
rates, annual percentage change in the consumer price
and less volatile than the actual inflation rate.
index. For expected inflation rates, forecasts of annual
Third, during the period from 1982 through percentage change in the consumer price index by econ¬
1986, the expected inflation rate seems to have omists surveyed in the ASA-NBER Business Outlook
been systematically above the actual infla¬ Survey.
ro
cD
Cl
+->
c
cu
u
cu
a_
'82 '85 '88 '91 ■92

Notice that the aftertax interest rate (/T) decreases as the marginal tax rate
(t) increases. This illustrates that an increase in the marginal tax rate actually
reduces the effective (aftertax) cost of borrowing and thus increases the
demand for tax-deductible loanable funds such as mortgages.
Taste for Borrowing. Tastes also affect the demand for loanable
funds. Households, businesses, or governments may wish to borrow more
or fewer funds at each interest rate due to changes in family composition,
business status, or voter preferences. This change in tastes shifts the demand
for loanable funds to the right or the left, depending on the nature of the
change. Taste changes that result in a desire to borrow more at each interest
rate shift the demand curve to the right. When borrowers desire fewer
loanable funds at each interest rate, the demand curve shifts to the left.
Profitability of Business Projects. Changes in the expected prof¬

itability of business projects alter the willingness of firms to borrow funds
to finance the business projects and therefore affect the demand for loanable
funds. An increase in the profitability of business projects shifts the demand
for loanable funds to the right. This is because businesses now desire more
loanable funds at each interest rate to be able to undertake the more profitable
projects.
Magnitude of Government Budget Deficits. Federal, state, and

local governments also demand loanable funds to pay for expenses on such
things as education, defense, welfare payments, and highways. The larger
the projected budget deficit—that is, the greater the shortfall between tax
receipts and government expenditures—the greater the amount the govern¬
ment must borrow to pay for the projects. For this reason, increases in the
portion of the budget deficit financed through borrowing lead to increases
in the government’s demand for loanable funds. We will say much more
about this important issue in Chapters 16 and 21.
Summary | Graphically illustrate the impact of the following on the demand for loanable
Exercise 4A funds: (a) a reduction in inflationary expectations, (b) an increase in the
profitability of business projects, and (c) an increase in the size of the federal
government budget deficit.
Answer: (a) A reduction in inflationary expectations decreases the demand

for loanable funds from D] to D2 in Figure 4.9 on page 105. (b) An increase
in the profitability of projects funded through borrowing shifts the demand
curve for loanable funds from D] to D3 in Figure 4.9. (c) An increase in the
budget deficit leads to an increase in the demand for loanable funds from
D] to D3 in Figure 4.9.
The Supply of Loanable Funds

Demand reflects only one side of the market for loanable funds; the other
side of the market is supply. Suppliers of loanable funds give up the current
use of funds in exchange for interest payments. The law of supply as it
relates to loanable funds says that, holding other things constant, the higher
the nominal interest rate, the greater the quantity of loanable funds supplied
by lenders. Thus, a typical supply curve for loanable funds is upward sloping,
as in Figure 4.10. When the nominal interest rate is 5 percent, lenders provide
$200 million in loanable funds. When the interest rate rises to 10 percent,
the quantity supplied increases to $300 million.
The movement along a given supply curve for loanable funds, such as
the movement from A to B in Figure 4.10, denotes a change in the quantity
of loanable funds supplied. A change in the position of the supply curve
denotes a change in supply.
Determinants of the Supply of Loanable Funds. The major

determinants of the supply of loanable funds—those factors that affect its
position—are interest rates on alternative uses of funds, inflationary expec¬
tations, the tax rate on interest income, wealth, risk, and liquidity. Table 4.2
Table 4.2
Determinants of the Supply of Loanable Funds
The position of the supply curve for loanable funds depends on the interest rate on alternative
uses of funds, inflationary expectations, the tax rate on interest income, wealth, the riskiness
of the loan, and the liquidity of the loan. The pluses in the right-hand column indicate that
increases in wealth and liquidity shift the supply of loanable funds to the right. The minuses
indicate that increases in interest rates on alternative uses of funds, inflationary expectations,
the tax rate on interest income, or the riskiness of the loan shift the supply curve to the left.
Effect of an Increase in the Variable

Variable on the Supply of Loanable Funds
Interest rate on alternative uses of funds —
Inflationary expectations —
Tax rate on interest income —
Wealth +
Riskiness of loan —
Liquidity of loan +
summarizes the impact of these variables on the position of the supply curve
for loanable funds.
Interest on Alternative Use of Funds. The greater the interest

that could be earned by investing funds in some alternative use, the lower
the supply of loanable funds. For instance, Figure 4.11 shows the initial
supply of loanable funds, S[. When the interest rate that could be earned by
investing funds in an alternative investment rises, the supply of loanable
funds shifts to the left to S2. This means that at each interest rate, there is a
lower quantity of loanable funds than before. Similarly, if the interest that
could be earned on an alternative investment decreases, the supply curve
shifts to the right, such as from S] to S3 in Figure 4.11.
Inflationary Expectations. Suppliers of loanable funds give up the

use of money today in return for principal and interest at some future date.
The higher the expected rate of inflation, the less that can be purchased when
the principal and interest are repaid. Thus, increases in the expected rate of
inflation reduce the willingness of suppliers of loanable funds to lend money
Figure 4.11
Changes in the Supply of Loanable Funds
Any of the following will cause a decrease in the supply of loanable funds from S1 to 52: a
higher interest rate earned by investing funds in alternative uses, a higher expected infla¬
tion rate, an increase in the marginal tax rate, a decrease in wealth, an increase in the
riskiness of the loan, and a decrease in the liquidity of the loan.
Any of the following will cause an increase in the supply of loanable funds from S1 to
53: a lower interest rate earned by investing funds in alternative uses, a lower expected
inflation rate, a decrease in the marginal tax rate, an increase in wealth, a decrease in the
riskiness of the loan, and an increase in the liquidity of the loan.
Interest
Rate
at the given interest rate. This is reflected in Figure 4.11 as the decrease in
supply from S[ to S'2. Conversely, a reduction in the expected rate of inflation
increases supply from S[ to S3.
Tax Rate on Interest Income. Increases in the marginal tax rate

reduce the amount of interest income lenders get to keep, which reduces the
incentives to lend funds. In Figure 4.11, an increase in the marginal tax rate
decreases the supply of loanable funds from S1 to S2, since lenders are willing
to provide fewer funds at each interest rate. Similarly, a reduction in the
marginal tax rate increases the supply of loanable funds from Sl to S3. As
noted in Chapter 3, some forms of loanable funds (notably, municipal bonds)
are exempt from federal taxes on interest income. Box 4.2 shows how this
affects the relative rewards to investors on municipal versus other bonds.
Wealth. The wealth of lenders obviously affects their willingness and/

or ability to supply loanable funds. In general, the greater a lender’s wealth,
the greater the lender’s capacity to make loans, and thus the greater the
supply of loanable funds. Graphically, an increase in wealth increases the
supply of loanable funds, which is depicted in Figure 4.11 as the shift from
Sl to S3. Similarly, a decrease in wealth decreases the supply of loanable
funds from S] to S2.
Risk. The level of risk associated with a loan affects the lender’s willing¬
ness to make the loan, and thus affects the supply of loanable funds. We
will deal more explicitly with various types of risk in Chapter 9. For now it
will suffice to think of this risk as default risk. The greater the level of default
risk, the greater the likelihood the borrower will fail to make interest and/or
principal payments. This risk increases the interest rate necessary to induce
the lender to make the loan. More generally, an increase in the riskiness of
a loan decreases the supply of loanable funds, as the shift from S] to S2 in
Figure 4.11 shows. A reduction in risk has the opposite effect: It increases
the supply of loanable funds from S] to S3 in Figure 4.11.
Liquidity. The liquidity of an investment refers to the ease with which

an investor can convert the loan into cash without loss. If a lender needs
cash before the borrower is obligated to repay the loan, the lender can attempt
to sell the loan to someone willing to pay for it now in return for the future
principal and interest payments the initial borrower will make. The more
liquid a loan, the easier it is to find a person willing to buy it.
The underlying liquidity of the loan thus affects the loan’s supply. The
more liquid the loan, the easier it is to sell at some future date, and thus the
more willing a lender is to make the loan in the first place. Thus, an increase
in the liquidity of loans will increase the supply of loanable funds, such as
the shift from S] to S3 in Figure 4.11. As you would expect, decreased
liquidity shifts the curve in the opposite direction, from S] to S2 in Figure
4.11.
Summary I One proposal that has floated around Washington, D.C., is to enact a trans¬
Exercise 4.5 action tax on financial transactions. This tax would be a fee of 1 percent or
so on the value of specified financial transactions. Such a fee would dis¬
courage short-term investment. How would such a fee affect the supply of
loanable funds? Would it have the same effect on the supply of one-year
and two-year loans?
Answer: A transaction tax means the supplier of loanable funds would

receive as payment the interest rate on loanable funds minus the tax. For
example, a one-year loan of $1,000 might earn 10 percent interest, but after
paying a 1 percent transaction tax, the net interest received by the supplier
Tax Rates and the Aftertax

Interest Rate in Two States
Income taxes are levied not only by the federal Interest Aftertax Interest
government but also by state governments and Rate Rate
sometimes county or city governments. Further¬ California Texas

more, the income tax rate on each additional Du Pont, 8.1% 4.70% 5.51%
dollar of income, called the maTginal tax rate, matures 2006
varies with the level of income. For the U.S. fed¬
U.S. Treasury bond, 6.11 4.15 4.15
eral income tax, the top marginal rate for 1992 matures May 2003
taxes was 32 percent. Along with this, we will
San Antonio bond, 5.87 5.28 5.87
consider the top marginal tax rate for the two matures August 2013
most populous states. The first, California, had a
California government 5.5 5.5 5.5
top marginal tax rate of 10 percent in 1992. The
bond, matures April 2019
second, Texas, does not have a state income
tax, so the marginal tax rate is zero.
Source: The Wall Street Journal, May 3, 1993.
The accompanying table illustrates what this
means for aftertax interest rates. Consider a
bond issued by Du Pont, which had a yield of Finally, what about municipal bonds, which
8.1 percent on April 30, 1993. What was the are exempt from federal income tax and from
aftertax interest rate for a person living in state tax in the state in which they are issued.
Texas? The aftertax interest rate was the interest We will look at two cases. First, consider a bond
rate minus the total tax rate times the interest issued by the city of San Antonio, Texas. This
rate, or 8.1% - (.32 X 8.1%) = 5.51%. How bond yielded 5.87 percent on April 30, 1993. In
does this compare to California? The aftertax in¬ Texas, which has no state tax, this bond also
terest rate for a person living in California was yielded 5.87 percent after tax. In California this
8.1% - [(.32 + .10) X 8.1%] = 4.70%. Thus, bond's interest is still subject to a state tax of 10
a California resident would receive a lower percent, so the aftertax interest rate is 5.87% -
aftertax return to investing in this Du Pont bond (.1 X 5.87%) = 5.28%.
than a person living in Texas would. Second, consider a municipal bond issued
Next, consider a U.S. Treasury bond, which by the state of California that matures on April
is exempt from state and local taxes. A Treasury 1, 2019, and yielded 5.5 percent on April 30,
bond maturing in May 2003 yielded 6.11 per¬ 1993. This bond is subject to neither federal nor
cent on April 30, 1993. What was the aftertax state tax, so a California resident receives an af¬
rate of return in Texas? It was 6.11 % - .32 x tertax rate of 5.5 percent. Moreover, the Texas
6.11% = 4.15%. In California it was the same, resident pays no state income tax, so that per¬
since these bonds are exempt from state tax. son also receives an aftertax rate of 5.5 percent.
would be 9 percent. This would lead to a reduction in the supply of loanable

funds.
Note, however, that on a two-year loan, the 1 percent transaction tax
would be spread over two years in calculating net interest. For example, a
two-year loan of $1,000 might earn 10 percent interest each year. After
paying the 1 percent transaction tax, the supplier would receive net interest
of 9 percent in year 1 and 10 percent in year 2, for an average of 9.5 percent.
Thus, the supply of loanable funds for two-year loans also would be reduced,
but by a smaller amount than the supply of one-year loans. Although both
supply curves would be reduced by the transaction tax, the supply of short¬
term loans would decrease more than the supply of long-term loans.
Equilibrium in the Market for Loanable Funds

As in any market, equilibrium in the market for loanable funds is determined
by the interaction of all suppliers of and demanders for those funds. Figure
4.12 graphs the demand for and supply of loanable funds. Suppose the
interest rate is relatively low, say, 5 percent. At this rate, the quantity de¬
manded of loanable funds is $400 million, while the quantity supplied is
only $200 million; there is a shortage of $200 million in loanable funds. In
contrast, when the interest rate is relatively high, such as 15 percent, the
Equilibrium in this
market for loanable Equilibrium in the Market for Loanable Funds
funds is achieved
when the interest rate
is 10 percent, where
the supply and de¬
mand curves intersect.
At this interest rate,
the quantity supplied
of loanable funds
equals the quantity
demanded. There is
no tendency for the
interest rate to rise
above or fall below 10
percent unless some¬
thing changes that
shifts the position of
the supply or demand
curves.

quantity demanded of loans is only $200 million, but the quantity supplied
is $400 million; thus, there is a surplus of loanable funds at this interest rate.
Notice, however, that when the interest rate is 10 percent, the quantity
demanded of loanable funds is $300 million and the quantity supplied is
$300 million. At this point the market is in equilibrium, with no tendency
to change further. Everyone willing and able to pay an interest rate of 10
percent is able to obtain a loan, and everyone willing and able to provide
loans at 10 percent is able to lend at this rate, so neither a shortage nor a
surplus exists. Thus, the equilibrium interest rate is 10 percent, and the
equilibrium quantity of loanable funds is $300 million.3
Applications of the Loanable Funds Model

Now that you understand how the demand for and supply of loanable funds
interact in a free market to reach a point of equilibrium, we may apply these
tools to analyze how changes in the economic environment affect the equi¬
librium interest rates paid by households, businesses, and government.
The Tax Reform Act of 1982 and

Interest Rates on Consumer Credit
Prior to the Tax Reform Act of 1982, interest paid by households on con¬
sumer loans (loans for car purchases, vacations, and other products) and
credit card balances was fully tax deductible. The Tax Reform Act phased
out the deductibility of interest on consumer loans and credit card balances,
and, as of 1994, such interest payments are no longer tax deductible. How
did this act affect the interest rate consumers pay for credit?
Figure 4.13 provides the answer. D] and S1 represent, respectively, the
demand for and supply of consumer credit when interest payments were tax
deductible. The effect of eliminating the tax deductibility of interest pay¬
ments on consumer loans was to decrease the demand for consumer loans
to D2, since households were willing and able to obtain fewer consumer
loans at each interest rate when the interest payments are not tax deductible.
Thus, the elimination of tax deductibility of consumer loans decreased the
equilibrium interest rate on those loans from 21 to 15 percent and decreased
the equilibrium quantity of the loans from $200 million to $150 million.
Over the past decade, other changes (most notably drastic reductions in
inflationary expectations) have led to further reductions in the interest rate
on consumer credit.
3 The loanable funds model used here determines the equilibrium stock (or quantity) of loanable
funds at a point in time. In contrast, some economists use a loanable funds model that determines
a flow of funds per unit of time. The flow model is more difficult to work with and will not be
used in this text.
Eliminating the tax de¬ Figure 4.13

ductibility of interest Effect of Eliminating Tax Deductions for Interest Payments
payments on con¬ on Consumer Loans
sumer loans reduces
demand for these
loans from D1 to D2.
This leads to a lower
equilibrium interest
rate and fewer con¬
sumer loans.
Consumer Loans (Millions $)
Inflationary Expectations and the Mortgage Interest Rate

As noted, one dramatic change during the past decade was a large reduction
in the inflation rate and inflationary expectations. At the same time, mortgage
interest rates fell substantially, reaching a 30-year low by 1994. We can use
our loanable funds framework to see why reduced inflationary expectations
contributed to the fall in mortgage rates.
Figure 4.14 presents an initial supply curve (S'1) and demand curve (Z)1)
for mortgages, with a corresponding equilibrium mortgage interest rate of
15 percent and equilibrium quantity of $400 million mortgages at point A.
Since inflationary expectations are a determinant of both the demand for and
supply of loanable funds, a reduction in inflationary expectations of 7 percent
will affect both the demand for and supply of mortgages. In particular, a
reduction in inflationary expectations decreases the demand for mortgages
to D2, since households are willing and able to obtain fewer mortgages at
each interest rate. On the supply side, a reduction in inflationary expectations
shifts the supply curve from S] to S2 in Figure 4.14, since suppliers are
willing to loan more funds at each interest rate. The intersection of the new
demand curve D and the new supply curve S corresponds to a lower
equilibrium interest rate of 8 percent at point B.
Notice that a lower interest rate results regardless of how far we shift
the demand and supply curves. In Figure 4.14, however, the supply and
demand curves shift such that the equilibrium quantity of mortgages is the
same as the initial level, and the equilibrium nominal interest rate declined
by the same amount as the reduction in inflationary expectations (7 percent).
As we discussed in Chapter 3, a situation where a 7 percent reduction in
inflationary expectations leads to a 7 percent reduction in the equilibrium
nominal interest rate (leaving the real interest rate unchanged) is known as
the Fisher effect. Box 4.3 shows the link between inflationary expectations
and nominal interest rates in the United States for the 1982-1992 period.
Returns on Real Estate Investments

and Corporate Interest Rates
In the late 1970s and early 1980s, many investors began using their funds
to invest in real estate, perceiving they could realize a greater return than
they could by investing in other investments such as bonds. What was the
impact of the shift into real estate on the corporate bond market, the major
source of long-term business borrowing?
Figure 4.15 shows the market for corporate bonds with a face value of
$1,000. The initial equilibrium is at point A, where S1 and D] intersect at an
interest rate of 10 percent. The increase in return on alternative investments
Nominal Interest Rates and

Expected Inflation Rates
The accompanying figure shows the interest rate equation implies that the real interest rate is the
on one-year U.S. Treasury bills and the expected nominal interest rate minus the expected rate of
inflation rate from 1982 through 1992. Notice inflation, which is the difference between the
that the interest rate and the expected inflation graph of the interest rate and the graph of ex¬
rate tend to move together, but the relationship pected inflation. In 1984 the real interest rate
is not very tight. Thus, both the interest rate and was 5 percent, while by 1988 it had fallen to
the expected inflation rate were high in 1982. about 2.5 percent. By the end of 1992, the real
However, expected inflation rates declined from interest rate was about zero. Thus, we find that
1982 through 1987, while the nominal interest the real interest rate has been quite variable
rate first declined and then rose sharply in 1984 over time.
before declining again. From 1987 to 1990,
both the nominal interest and expected inflation
rates increased, and both then declined through Source: Citibase electronic database. For the interest rate,
the one-year Treasury bill rate. For expected inflation,
the end of 1992.
forecasts of annual percentage changes in the consumer
Another way to look at this graph is to cal¬ price index by economists surveyed in the ASA-NBER
culate the implied real interest rate. The Fisher Business Outlook Survey.
such as real estate reduced the supply of funds in the corporate bond market,
resulting in a shift in the supply of loanable funds in that market from S1 to
r\
S . As we see, the impact of this shift was to increase the interest rate on
corporate bonds to 12 percent.
Of course, between 1980 and 1994, the return on real estate investments
was very low, and as a result investors have been leaving the real estate
market in droves. This recent decline in the return on real estate has increased
the supply of loanable funds in the corporate bond market, contributing
(along with lower inflationary expectations) to the low interest rates on
corporate bonds that prevail today.
Recessions and the Prime interest Rate

Creditworthy businesses can obtain short-term funds to use as working cap¬
ital in the prime lending market. The prime interest rate tends to vary over
time, especially during periods of economic boom and recession. We can
use our loanable funds model to see how a period of recession might affect
the prime interest rate.
The market for prime lending is presented in Figure 4.16, which depicts
the initial demand (Dl) and supply (Sl) of loanable funds, resulting in an
equilibrium at point E, where the interest rate is 7 percent and the corre¬
sponding equilibrium quantity is $500 million. Let us suppose the economy
is moving into a recession—a period of rising unemployment in which the
profitability of business projects declines. What impact does this trend have
on the market for prime lending? Since the profitability of business projects
is a determinant of the demand for loanable funds, businesses decrease their
demand for loanable funds to D2. This results in a movement in equilibrium
from point E to point A. The prime interest rate falls to 5 percent, and the
equilibrium quantity of funds falls to $400 million.
Figure 4,16
Recessions and the Prime Interest Rate
:
The initial equilibrium is at point E. When a recession occurs, one effect is to decrease the
demand for loanable funds due to a decline in the profitability of business projects. This is
represented as a decrease in demand from D1 to D2, which tends to decrease the equilib¬
rium interest rate and quantity of loanable funds to point A. However, recessions may also
reduce suppliers' wealth, which would shift supply from 51 to 52, resulting in an equilibrium
at point B. In addition, recessions may increase the risk of loaning funds, which could fur¬
ther decrease supply from S2 to S3. Notice that, depending on the magnitude of the shifts
in supply and demand, the interest rate may go up (point C) or down (point A) during
recessions. However, the equilibrium quantity of loans will clearly fall during recessions.
Short-term Business Loans (Millions $)

122 Chapter 4 S u pply an d Demand in Financial Markets
Do recessions necessarily result in a lower prime rate and less borrow¬

ing? If the only variable affected by recession is the profitability of business
projects, the answer is yes. However, recessions also frequently affect two
determinants of the supply of loanable funds: wealth and risk. Specifically,
during recessions many lenders reduce their wealth as they draw on their
assets. This decreases the supply of loanable funds—say, to S2 in Figure
4.16. Furthermore, the reduced profitability of business projects increases
the risk of lending in the prime market; this further decreases the supply of
lending—say, to S3. As we see, each of these changes tends to reduce the
amount of borrowing during recessions. Whether the interest rate falls (point
A), stays the same (point B), or rises (point C) depends on the relative
magnitude of these effects, although the prime rate frequently rises during
recessions.
The Federal Deficit and Interest

Rates on Government Bonds
How does the growing federal budget deficit affect the interest rate the
federal government must pay in the bond market? In Figure 4.17, the initial
equilibrium in the government bond market is at point A, where Sl and Dl
intersect. A larger budget deficit increases the demand for loanable funds to
the right to D2 as the government attempts to obtain more funds to pay for
its programs. This increases the equilibrium quantity of funds obtained from
$400 million to $500 million, but at a higher interest rate. In this case, the
increase in the budget deficit increased the interest rate from 5 to 7 percent.
We examine this link between budget deficits and interest rates in more
detail in Chapters 16 and 21.
Tax Policy and Interest Rates on Municipal Bonds

Municipal governments issue bonds to obtain funds to pay for local projects.
One of the more interesting features of municipal bonds is that the interest
lenders receive on the bonds is exempt from federal income taxes. This
makes the bonds attractive to individuals in higher income brackets.
With all the debate in Washington about raising tax revenue to reduce
the federal budget deficit, it is interesting to speculate what would happen
if Congress eliminated the exemption on municipal bond interest in an
attempt to gain additional federal income tax revenue. In other words, sup¬
pose Congress took action that made it necessary for lenders of funds in the
municipal bond market to pay federal income taxes on municipal bond
interest. This would raise tax revenue for the federal government, but what
would be the impact on interest rates your state paid on its borrowing?
Point A in Figure 4.18 represents the initial equilibrium in the market
for municipal bonds. Sl and D1 are drawn under the assumption that munic¬
ipal bond interest is tax deductible. If municipal bond interest became subject
to federal income taxes, lenders would provide fewer funds at each interest
rate, because the taxes would lower the aftertax rate of interest. This would
result in a decrease in the supply of funds in the municipal bond market,
shifting the supply curve to the left to S~ and moving equilibrium to point
B. Thus, such a tax change would raise federal government revenue at the
expense of municipal governments, which would have to pay a higher rate
of 7 percent. In short, if the interest on municipal bonds were not tax
deductible, municipal governments would pay a higher interest rate and
obtain fewer loanable funds for municipal projects.
Interest Rate Ceilings

Until now, all of our applications have assumed that interest rates are free
to rise or fall to achieve a point of equilibrium between the demand for and
supply of loanable funds. This is not the case when interest rate ceilings are
in effect. The term interest rate ceiling (also called a usury law) refers to a
specified amount above which the interest rate is not permitted to rise. Until
the early 1980s, interest rate ceilings were mandated by Regulation Q, which
prohibited banks from paying interest on certain checking accounts and set
maximum rates on savings and time deposit accounts. Although ceilings on
deposits were phased out by the Depository Institution Deregulation and
Point A represents a Figure 4.18

situation where the in¬ Federal Tax Policy and Municipal Bond Interest Rates
terest on municipal
bonds is tax deducti-
ble. If the federal gov- Interest Rate
ernment eliminates On Municipal
the tax deductibility of Bonds(%)
municipal bond inter- 5c2
est, the supply of
loanable funds shifts
\ i
\ 5
to S2, resulting in a
\
new equilibrium at
point B. As a conse- 7 \
quence, municipal \
governments will pay | \
a higher interest rate.
V
4
i \
* jf' ' *
' ;
400 500
Loanable Funds in Municipal Bond Market
(Millions S)
Monetary Control Act of 1980, some states still impose interest rate ceilings
through state laws that regulate credit card and other types of interest.
When the interest rate is not allowed to rise to clear the market for
loanable funds, a shortage results. This is illustrated in Figure 4.19, where
we examine the impact of a 5 percent interest rate ceiling on the market for
loanable funds. Given the ceiling of 5 percent, lenders provide only $100
million in funds, while the quantity demanded of loanable funds by borrow¬
ers is $400 million. There is a shortage of $300 million in loanable funds,
indicating that some would-be borrowers are unable to obtain loans. We
look at further ramifications of interest rate ceilings in Chapter 7.
Interest Rate Floors

Some financial instruments, such as U.S. savings bonds, have features that
prevent the interest rate from falling below some specified amount, or in¬
terest rate floor. Other examples include pension funds that stipulate a
minimum guaranteed interest rate and adjustable-rate mortgage contracts
that stipulate that the interest rate, which varies throughout the term of the
loan, cannot fall below some specified amount.
Figure 4.20 shows the impact of an interest rate floor of 15 percent on

the market for loanable funds. When the interest rate is 15 percent, the
quantity of funds supplied is $400 million, whereas only $100 million in
funds is demanded at that interest rate. In other words, the interest rate floor
creates a surplus of loanable funds of $300 million. Normally a free market
automatically corrects this problem by driving down the interest rate until
equilibrium between the quantity demanded of loanable funds and the quan¬
tity supplied is reached. The interest rate floor “disables” the mechanism
that would eliminate the surplus.
Summary The federal government is projected to have a large deficit next year. Suppose
Exercise 4.6 that, instead of borrowing funds to finance the deficit, the government raised
the tax rates for all individuals, borrowers and lenders alike. What would
happen to the equilibrium interest rate and quantity of loanable funds in (a)
the government bond market and (b) the mortgage market?
(a) The initial equilibrium in the market for government bonds

is at point E in part a. Since the government deficit is not funded through
borrowing, the projected deficit does not affect the federal government’s
demand for loanable funds, which remains at D°. However, higher tax rates
reduce the aftertax interest rate lenders (suppliers of loanable funds) receive
at every interest rate. This decreases the supply of loanable funds from 5°
to S1, resulting in a higher interest rate on government borrowing (i]) and a
lower quantity of loanable funds in the government bond market (L1). (b)
The initial equilibrium in the mortgage market is at point E in part b. An
increase in the tax rate increases the demand for mortgages from Z)° to Dl,
because mortgage interest is tax deductible. The higher tax rate means the
aftertax interest rate paid by borrowers has fallen. Furthermore, the increase
in the tax rate reduces the supply of loanable funds from S° to some curve
to the left, such as S1, S2, or S3. If the supply curve shifts by a small amount—
say, to Sl—the new equilibrium occurs at point A (a higher interest rate and
a greater quantity of mortgages). If the supply curve shifts farther to the left,
such as to S", the equilibrium occurs at point B (a higher interest rate but
the same quantity of mortgages as that before the tax increase). But if the
supply curve shifts even farther to the left, such as to S3, the new equilibrium
is at point C (a higher interest rate and a lower quantity of mortgages). Thus,
regardless of the magnitude of the shifts in the supply and demand curves,
the mortgage interest rate rises. However, the impact of the increase in the
marginal tax rate on the equilibrium quantity of mortgages is indeterminant;
it could rise (point A), stay the same (point B), or fall (point C), depending
Key Terms 127
on the relative magnitude of the increase in demand and the decrease in

supply.
Interest Rate
Government
Bonds
Loanable Funds in Loanable Funds in

Government Bond Market Mortgage Market
(a)
Conclusion
In this chapter, we examined the fundamental tools of supply and demand

analysis in the context of financial markets. In particular, we analyzed the
market for fee-based financial services and the loanable funds market. Equi¬
librium in each market is determined by the intersection of the supply and
demand curves. In markets for fee-based financial services—markets for
insurance, ATM transactions, wire transfers, and the like—equilibrium de¬
termines a price for the services such that quantity demanded equals quantity
supplied. In markets for loanable funds, equilibrium determines the interest
rate borrowers pay lenders for the use of funds.
We also discussed the major determinants of demand and supply and
analyzed how changes in the determinants of supply and demand lead to
changes in the equilibrium price and quantity, or interest rate and quantity,
in these markets. These tools of supply and demand analysis will serve us
well in the remainder of this text. We will have occasion to apply them in a
wide variety of settings and to a broad array of markets.
KEY TERMS
fee-based financial service law of demand

demand curve change in quantity demanded
KEY TERMS continued
change in demand determinants of supply

determinants of demand shortage
substitutes equilibrium
complements surplus
supply curve liquidity
change in quantity supplied loanable funds
law of supply interest rate ceiling
change in supply interest rate floor
1. What is the price of a fee-based financial 5. During the 1980s, federal budget deficits
service? What is the price of an interest- were financed through borrowing. Graphi¬
rate-based financial service? Is it possible cally illustrate the impact of financing the
to convert the interest-rate-based price of budget deficit with federal borrowing on
a given financial service into a fee-based (a) . The market for government bonds
price, and vice versa? Explain. (b) . The market for mortgages
(c) . The market for new-car loans
2. What is the difference between (a) an in¬
crease in demand and (b) an increase in 6. Other things being the same, would you
quantity demanded? Explain and provide expect the equilibrium interest rate to be
a diagram that illustrates the difference. higher on a 1-year loan or on a 20-year
3. Mandatory seat belt laws have reduced the loan? Use supply and demand analysis to
number of automobile liability insurance explain your answer.
claims. What is the impact of this result
on the market price and quantity of auto¬ 7. Other things being the same, would you
mobile liability insurance? Explain care¬ expect the equilibrium interest rate to be
fully. higher or lower on a small-business loan
or on a large-business loan? Explain using
4. During the late 1970s, the interest rate on supply and demand analysis.
money market mutual funds increased
from 10 to 15 percent. Banks and savings 8. Suppose the mortgage market is initially
and loans, however, continued to pay only in equilibrium (label this point A in a sup¬
5 '/4 percent interest on savings accounts. ply and demand diagram). Lenders expect
(a) . What impact did this increase in the the inflation rate to decline, whereas bor¬
interest rate have on the market for sav¬ rowers expect it to rise. Show the new
ings accounts at banks? equilibrium in the market for mortgages,
(b) . Graphically illustrate the impact of and label it point B. What can you say
the Depository Institution Deregulation about the direction of the change in (a)
and Monetary Control Act of 1980 on the the interest rate and (b) the quantity of
market for money market mutual funds. mortgages?
Selections for Further Reading 129
9. Show the impact on the market for per¬ (d) . Savings deposits
sonal loans of (e) . New-car loans
(a) . A government law that permits bor¬ (f) . Bank wire transfers
rowers to deduct interest payments on per¬ (g) . Long-term business loans
sonal loans for tax purposes
13. An article in the Wall Street Journal re¬
(b) . An increase in the marginal tax rate
ported that the return on investing in
(c) . A simultaneous allowance for bor¬
rowers to deduct interest payments and a stocks in 1996 is expected to be 5 percent
higher than normal. Assuming investors
rise in the marginal tax rate
believe this report, what would you expect
10. Use supply and demand analysis to ex¬ to happen to (a) mortgage interest rates,
plain the likely impact of a recession on (b) consumer loan rates, and (c) interest
the market for rates on federal and municipal borrowing?
(a) . ATM transactions
(b) . Life insurance 14. Moody’s bond rating service recently low¬
(c) . Mortgages ered its rating of bonds issued by several
(d) . Savings deposits large insurance companies and pension
(e) . New-car loans funds, which means loans to these compa¬
(f) . Bank wire transfers nies are now more risky. Graphically il¬
(g) . Long-term business loans lustrate the impact of this action on the mar¬
ket for bonds issued by these companies.
11. Use supply and demand analysis to ex¬
plain the likely impact of an increase in 15. Critically evaluate this statement: “An in¬
the marginal tax rate on crease in the demand for mortgages in¬
(a) . ATM transactions creases the mortgage interest rate, which
(b) . Life insurance in turn leads to an increase in the supply
(c) . Mortgages of funds for mortgages. This increase in
(d) . Savings deposits supply reduces the mortgage interest rate.
(e) . New-car loans Thus, an increase in the demand for mort¬
(f) . Bank wire transfers gages leads to a self-correcting increase in
(g) . Long-term business loans supply, which tends to stabilize interest
rates at their initial level.”
12. Use supply and demand analysis to ex¬
plain the likely impact of a reduction in 16. Suppose the marginal tax rate is 33 per¬
inflationary expectations on the market for cent. Determine the effective aftertax in¬
(a) . ATM transactions terest rate paid by a borrower with
(b) . Life insurance (a) . A 12 percent mortgage
(c) . Mortgages (b) . A 10 percent consumer loan
Fleisher, B. M., and K. J. Kopecky. “The Loanable kets.” Journal of Urban Economics, 26 (Septem¬
Funds Approach to Teaching Principles of Macro¬ ber 1989), pp. 212-230.
economics.” Journal of Economic Education, 18 Mitchell, W. E., and R. L. Sorensen. “Pricing, Price
(Winter 1987), pp. 19-33. Dispersion, and Information: The Discount Broker¬
Flenderson, J. V., and Y. M. Ioannides. “Dynamic age Industry.” Journal of Economics and Busi¬
Aspects of Consumer Decisions in Housing Mar¬ ness, 38 (December 1986), pp. 273-282.
CHAPTER
The Bank as a Firm: Loans
n the previous chapter, we used the supply and demand model to

analyze financial markets. We learned that the intersection of the supply and
demand curves for funds determines the funds’ market “price,” the interest
rate. How does this fact relate to banks, and in particular to the loan decisions
of banks? In this chapter, we answer this question. We see how the market
demand and supply for loans affects individual banks. We then focus on the
behavior of individual banks, with particular emphasis on viewing banks as
firms, similar to firms in other markets in the economy. Banks produce an
output—loans—using inputs such as deposits, labor, and capital equipment.
The banking industry is heavily regulated, and in Chapter 7 we discuss
how the regulation of banking has changed from the early days of the
republic to the present. One important set of regulations that governed bank
behavior since before World War II limited the ability of banks to change
the interest rates offered to depositors or those charged for loans. The passage
of the Depository Institution Deregulation and Monetary Control Act
(DIDMCA) in 1980 altered these regulations and began a process of partial
deregulation, eliminating many of the regulations on deposit interest rates.
These tremendous changes in the banking industry make it necessary to look
at individual bank behavior to analyze how these banks determine the most
profitable quantity of and interest rates on loans and deposits.
Modern banks produce a variety of banking services, and in this chapter
we focus on one of the most important outputs banks produce: loans. (In the
next chapter, we examine bank decisions related to deposits.) We look at
how banks determine the profit-maximizing quantity of loans to issue. As
we will see, the quantity of loans and the interest rate charged depend on
the degree of market power the bank enjoys. We begin by analyzing the
situation where a bank is small relative to total loan demand and thus has
no market power. Then we turn to the case where a bank is large relative to
total loan demand and thus has some market power. Finally, we examine
the impact on bank decisions of imperfect information regarding whether
borrowers will repay loans.
Purely Competitive Banks
When individual banks are small relative to the market, they cannot influence
the interest rates they charge for loans. In other words, they are price takers.
130
This situation typifies a model called pure competition among banks. The
market for bank loans is purely competitive if
1. There are many buyers and sellers of bank loans, each of which is
“small” relative to the whole market.
2. Each bank provides similar bank loans; that is, no product differentiation
exists.
3. Buyers and sellers of banking services have full information about cur¬
rent market interest rates.
4. There are no transactions costs (the costs associated with securing and
making loans).
The conditions for pure competition are unlikely to hold exactly, but
they serve as a useful benchmark and may approximate reality in some cases.
Many markets for banking services have numerous buyers and sellers (or
demanders and suppliers) of loans, especially those in large cities. Moreover,
loans made by different banking firms are similar in nature. It makes little
difference in most cases whether a loan is provided by First City or First
National. All that matters is the terms of the loan, which include the interest
rate on the borrowed funds. Suppliers and demanders of loans are likely to
be well informed about alternative sources of loans and the rates charged.
Finally, while transactions costs are seldom zero, they are unlikely to be a
dominant portion of the cost of demanding or supplying a loan.
If you remember the theory of perfect competition from your principles
of economics course, you may notice a similarity between pure competition
and an alternative industry model known as perfect competition. The key
difference between these two models is that a perfectly competitive market
has, in addition to the four characteristics listed above, a fifth characteristic
known as free entry and exit. Pure competition does not require that there
be free entry into and exit from the banking industry.
The reason we use the model of pure competition to analyze banks that
lack market power is that there is no free entry into banking. Despite the
deregulation of bank interest rates, entry into the banking industry is still
heavily controlled by federal, state, and local regulations. Among other
things, this implies that it is difficult for an entrepreneur to open a bank. A
maze of applications and other red tape must be approved at many levels of
government before a bank can be chartered. Exit is less of a problem, since
banks can be sold or, with sufficient time and care, liquidated.
In a purely competitive banking market, no individual bank has market
power and in effect must charge the “market-determined” interest rate. The
reason is simple: Since there are many banks, each offering loans that are
essentially identical, borrowers would just as soon obtain a loan from one
bank as from any other. Since there are no transactions costs, borrowers will
obtain loans from the bank that offers the lowest interest rate. If one bank
attempted to charge a higher interest rate than all other banks, borrowers
would go elsewhere for their loans.
132 Chapter 5 The Bank as a Firm: Loans
Market Demand and the Demand

for an Individual Bank's Loans
v
In a purely competitive banking market, the intersection of the market supply

and demand curves determines the equilibrium interest rate on loans, as in
part a of Figure 5.1. Banks supply loanable funds, while households, busi¬
nesses, and government demand them. The equilibrium loan interest rate is
10 percent. The equilibrium dollar value of loans, measured on the horizontal
axis, is $500 million.
Recall that in purely competitive markets, individual banks are price
takers; that is, they take the market interest rate as given. This is illustrated
in part b of Figure 5.1 by the horizontal demand for loans an individual bank
faces. Individual banks must take the equilibrium interest rate on loans as
given and can provide as many loans at this rate as they choose to make.
It is important to stress that the demand for loans at the market level is
downward sloping, whereas the demand for loans provided by an individual
purely competitive bank is horizontal. Borrowers are much more sensitive
to a change in the interest rate an individual bank charges than to a change
in the market interest rate, because at the market level, few substitute sources
of funds exist. In contrast, there are many substitutes for loans provided by
an individual bank (namely, loans at other banks).
Elasticity of Demand for Loans. Economists use a concept called

the elasticity of demand to measure the sensitivity of demand to price
changes. In the loan market, the elasticity of demand for loans measures
how responsive the quantity demanded of loans is to a change in the interest
rate; it is the percentage change in the quantity demanded of loans divided
by the percentage change in the interest rate on loans. Formally, if L is the
quantity demanded of loans and iL is the loan interest rate, the interest
elasticity of demand for loans is
_ %AL
E ~ %Ml
where A means “a change in.” If E is greater than 1 in absolute value, a 1

percent increase in the interest rate leads to a greater than 1 percent reduction
in the quantity demanded of loans. In this case, the demand for loans is
elastic (that is, responsive) to a change in the interest rate. In contrast, if E
is less than 1 in absolute value, a 1 percent increase in the interest rate
reduces the quantity demanded of loans by less than 1 percent. In this case,
the demand for loans is inelastic (that is, not very responsive) to a change
in the interest rate.
The formula for interest elasticity of loan demand contains a pitfall you
should avoid. In particular, since interest rates are stated in percentages, such
as 5 percent or 10 percent, you may be tempted to think that an increase or
decrease in the interest rate of 1 percent—say, from 5 to 6 percent—is a
percentage change of 1 percent. This is not correct. The percentage change
formula is AiLliL, just as it is for prices. Thus, an increase in the interest rate
from 5 to 6 percent is a percentage change of (6 — 5)/5 = .20, or a 20
percent change.
The market demand for loans is more inelastic than the demand for loans
by an individual bank. In fact, the demand for an individual bank’s loans is
perfectly elastic when the market is purely competitive. This is because if a
purely competitive bank increases its interest rate even slightly above the
market rate, it will lose all of its loan customers. Thus, the elasticity of
demand for a purely competitive bank’s loans is infinite in absolute value,
or perfectly elastic. This is why the demand for the loans provided by an
individual bank is horizontal at the market-determined interest rate on bank
loans.
134 Chapter5 The Bank as a Firm: Loans
The Purely Competitive Bank's Loan Decision

Supply and demand at the market level determine the market interest rate
on loans. A purely competitive bank must take this loan rate as given and
determine how many loans to issue at this rate. Remember, the bank’s goal
is to maximize its profits. How many loans will a profit-maximizing bank
issue? To answer this question, we must understand the nature of bank
revenues and costs, since profits are the difference between revenues and
costs. Remember too from your principles of economics course the distinc¬
tion between accounting profits and economic profits. Economic profits are
the difference between revenues and total opportunity cost. Since accounting
costs do not include all opportunity costs, accounting profits are higher than
economic profits. Thus, some banks can earn zero economic profits but report
earning (accounting) profits on their income statements. These banks would
be earning just enough accounting profits to cover the opportunity cost of
resources tied up in the banks.
The bank’s revenues from issuing L loans are given by
R = iL X L,
where iL is the market interest rate on loans and L is the dollar value of loans
provided by the individual bank; that is, revenues are merely the interest rate
on loans times the dollar value of loans issued. For example, if a bank issues
$1 million in loans at an interest rate of 10 percent, the bank’s revenue
(interest income) is $1,000,000 X .10 = $100,000. Part a of Figure 5.2
graphs the revenues of an individual bank as a function of the dollar amount
of loans on the horizontal axis. Notice that this relation is linear, since the
bank receives the same interest rate, iL, on each additional dollar in loans.
Furthermore, revenues increase as more loans are made. When LA loans are
made, revenues are $150 million. When loans increase to LM, revenues
increase to $500 million.
The cost of issuing L loans is represented as C(L) and consists of two
components. The first is the cost of funds, since banks require funds in the
form of reserves to make loans. These reserves can be obtained by attracting
deposits, and the cost of deposits is the interest rate paid on deposits in the
bank times the quantity of deposits. The second component is the general
cost of administering the bank, which includes the cost of staffing the deposit
window with tellers, the cost of the loan officers and staff needed to process
loan applications, other operating costs such as the building and utilities,
and the cost of covering loans that are in default. In part a of Figure 5.2,
costs, C(L), increase as more loans are made. For example, when LA loans
are made, costs are $150 million. When loans increase to LM, costs increase
to $200 million. Note too that the cost function intersects the vertical axis
at a cost above zero. Even when the bank makes no loans, it must pay certain
costs, such as the costs associated with the building and utilities. These are
called fixed costs, since they are paid at any level of loans, including zero.
Costs that change when the bank changes the quantity of its loans are called
variable costs.
Why is the revenue curve a straight line while the cost curve is not?
Remember that the individual firm in pure competition cannot affect the
market price, so revenue to that firm is simply the interest rate times the
dollar amount of loans, or iL X L. This is a straight line through the origin,
labeled R (for bank revenue), in Figure 5.2. The shape of the cost curve
reflects the fact that as a bank makes more and more loans, costs first rise
at a decreasing rate—up to loan amount LA in the figure—and then begin
rising at an increasing rate. Why? Consider a bank that opens up in a new

building of a given size and has a given number of tellers, loan officers, and
other workers. As the bank begins operations and starts accepting deposits
and making loans, its costs rise somewhat, but due mostly to the cost of
paying interest on deposits. The building and labor costs are being paid
regardless of the number of deposits or loans. Thus, costs rise, but at a
decreasing rate. At some point, however, the number of loans (and deposits)
gets so large that costs start rising at an increasing rate. The number of loan
applications and transactions engaged in to service depositors becomes so
large that it taxes the ability of the bank’s building and labor force to handle
the work. At this point, the cost of making additional loans is not only the
interest on deposits but also the cost of hiring and training more workers
and the cost of expanding the bank building. Thus, the shape of the cost
curve is based on the fact that as a bank makes a larger volume of loans,
eventually the additional resources required to make each additional dollar
of loans will increase, and hence costs will rise at ever increasing rates.
Using Total Revenue and Total Cost t© Determine the Op¬

timal Quantity of Loans. The individual bank’s profits are the dif¬
ference between total revenues and total costs. In part a of Figure 5.2, profits
are given by the vertical distance between the revenue and cost curves.
Notice that when fewer than LA loans are made, the cost curve lies above
the revenue curve, and thus profits are negative (the bank experiences a loss).
For loan amounts between LA and LB, the revenue curve lies above the cost
curve, and the bank earns positive profits. But for loans in excess of LB,
costs again exceed revenues, and losses result at these levels of loans.
Part b of Figure 5.2 graphs the difference between the revenue and cost
curves in part a. This profit curve summarizes the profits the bank could earn
for different quantities of loans. Notice that profits are exactly zero at LA
and Lb. This is consistent with part a, since at these levels of loans costs
exactly equal revenue. Furthermore, profits are maximized at point E, where
the profit function (R — C) is at its highest point. This level of loans, LM,
is the level that maximizes the bank’s profits. Notice in part a that this point
corresponds to the point where the vertical distance between revenue and
costs is the greatest. Moreover, the slope of the cost curve at this level of
loans equals the slope of the revenue curve.
It is no accident that the slope of the revenue curve equals the slope of
the cost curve at the profit-maximizing level of loans. The slope of the cost
curve is called marginal cost (MCL) and reflects the cost to the bank of
loaning out an additional dollar. For example, if the cost to the bank of
making an additional $1 loan is 7 cents, the marginal cost of the loan is .07.
Similarly, the slope of the revenue curve is the interest rate on loans (iL) and
reflects the revenue the bank would generate if it loaned out an additional
dollar. If the interest rate is 8 percent, for example, each dollar in loans
yields the bank 8 cents in interest income.
To maximize profits, a bank in pure competition issues loans at the point

where the marginal cost of a loan equals the interest rate on loans. The
reason is simple: If the marginal cost were less than the interest rate (a point
to the left of LM in part a of Figure 5.2), the bank could add more to revenue
than to cost by loaning out an additional dollar. This is why it is not profitable
for the bank to make fewer than LM loans. Similarly, if marginal cost were
greater than the interest rate, the bank would reduce costs by more than it
lowered revenue if it reduced its loans by $1. Thus, it would not be profitable
for the bank to issue more than LM loans. To summarize, to maximize profits
a bank in pure competition produces the quantity of loans such that the
interest rate on loans equals the marginal cost of loans:
iL — MCl.
This condition simply means that the revenues derived from issuing an
additional $1 in loans (iL) equals the cost to the bank of issuing an additional
$1 in loans (MCL).
Using Marginal Revenue and Marginal Cost to Determine

the Quantity of Loans. Figure 5.3 shows another way to summarize
the profit-maximizing loan decision of an individual bank in pure competi¬
tion. Here the interest rate on loans, iL = .10 or 10%, is determined in the
market and defines the demand curve for loans offered by an individual
bank. The curve labeled MCL represents the marginal cost to the bank of
loaning out an additional dollar. Notice that at point A, marginal cost equals
the interest rate on loans, which is the condition for maximizing profits. The
level of loans that corresponds to this point is LM = $100 million, so the
purely competitive bank maximizes profits by issuing LM = $100 million
in loans. The bank issues the loans at the market interest rate, iL = 10%.
The curve labeled ACL in Figure 5.3 represents the average cost to the
individual bank of issuing loans. The average cost of issuing L loans is
defined as the ratio of costs to the total dollar value of loans issued:
C(L)
ACL =
L
For example, if the cost to a bank of issuing $100 million in loans is $5
million, the average cost of each $1 in loans is ACL = $5,000,000/
$100,000,000 = 5%. Notice that the average cost of loans represents costs
as a fraction of the dollar value of loans rather than a dollar amount. Like
the interest rate, average cost is measured as a percentage of the dollar value
of loans.
We can use Figure 5.3 to determine the profits of the individual bank.
To maximize profits, the bank issues LM — $100 million in loans. The
distance between points A and B reflects the spread between the interest rate
received on loans (10 percent) and the average cost to the bank of issuing
Lm = $100 million in loans (5 percent). In this example, the spread is . 1 —

.05 = 5%. Thus, net of costs, the bank earns an average of 5 percent on the
$100 million in loans it issues. The bank’s profits when it produces the
profit-maximizing level of loans is
.05 X $100,000,000 = $5,000,000.
This corresponds to the shaded area of the rectangle in Figure 5.3. It is the
base ($100 million) times the height (.05), or $5 million.
Summary Suppose the market demand for bank loans increases. Use a graph to illus¬
Exercise 5.1 trate what happens to the number of loans and the profits of an individual
bank that operates in a purely competitive banking market.
Answer: In part a, the initial market equilibrium is at point A, where the

market interest rate is i°L. Given this interest rate, the individual bank maxi¬
mizes profits by issuing L0 loans in part b to earn profits represented by the
lightly shaded region. An increase in the demand for bank loans shifts the
market demand for bank loans to the right from D° to Dl in part a. This
increases the market interest rate on bank loans from i°L to ilL. The individual
bank can now issue loans at a higher interest rate, which effectively shifts
the demand for the bank’s loans upward from DF to DlF in part b. To
maximize profits, the individual bank increases its loans from L0 to Lx and
earns higher profits corresponding to the darker shaded region in part b.
Loan Interest Rate Loan Interest Rate
Market Dollar Amount of Loans Individual Bank's Dollar Amount of Loans

(a) (b)
Banks with Market Power
In contrast to purely competitive banks, banks with market power have some
control over the interest rate they charge for loans. The interest rate charged
on loans by a bank with market power is not market determined but depends
on the quantity of loans the bank chooses to issue. A bank with market
power faces a downward-sloping demand curve, such as the demand curve
for loans issued by Bank One in Figure 5.4. This downward-sloping demand
curve indicates that some borrowers will obtain a loan from the bank even
if the rate it charges is higher than the rates charged by other lenders of
funds. For example, if Bank One charges an interest rate of 7 percent on
loans, it will be able to issue $150 million in loans. If it raises the interest
rate to 10 percent, it will issue fewer loans—$100 million—as some borrow¬
ers either decide not to borrow or choose to borrow from another bank.
Thus, to issue more loans, a bank with market power must lower the interest
rate it charges on loans.
Sources of Market Power for Banks

Before we examine how a bank with market power chooses which interest
rate to charge for a loan, it is useful to briefly discuss factors that give rise
to market power. A complete description of sources of market power is far
beyond the scope of this text. Here we will review two sources of market
power that are important for understanding why some banks enjoy market
power: economies of scale and location.
Economies of Scale and Monopoly Banks. A monopoly bank

is a single bank that effectively services the entire market for loans. Many
small towns have a single bank, and the transactions costs of obtaining a
loan from a bank located in some distant location make it impractical for
households to obtain loans elsewhere. A natural question, however, is why
small towns tend to have a single bank. The answer lies in what economists
refer to as economies of scale: Larger banks can provide loans at lower
average cost than smaller banks can. Economies of scale exist whenever the
average cost curve decreases as the quantity of loans increases.
Consider the average cost curve for a bank illustrated in Figure 5.5,
which exhibits economies of scale. Suppose borrowers desire $200 million
in loanable funds. If a single bank provided these funds, the average cost to
the bank of providing the loans would be 8 percent. (Remember, AC is not
expressed as a dollar amount). To stay in business, this bank would have to
receive an interest rate on loans of at least 8 percent to cover the costs of
providing loans. In contrast, if two banks shared the market and each pro-
vided half of the $200 million in loans, the average cost for each bank of
providing $100 million in loans would be 15 percent, because neither bank
would be able to take advantage of the economies of scale reflected in the
shape of the average cost curve.
What does all this imply about the number of banks in this hypothetical
small town? If there were two banks, each charging 15 percent for loans,
the banks would have an incentive to merge into a single bank (which would
lower the average cost of loans). Alternatively, one of the banks could lower
its interest rate on loans to attract customers from the other bank and drive
its competitor out of business thanks to its lower average cost of servicing
$200 million in loans. In short, when sufficient economies of scale exist, a
single bank will dominate the market for loans. A monopoly bank faces a
downward-sloping demand curve for loans to borrowers located in its vicin¬
ity and thus can issue more loans only if it lowers the interest rate. Box 5.1
discusses one study that looks at economies of scale in banking.
Location Advantages. Market power does not necessarily imply that

a single bank services the entire market. In many instances, borrowers value
the convenience of using a bank that is close to their homes or offices. If a
If the average cost of Figure 5.5

making loans exhibits Economies of Scale
economies of scale, a
single bank can serve
the market at a lower
cost than can several
banks serving the
same market. When
$200 million in loans
is demanded this mar¬
ket, two banks sharing
the market must
charge 15 percent for
each loan to cover
costs, while one bank
can provide the same
number of loans at
only 8 percent.
Dollar Value of Loans

(Millions $)
Economies of Scale in Banking
Economies of scale are one reason some indus¬ with $25 million in assets, average cost is 10
tries are dominated by a few large firms. Do cents per $1 of assets. Thus, as banks grow
firms in the U.S. banking industry enjoy econo¬ from $1 million to $25 million in assets, average
mies of scale? It turns out the evidence is rather cost per dollar of assets falls by about one-
mixed, but the main conclusion is that the aver¬ half of 1 percent. As bank size grows from $25
age cost curves of banking firms have a flat¬ million to $300 million, average cost fluctuates
tened U shape, with slight economies of scale at in the range of 10 cents per $1 of deposits,
small sizes and slight diseconomies of scale (an suggesting a flat average cost curve over this
upward-sloping average cost curve) at large range. But as bank size grows to $5 billion, av¬
sizes. The accompanying figure illustrates this erage cost rises to about 10.3 cents per $1 of
phenomenon. assets.
Empirical data for 1984 suggest that for During 1984, the prime interest rate was
banks with $1 million in assets, average cost is about 12 percent. Thus, banks received an inter-
about 10.5 cents per $1 of assets. For banks Continued on p. 143
ATC
25 300
Bank Size (Millions $)
Continued from p. 142 loan interest rates are relatively close to the av¬
erage cost of funds to banks.
est rate that was between 1.7 and 2 percent
above average cost, depending on the econo¬
Sources: David B. Humphrey, "Why Do Estimates of Bank
mies of scale they enjoyed. The conclusion is
Scale Economies Differ?" Federal Reserve Bank of Rich¬
that economies of scale in the U.S. banking in¬ mond Economic Review (September-October 1990), 38-
dustry are present but fairly modest and that 50, Economic Report of the President, January 1993, 428.
bank across town offers a better deal on banking services, some borrowers
will still choose to use the more convenient bank. Thus, by raising the interest
rate on loans, a bank with a location advantage will lose some loan custom¬
ers, but not all of them.
Of course, given modern technology such as the modem and the fax
machine, the location advantage may be somewhat diminished. It is still
important, however, if for no other reason than that bank customers still find
it convenient to visit their local bank for a number of services. While pos¬
sible, it is usually very inconvenient for a borrower to arrange a loan without
visiting his or her bank. Perhaps it is possible to open accounts by mail, but
this too is inconvenient. Ask yourself whether you would like to bank with
First National Bank of Fairbanks, Alaska. The answer should convince you
that location advantage is still important.
Profit-Maximizing Loan Decisions

for Banks with Market Power
We can now examine how a bank with market power determines its profit-
maximizing interest rate and level of loans. These decisions are a bit more
complicated than those of purely competitive banks, because the interest rate
a bank with market power receives on its loans depends on the number of
loans it issues.
Consider the demand curve for loans at Bank One in part a of Figure
5.6. If the bank charges an interest rate of 30 percent, the quantity demanded
of its loans is zero, so its revenues (graphed as a function of loans in part b)
will also be zero, as shown by point A. On the other hand, part a reveals
that if the bank lowers the interest rate on loans to 0 percent, it can issue
$100 million in loans. But since nothing times something is nothing, its
revenues will be zero, corresponding to point B in part b of Figure 5.6. For
interest rates between zero and 30 percent, however, the bank is able to issue
varying amounts in loans. The revenue function in part b of Figure 5.6
represents the revenues associated with these different levels of loans be¬
tween $0 and $100 million. For example, if the bank charges an interest rate
Part a illustrates the Figure 5.6

demand curve for Demand and Revenues of a Bank with Market Power
loans issued by Bank
One. If Bank One
charges an interest
rate of 30 percent, no
one will demand any
loans. If it charges no
interest for its loans,
$100 million worth of
loans will be de¬
manded. Part b shows
Bank One's revenue
curve, which depends
on the quantity of
loans issued. An inter¬
est rate of 30 percent
corresponds to point
A in part b; an interest
rate of zero corre¬
sponds to point B. As
the interest rate in¬
creases from zero to
30 percent, revenues
initially rise and then
begin to decline.
Loans at Bank One (Millions $)
of 20 percent, it can provide $10 million in loans and earn revenues of .20
X $10,000,000 = $2,000,000.
To determine the level of loans that maximizes Bank One’s profits, we
superimpose the revenue curve in part b of Figure 5.6 on the bank’s cost
curve to obtain the graph in part a of Figure 5.7. Notice that costs exceed
revenues for points to the left of A or to the right of B, and the bank makes
a loss. Part b of Figure 5.7 depicts these losses by graphing profits as a
function of loans. At points A and B, revenues from loans exactly equal the
bank’s costs and profits are zero, which is consistent with the profit curve
graphed in part b.
For loans between points A and B, revenues exceed costs and the bank
earns profits. Notice that the vertical distance between revenues and costs is
greatest at $40 million in loans. At this point, revenues are $3 million and
costs are only $1 million. Profits are thus $2 million, which corresponds to
point C in part b of Figure 5.7, where the profit curve achieves its maximum.
Notice that at the point of profit maximization, the slope of the bank’s
revenue curve (denoted MRL in part a) equals the slope of the bank’s cost
curve (denoted MCL). The slope of the revenue curve, MRL, is the marginal
revenue to the bank of issuing additional loans. For example, if by issuing
an additional $1 in loans the bank increases its revenue by 10 cents, the

marginal revenue of additional loans is .10. Similarly, marginal cost (MCL)
reflects the cost to the bank of issuing an additional $1 in loans. At the point
of profit maximization, MRL = MCL.
To understand why MRL = MCL at the point of profit maximization,
suppose the bank issues fewer than $40 million in loans. This corresponds
to a point where the slope of the revenue function (MRL) is greater than the
slope of the cost function (MCL). If MRL > MQ, the bank will be able to
add more to revenue than to cost by offering more loans. As the bank expands
loans up to $40 million, MRL will equal MCL. Would the bank want to
continue issuing loans beyond $40 million? The answer is no. For loans in
excess of $40 million, MRL is less than MCL: Issuing loans in excess of $40
million would increase costs more than it would increase revenues. In short,
to maximize profits, a bank with market power issues loans such that mar¬
ginal revenue equals marginal cost.
Figure 5.8 shows an alternative way to illustrate the profit-maximizing
loan decision of a bank with market power. The demand curve for the bank’s
loans is given by D and is downward sloping since the bank has market
power. The marginal revenue associated with loans is denoted MRL and lies
Oligopoly Banks 147
below the demand curve. Notice that marginal revenue equals marginal cost
at point A. Thus, the profit-maximizing level of loans for this bank is LM.
The interest rate a bank with market power will charge for a loan is the
maximum rate borrowers will pay for the profit-maximizing level of loans.
This interest rate corresponds with point B on the demand curve, so the
profit-maximizing interest rate charged by the bank is i\f. Notice that the
bank charges an interest rate that exceeds its marginal cost of issuing loans,
reflecting the bank’s market power.
To determine bank profits, notice that the vertical distance BC in Figure
5.8 reflects the spread between the interest rate received on each loan and
the average cost of issuing loans. Multiplying this spread by the total value
of loans (Lm) determines the profits of the bank with market power. The
shaded region in Figure 5.8 shows these profits.
Summary 1 Suppose a bank with market power has a constant marginal cost of issuing
Exercise 5.2 loans and that marginal cost equals the average cost at each level of output;
that is, marginal and average costs are equal at every level of loans and also
neither increase nor decrease as the number of loans expands, (a) Graph the
bank’s marginal and average cost curves, (b) Show the profit-maximizing
value of loans, the interest rate on loans, and the bank’s profits.
Answer: (a) The bank’s marginal and average costs are a horizontal line,
such as the one corresponding to MCL = ACL. (b) The profit-maximizing
level of loans is that level where MRL = MCL. This corresponds to loans
of L°. The bank charges the maximum interest rate consistent with this level
of loans, which is i{[. The bank’s profits are given by the shaded region.
Oligopoly Banks
Oligopoly is a situation in which only a few firms in a market produce a

good or service. Thus, an oligopoly bank competes for customers with only
148 Chapter5 The Bank as a Firm: Loans
a few other banks. In this section, we focus on the case of duopoly banks:
two banks that compete with each other for loan customers. (The basic
principles apply even if there are, say, three to five banks in the relevant
market.) The distinguishing feature of an oligopolistic market for banks is
the high degree of interdependence among banks. One way to measure the
degree of interdependence is to look at concentration ratios, as we discuss
in Box 5.2.
The Nature of Oligopoly Interdependence

Let us first examine the nature of oligopolistic interdependence. Consider
two banks that are located on the same block in a town, so neither bank has
a location advantage over the other. In this instance, actions by one bank
will have a dramatic impact on the profits of the other bank. For example,
if one bank lowers its interest rate on loans below that of the other bank,
shoppers for loans will switch to the bank offering the lower interest rate.
This will drastically reduce the profits of the bank offering the higher interest
rate. This effect characterizes oligopolistic interdependence.
Using Game Theory to Model

Oligopolistic interdependence
The interdependence of oligopolistic banks means the actions these banks
take contain important strategic elements. Economists use game theory to
analyze these strategic interactions, and we will use a simple game to illus¬
trate these interactions between two banks. Table 5.1 depicts this oligopo¬
listic interdependence between two banks, Bank One and Bank Two. For
simplicity, we assume each bank has a choice between two interest rates on
loans: a high interest rate of 15 percent and a low interest rate of 8 percent.
The first entry of each cell of the matrix in Table 5.1 corresponds to the
profits of Bank One, and the second entry corresponds to the profits of Bank
Two. Looking at the table, we see that if both banks charge an interest rate
of 8 percent, each bank earns zero profits. If Bank One charges 8 percent
Table 5.1
A Hypothetical Payoff Matrix
Bank Two
Interest Rate 8% 15%

o
oH
8% $0, $0
1
Bank One
15% — $10. $40 $10, $10
Oligopoly Banks 149
Concentration Ratios in Banking
One important determinant of market structure called concentration ratios. In the case of bank¬
is the number of firms. For a monopoly, obvi¬ ing, we might ask what percentage of banking
ously, there is a single firm, while an oligopoly industry assets are held by the four or eight larg¬
consists of several firms and a purely competi¬ est firms. These data are called the four-firm
tive market contains very many firms. To obtain and eight-firm concentration ratios.
a gauge of how concentrated a particular indus¬ An international comparison of concentra¬
try is, economists use a measure of the share of tion ratios reveals an interesting feature of the
output or revenue belonging to the largest four U.S. banking industry relative to other countries:
or eight firms in the market. These statistics are Continued on p. 150
U.S. Japan U.K. France Germany

Total assets of
four largest banks1
576.926 1,717.836 706.116 1,113.228 792.136
(billions of
U.S. dollars)
Total assets of
eight largest banks1
916.258 3,157.847 947.335 1,649.382 1,316.312
(billions of
U.S. dollars)
Total assets of
banking industry2
3,237.600 5,545.452 2,210.785 1,841.464 2,572.393
(billions of
U.S. dollars)
Concentration
ratio of four 17.82% 30.97% 38.35% 60.45% 30.79%
largest banks
Concentration
ratio of eight 28.30% 56.94% 42.85% 89.57% 51.17%
largest banks
]The Banker, July 1992.

2lnternational Financial Statistics, February, 1993.
Continued from p. 149 17.82 percent of American banking assets. In

In the United States, banking is less concen¬ contrast, the four largest banks in France hold
trated than in almost all other countries. The ac¬ 60.45 percent of French banking assets. This il¬
companying table reports the total assets held lustrates the general feature of the U.S. banking
by the four and eight largest banks as well as system: It is relatively unconcentrated by interna¬
the entire banking industry. Data are for the tional standards. One reason for this is the leg¬
United States, Japan, the United Kingdom, acy of laws prohibiting interstate banking in the
France, and Germany, and all are converted into United States and some state laws prohibiting
billions of U.S. dollars. branch banking. These regulations severely limit
Notice that the four- and 8-firm concentra¬ concentration in the banking industry.
tion ratios are lower for the United States than
those for any of the other four countries. For in¬ Sources: The Banker, July 1992; International Financial
stance, the four largest U.S. banks account for Statistics, February 1993.
and Bank Two charges 15 percent, Bank One earns profits of $40, while
Bank Two suffers losses of $10. Put simply, at the lower interest rate, Bank
One steals Bank Two’s customers and earns large profits at its rivals’ ex¬
pense. Likewise, if Bank One charges the high interest rate of 15 percent
when Bank Two charges the low rate of 8 percent, Bank Two steals Bank
One’s customers. In this instance, Bank One loses $10, while Bank Two
earns $40 in profits. But if both banks charge the high interest rate of 15
percent, each bank earns profits of $10 since they share the market equally.
Table 5.1 thus reveals that the profits Bank One earns depend not only
on the interest rate it charges but also on the interest rate the rival bank
charges. Given this situation, what interest rate would each bank choose to
charge? Your initial answer might be 15 percent, since if each bank charged
this rate, profits would be $10 for each bank. However, note that if Bank
Two did charge an interest rate of 15 percent, Bank One could increase its
profits from $10 to $40 by lowering its rate to 8 percent.
To determine the equilibrium outcome for this oligopoly, consider Table
5.1 from Bank One’s perspective. If Bank Two charged 8 percent, Bank
One would maximize profits by charging 8 percent, since the corresponding
profits of zero are preferable to the loss of $10 it would incur by charging
the high rate against the competitor’s low rate. If Bank Two charged 15
percent, Bank One would still be better off charging 8 percent, since the
profits of $40 are higher than the profits of $10 it would earn by charging
the higher interest rate. In short, regardless of the rate Bank Two charges,
in this example Bank One is always better off charging the lower interest
rate. Notice that because the payoff matrix is symmetric, the same is true
from the viewpoint of Bank Two. Thus, the equilibrium outcome for this
oligopoly market is for each bank to charge the low interest rate to earn
profits of zero.
Oligopoly Banks 151
This result may surprise you. Each bank finds it in its interest to charge
a low interest rate, even though both banks would benefit if they agreed to
each charge a high rate. Why wouldn’t the banks conspire to charge a high
rate? Conspiring to charge a high interest rate is an example of collusion:
acting in concert to make both banks better off. In the United States, collusion
is illegal; banks cannot conspire to set high interest rates. There are other
reasons, however. Suppose the presidents of the banks secretly met and
agreed to charge high interest rates. Would they have an incentive to live up
to their promise? Consider Bank One’s point of view. If it cheated on the
collusive agreement by lowering its interest rate, it would increase its profits
from $10 to $40. Thus, Bank One has an incentive to induce Bank Two to
charge a high rate so that Bank One can cheat to earn higher profits. Of
course, Bank Two recognizes this incentive, which reduces the likelihood
that the agreement will be made in the first place.
Suppose, however, that the president of Bank One is honest and would
never cheat on a promise to charge a high interest rate. (She is honest enough
to keep her word to the other president, but not so honest as to obey the law
against collusion.) What happens to Bank One if the president of Bank Two
cheats on the collusive agreement? Bank One experiences losses of $10
when Bank Two cheats. When the stockholders ask the president of Bank
One why they lost $10 when the rival bank earned profits of $40, how can
the president answer? She cannot admit she was cheated on in a collusive
agreement, for doing so would send her to jail for violating the law. Whatever
her answer, she risks being fired or sent to jail, and this reduces her incentive
to enter into a collusive agreement in the first place. Thus, the fact that only
two banks service a market need not imply the banks have market power.
Repeated interaction
Our description of the duopoly market for loans illustrates the difficulty
banks have in reaching collusive agreements. An important feature of the
analysis, however, is that a bank incurs no cost if it cheats on the collusive
agreement; it receives only benefits. Technically, the reason this occurred in
the previous section is that we assumed the underlying competition for
customers was a “one-shot” (or one-time) interaction; that is, the banks
faced the payoff matrix presented in Table 5.1 only once.
In reality, of course, banks compete for customers day after day and
year after year. When banks repeatedly face a payoff matrix such as the one
in Table 5.1, they can in some cases collude without fear of being cheated
on. To see this, suppose Bank One and Bank Two secretly met and agreed
to the following: “Let’s each charge the high interest rate, provided neither
of us has ever cheated before. If one of us cheats and charges the low interest
rate, let’s charge the low interest rate in every period thereafter.” It turns
out that if both banks agree to behave this way, conditions exist under which
neither has an incentive to cheat. Before we look at this formally, let us
examine the basic intuition. Under this agreement, a bank that cheats earns
an immediate profit of $40 instead of $10. Thus, there is still the immediate
benefit to a bank of cheating on the agreement. However, because the banks
compete repeatedly over time, cheating has a future cost: The agreement
stipulates that if either bank ever cheats, each bank will charge a low interest
rate in all future periods. Thus, the bank that cheats will earn $0 instead of
$10 in these periods. Hence, the benefit of cheating today on the collusive
agreement is earning $40 instead of $10 today, but the cost is earning $0
instead of $10 in each future period. If the present value of the costs of
cheating exceed the one-time benefit of cheating, it does not pay for a bank
to cheat, and high interest rates can be sustained.
Now let us formalize this idea. Suppose the banks agreed to the collusive
plan outlined earlier, and Bank One believes Bank Two will live up to the
agreement. Does Bank One have an incentive to live up to the agreement?
If Bank One cheats, its profits will be $40 today but $0 in all subsequent
periods, since cheating today leads Bank Two to charge a low interest rate
forever after. The best Bank One can do when Bank Two charges the low
rate in these periods is earn $0. Thus, if Bank One cheats, the present value
of its profits will be
Profits if it does cheat = $40 + 0 + 0 + 0 + 0 + ...
If Bank One does not cheat, it will earn $10 each period forever. Thus, the
present value of the profits of Bank One if it does not cheat will be
10 10 10
Profits if it does not cheat = 10 H-1-~ H-r + • • •
1 + r (1 + rf (1 + r)3
10(1 + r)
r
where r is the real interest rate the bank uses in its present value calculations.* 1
Bank One has no incentive to cheat if its earnings from cheating will be less
than its earnings when it does not cheat. In this example, there is no incentive
to cheat if
$10(1 + r)
Profits if it does cheat = $40 < Profits if it does not cheat,
r
which is true if r < 1/3. In other words, if the real interest rate Bank One
uses in its present value calculation is less than 33 percent (or 1/3), it will
lose more (in present value) by cheating than it will gain. The same is true
1 The above calculation uses a mathematical result from summing an infinite series. That result is
1 + X + x2 + x3 + . . . + x' + ... — X,- = o° X1 — 1/(1 — x)
for any fraction x that is greater than — 1 and less than 1. In the present value calculation, x is
1/(1 + r). Substituting this for x, we find that the infinite sum is (1 + r)/r.
for Bank Two. Thus, since oligopoly banks compete repeatedly over time,
they can collude and charge high loan interest rates to earn $10 each period.
This benefits banks at the expense of borrowers, which explains why there
are laws against collusion.
Summary Suppose Bank One and Bank Two repeatedly face the situation presented in
Exercise 5.3 Table 5.1 and that the interest rate they use in their present value calculations
is 40 percent, (a) What are Bank One’s profits if it cheats on the collusive
agreement described earlier in the text? (b) What are Bank One’s profits if
it does not cheat on the collusive agreement? (c) Will both banks charge the
high interest rate each period?
Answer: (a) If Bank Two lives up to the collusive agreement, Bank One
will earn $40 today if it cheats and $0 forever after, (b) If Bank Two lives
up to the collusive agreement and Bank One does not cheat, the present
value of Bank One’s profits will be
13
$10 $10 $10 $10(1 + .4)
$10 + + • • • $35.
1 + .4 + (1 + A)2 + (1 + A)3 .4
(c) Since $40 $35, the present value of Bank One s profits is higher if it
cheats than if it does not cheat on the collusive agreement. Since the situation
is symmetric, each bank has an incentive to cheat on the agreement, even if
it believes the other bank will not cheat. In equilibrium, each bank will
charge the low interest rate each period to earn profits of $0 each period.
Banks with Imperfect Information

Thus far we have ignored an important consideration by individual banks in
making loan decisions: We have assumed all borrowers are equally likely to
repay their loans. In reality, some borrowers fail to repay loans, and this
reduces bank profits. Of course, if a bank knows a particular borrower will
not repay the loan, it will not agree to lend that person money in the first
place. But banks suffer from imperfect information: They do not know for
certain which borrowers will and which will not repay their loans. In this
section, we analyze the impact of imperfect information on banks’ loan
decisions.
Symmetric Information
First, we assume there is symmetric information; that is, borrowers and
banks have the same information about whether a loan will be repaid. Sup¬
pose the likelihood that a borrower with a low income is able to repay a
loan is 10 percent, while the probability that a borrower with a high income
will be able to repay a loan is 90 percent. Thus, on average only 1 out of

every 10 low-income borrowers will repay loans, while 9 out of every 10
high-income borrowers will repay loans. In this case, symmetric information
means that when a low-income borrower applies for a loan, both the bank
and the borrower both know there is only a 10 percent chance the loan will
be repaid. Similarly, when a high-income borrower applies for a loan, both
the bank and the borrower know there is a 90 percent chance the borrower
will repay the loan.
Therefore, when symmetric information exists, the expected return to
the bank of lending to the high-income borrower at a given interest rate is
greater than that of lending money to a low-income borrower. To be willing
to lend money to a low-income borrower, the bank must receive a higher
interest rate to compensate for the additional default risk associated with the
loan. In short, the bank discriminates against low-income borrowers by
requiring them to pay a higher interest rate for loans than high-income, more
creditworthy borrowers pay. Individuals in each risk class pay an interest
rate that fully compensates the bank for the riskiness of their own loans.
Better credit risks do not subsidize poor credit risks when symmetric infor¬
mation exists.
Figure 5.9 illustrates why a bank with market power will charge a lower
loan interest rate to borrowers who are good credit risks (part a) and a higher
interest rate to borrowers who are bad credit risks (part b). For simplicity,
we assume there is a constant marginal cost of providing loans. However,
one component of the marginal cost of issuing loans is the likelihood of
default, since defaults increase the cost to the bank of issuing loans. The
marginal cost of issuing loans to good credit risks is denoted MCG in part
a, while the marginal cost of issuing loans to a bad credit risk is denoted
MCS in part b. Notice that MCfi > MCG, reflecting the higher marginal cost
of issuing more risky loans. The marginal revenue of issuing a loan to a
good credit risk equals the corresponding marginal cost at point A in part a
of Figure 5.9. Thus, LG loans are issued to borrowers who are good credit
risks at an interest rate of iG = 8%. Similarly, marginal revenue equals
marginal cost for borrowers who are bad credit risks at point B in part b.
Thus, the bank issues LB loans to borrowers who are bad credit risks at an
interest rate of iB — 15%. Since iB > iG, individuals who are bad credit risks
obtain funds, but at a higher interest rate than do creditworthy borrowers.
Asymmetric Information and Adverse Selection

We now consider asymmetric information, the situation in which borrowers
have better information about their ability to repay a loan than the bank
does. To be concrete, suppose there are two types of borrowers, honest and
dishonest. Honest and dishonest borrowers are identical in every observable
respect (they have the same income, etc.) but differ in character, which is
unobservable. For simplicity, assume honest borrowers repay loans 90 per-
cent of the time, whereas dishonest borrowers repay only 10 percent of the
time. Asymmetric information arises because borrowers know how honest
they are, but banks do not.
Because a bank cannot distinguish between honest and dishonest bor¬
rowers, it must charge the same interest rate to both types. This interest rate
will be an average of the interest rates it would charge to each type of
borrower if symmetric information existed. Asymmetric information creates
a higher interest rate than the bank would charge to honest borrowers if there
were symmetric information, but a lower rate than it would charge to dis¬
honest borrowers if symmetric information were available. In short, honest
borrowers pay a higher interest rate to compensate for the fact that dishonest
borrowers default most of the time. Honest borrowers thus subsidize dis¬
honest borrowers.
Unfortunately, this is not the end of the story. As the interest rate rises
above the rate honest borrowers would have to pay in the presence of
symmetric information, some honest borrowers decide not to borrow, and
the quantity demanded of loans by honest borrowers falls. This increases the
proportion of loans issued to dishonest borrowers, thus increasing the number
of defaults as a fraction of all bank loans. As defaults increase, the bank
further raises the interest rate to offset the higher marginal cost of issuing
loans. Because of the higher interest rate, even fewer honest borrowers seek
loans. Ultimately, by continuing to increase the interest rate, the bank ends
up in a situation where it issues loans only to dishonest borrowers (who have
no intention of repaying the loans in the first place), driving honest borrowers
out of the market.2
This phenomenon is known as adverse selection: As the interest rate
rises, honest borrowers decide not to borrow, and the bank is left with an
adverse pool of borrowers—those who know they are more likely to default.
When asymmetric information exists, an increase in the loan interest rate
will raise the fraction of loans that will not be repaid.3
Adverse selection does not occur only because of asymmetric informa¬
tion about the honesty of borrowers. It also occurs because borrowers have
inside information about the riskiness of the projects they are borrowing to
finance. For example, suppose you have decided to open a small restaurant.
Your aunt Alice has agreed to be your chef, and she has a great collection
of Italian recipes. The bank knows most new restaurants fail but is convinced
that your restaurant has a reasonable chance to succeed. It charges you a
relatively high interest rate but agrees to make the loan. What the bank
doesn’t know, but you do, is that Aunt Alice has a weak heart and may not
be able to work the long hours it will take to make the restaurant a success.
Furthermore, you are not sure whether anyone else can replace Aunt Alice
as chef. In this case, you have more information than the bank does. You
know the restaurant is actually a riskier venture than the bank thinks because
of the risk to your aunt’s health.
Figure 5.10 illustrates the impact of adverse selection on the loan interest
rate for a bank with market power. Since the bank cannot distinguish between
good and bad credit risks, there is a single demand curve composed of good
and bad credit risks alike. This demand curve is labeled DG + DB, signifying
that it is the total demand by both types of borrowers. The corresponding
marginal revenue curve is denoted simply as MRL. Let MQv denote the
marginal cost of issuing loans in the absence of any defaults. If borrowers
never defaulted, banks would choose to make a quantity of loans at point A,
where MCN intersects MRL. We see that the bank would issue $200 million
2 This analysis was first applied to the market for used cars by George Akerlof in “The Market
for ‘Lemons’: Quality Uncertainty and the Market Mechanism,’’ Quarterly Journal of Economics,
August 1970, pp. 488-500.
3 In the next chapter, we examine the related phenomenon of moral hazard.
Figure 5.10
Adverse Selection and Loans at a Bank with Market Power
When borrowers have better information about their status as credit risks than the bank
does, the bank will face a single demand curve, D = DG + De, composed of both honest
and dishonest borrowers. In the absence of any defaults, the marginal cost curve would be
MCW. As dishonest borrowers default, the marginal cost increases to MCA This lowers the
quantity of loans from $200 million to $150 million, and the interest rate goes up from 6
to 13 percent. In effect, honest borrowers must pay higher interest rates because of the
behavior of dishonest borrowers. Since dishonest borrowers do not plan to pay back loans,
they continue to seek loans even at the higher interest rate. Honest borrowers, on the
other hand, are less likely to borrow at higher interest rates. Therefore, the proportion of
dishonest borrowers increases, leading to an increase in defaults and further raising the
marginal cost of making loans to MCT The result is an even higher rate of 13 percent.
This phenomenon, known as adverse selection, leaves the bank with a pool of borrowers
more likely to default.
in loans and, with this quantity, the demand curve tells us the interest rate
would be 6 percent.
Next, suppose dishonest borrowers default, but asymmetric information
precludes the bank from distinguishing dishonest from honest borrowers. An
increase in defaults raises the marginal cost of loans, since some loans now
go unpaid. This higher marginal cost is represented in Figure 5.10 as MC^,
which lies above the marginal cost curve when there are no defaults (MQv).
This increase in the marginal cost of issuing loans results in MRL = MC^
at point B. In turn, this increases the interest rate from 6 to 13 percent,
resulting in a reduction in loans from $200 million to $150 million. Unfor¬
tunately, however, the reduction in quantity demanded consists solely of
honest borrowers; dishonest borrowers have no intention of repaying loans
and could care less what rate the bank charged. Thus, the fraction of loans
issued to dishonest borrowers increases, leading to a further increase in
marginal cost to MQy This new marginal cost equals marginal revenue at
point C, resulting in a rise in the interest rate to 18 percent. In the presence
of asymmetric information, increases in the interest rate can ultimately lead
to a situation where the only individuals willing to pay the high rate are
those who know they will default. In the absence of a mechanism for alle¬
viating the problems generated by asymmetric information, banks would
ultimately refuse to issue any loans at all.
Bank Strategies for Countering Asymmetric Information

We have seen the difference between symmetric imperfect information (im¬
perfect because there is still uncertainty about whether a loan will be repaid,
but this uncertainty is known by both borrower and lender) and asymmetric
information (where borrowers have better information than lenders about
ability or willingness to repay). We have also seen how information asym¬
metries can lead to problems of adverse selection. Yet, even though consid¬
erable information asymmetries exist in the banking industry, we all know
that banks continue to make loans. In fact, as we saw in Chapter 1, loans
are the engine that drives money creation in the economy. Because of this
crucial role of loans, banks have developed several effective mechanisms to
overcome default risks and other information asymmetries. We discuss these
strategies in the next section.
Credit Reports. Banks rely heavily on credit reports for information

about loan applicants’ credit histories. By examining the past credit history
of a potential borrower, a bank reduces the level of asymmetric information
about that person. In effect, the bank can infer the probability that a potential
borrower will (or will not) default on a new loan by examining the frequency
with which that borrower has defaulted in the past. If the credit report is
sufficiently accurate, symmetric information between the bank and the po¬
tential borrower exists. In this case, the bank can set an interest rate for the
borrower that is consistent with his or her likelihood of default and charge
individuals with poor credit reports higher interest rates than those with good
reports.
Unfortunately, this is not always easy to do. Credit reports pose two
problems. First, if they are inaccurate or incomplete, some asymmetric in¬
formation will remain which will still lead to an adverse selection problem.
Second, some potential borrowers have no credit history because they have
never borrowed funds before. In this case, credit reports tell the bank nothing
and therefore do not help reduce asymmetric information. Some first-time
borrowers may develop poor credit histories and others impeccable ones.
The bank has no way of knowing which will be true for a given first-time
borrower and thus needs some alternative strategy to deal with these cases.
Reputation. In the absence of any credit history, banks need some other
method to deal with the problem of asymmetric information. Many banks
attempt to build a reputation for being tough on borrowers who default.
Toughness might include foreclosing on the assets purchased with the loan
money or seeking legal action to receive payment for the funds in default.
In historical periods, being tough even included sending the defaulter to
debtor’s prison.
Is it “ethical” to be tough on those who default? Your answer might
depend on whether you think the person who defaulted was dishonest (he
or she was able to pay but chose not to) or was truly unable to pay back the
loan. From the viewpoint of the bank, however, it is not always clear whether
a defaulter is honest or dishonest. Dishonest defaulters can easily hide their
assets to make themselves appear penniless. Asymmetric information makes
it virtually impossible for a bank to determine the reason for default.
Given asymmetric information about the reason for default, suppose a
bank adopts a policy of being relatively lenient on those who default. This,
of course, will lead to an adverse selection problem. The policy will attract
dishonest borrowers in droves, ultimately driving interest rates on loans so
high that only dishonest borrowers will choose to borrow money from the
bank.
In contrast, consider a bank that adopts a policy of always being tough
on defaulters. This increases the cost to dishonest borrowers of doing busi¬
ness at the bank. Indeed, if the bank is tough enough, dishonest borrowers
will either borrow funds from a 4 ‘kinder, gentler’ ’ bank, do without borrowed
funds, or actually behave honestly (i.e., repay loans). By investing in a
reputation for being tough on defaulters, a bank can reduce the negative
impact of asymmetric information. The bank benefits by having fewer de¬
faults and, since a lower number of defaults reduces the loan interest rate,
those who borrow from the bank benefit as well.
By adopting a policy of toughness, the bank rids itself of borrowers who
default for reasons of dishonesty. Since anyone who defaults thereafter must
really be unable to pay back the loan, why not let that person off easy?
Unfortunately, if the bank started letting defaulters off easy, dishonest bor¬
rowers would again have an incentive to default. In the presence of asym¬
metric information, the only way the bank can avoid dishonest defaults is to
be tough on everyone who defaults. In short, banks are tough on defaulters
not because the banks are mean but because if they failed to do so, there
160 Chapter 5 The Bank as a Firm.-Loans
would soon be no money available for honest borrowers due to adverse

selection. Honest defaulters are, in effect, the sacrificial lamb that reduces
the adverse selection problem, thus allowing all borrowers to obtain funds
at a relatively low interest rate.
Collateral. Many banks require borrowers to put up collateral to obtain

a loan. Collateral is property or other assets pledged as security against
default on a loan. If default occurs, the lender gets the collateral. Collateral
is, in essence, a “hostage” the bank uses to induce the borrower to repay
the loan. If the borrower does not repay the loan, the bank keeps the hostage;
if the borrower does repay, the bank releases the hostage. New-car loans and
mortgages typically use the underlying asset purchased with the loan money
as collateral. If the borrower defaults on the loan, the bank can seize the car
or the house and sell it to recoup some or all of the loan proceeds.
Down Payments. To successfully induce borrowers not to default, the

collateral must be valuable enough to give individuals an incentive to repay
their loans. This can be a problem in the case of, say, a mortgage or a new-
car loan. Suppose the only collateral put up for a mortgage loan is the house,
and the money used to purchase the house comes solely from the bank.
Then, in effect, no collateral exists. If the borrower defaults, the bank seizes
the house, but the borrower receives the free use of it for the period up until
the bank repossesses it. Clearly this strategy will not induce dishonest bor¬
rowers to repay loans.
One common mechanism banks use to counter this problem is to require
a down payment. For example, suppose a bank agrees to lend a borrower 90
percent of the value of a house; the other 10 percent must come from the
borrower’s savings. In this case, the bank’s stake in the house is only 90
percent of the house’s market value. If the borrower defaults and the bank
repossesses the house, it can sell it and use the proceeds to repay both the
loan and the costs of foreclosing on the house. The borrower, in turn, loses
the 10 percent down payment. Thus, a down payment reduces the incentive
for even dishonest borrowers to default.
Does the use of down payments and collateral work in the real world?
Mortgage lenders in the Southwest experienced numerous defaults in the
late 1980s, despite the fact that loans required down payments and were
collateralized. Is this consistent with our previous discussion? The answer
is yes. During the late 1980s, real estate prices in the Southwest fell by as
much as 50 percent in some places. To see the impact of this on the incentive
to default, consider a borrower who obtained a mortgage on a $100,000
house. The bank required a 10 percent down payment, so the mortgage itself
was $90,000, with the additional $10,000 coming from the borrower. As the
above discussion indicates, when the market value of the house is $100,000,
the borrower has no incentive to default; doing so would effectively give
the bank an asset worth more than what was owed to the bank.
Now suppose that after the buyer has obtained the loan, the market price
of the house falls, leaving the house worth only $50,000. The house is now
worth less than the amount owed to the bank; thus, it is as if the bank had
no down payment. If the borrower does not default, he pays the bank $90,000
in exchange for a house worth only $50,000. If the borrower defaults, he
pays nothing, and the bank receives an asset worth only $50,000. Thus, since
the price of the house fell below the value of the loan, the borrower has an
incentive to default. To induce a borrower to repay the loans, the down
payment must be sizable enough to keep the loan value below the market
value of the house throughout the life of the loan. This did not happen in
the Southwest, and consequently the number of defaults skyrocketed during
that period.
Of course, lenders can always sue for damages, even after seizing col¬
lateral. Thus, banks in the Southwest could sue borrowers for losses not
covered by the sale of the collateral. However, this process would be both
expensive and time consuming. The value of collateral and down payments
is that they are relatively inexpensive means of recovering losses from
defaults on certain loans. However, when housing prices in the Southwest
fell drastically, the value of collateral and down payments was insufficient
to allow recovery of the loan amounts. As a result, banks and savings and
loan institutions throughout the Southwest were severely weakened. Box 5.3
further illustrates the plight of savings and loans in the 1980s.
Summary j Banks that issue car loans require a much larger down payment on a used
Exercise 5.4 car than on a new one. Furthermore, interest rates on used-car loans are
higher than rates on new cars. Why?
Answer: The seller of a car has better information about its condition than
the buyer does. This leads to adverse selection: Car owners are more likely
to keep good cars and sell bad ones, so the market for used cars will likely
be composed of many “lemons.” (This situation is similar to the loan
market, in which the group of potential borrowers consists of many “lem¬
ons” who have a high probability of borrowing and not repaying). Before
purchasing a used car, a buyer can hire a mechanic to inspect the car and
determine whether it is a jewel, a lemon, or something in between. But
ultimately, the true test of whether or not a used car is a lemon comes when
the owner regularly drives the car.
Banks, of course, know this and recognize that a used-car buyer will
have an incentive to default on a used-car loan if the car turns out to be a
lemon. Therefore, the marginal cost to a bank of issuing a used-car loan is
higher than the marginal cost of issuing a loan for a new car. As the figure
shows, the higher marginal cost of issuing used-car loans (MC^) leads to a
higher interest rate (iuL) relative to the lower marginal cost for new-car loans
(MQv) and the corresponding interest rate on such loans (i%).
Inside Money
Going for Broke in the
Savings and Loan Industry
In the early 1980s, many savings and loans ex¬ (50,000,000 + — $100,000,000)/2 =
perienced severe financial difficulty due to the - $25 million,
rise in interest rates they had to pay on deposits,
which is negative and considerably lower than
while the interest rates on outstanding mort¬
that of alternative 1. However, if things go well,
gage loans were fixed and often below the in¬
the savings and loan will earn $50 million and
terest paid on deposits. Thus, many savings and
end up solvent, with a net worth of zero. If
loans were insolvent according to Generally Ac¬
things go badly, of course, it will have a net
cepted Accounting Practices. Bank managers
worth of - $1 50 million.
faced perverse incentives to choose riskier in¬
What do you think the manager facing
vestments, which tended to accentuate the in¬
these two hypothetical alternatives will do? You
solvency of S&Ls.
might think she would choose alternative 1,
An example will illustrate the nature of this
since it both pays a higher average return and
perverse incentive. Consider a manager of an in¬
involves less risk. However, because the savings
solvent savings and loan with a net worth of
and loan is already insolvent, the manager actu¬
— $50 million. This manager has only two in¬
ally has an incentive to choose the riskier option,
vestment alternatives. Alternative 1 yields a re¬
alternative 2. The reason is as follows. If things
turn of $25 million if things go well, which will
go well, the savings and loan will be restored to
happen 50 percent of the time, and a return of
solvency, and the manager will be a hero. If
$0 if things go badly, which also will happen 50
things go badly, the S&L will still be insolvent,
percent of the time. Thus, this alternative yields
but more deeply in the hoie. But the magnitude
an average (or expected) return of
of the insolvency is not important, since the
($25,000,000 + $0)/2 = $12.5 million. manager and stockholders are already in a posi¬
tion of negative net worth. The limited liability
If things go well, the savings and loan will earn of corporations prevents stockholders from hav¬
$25 million, reducing its negative net worth to ing to come up with funds to cover the negative
-$25 million, whereas if things go badly, the net worth. Furthermore, the depositors are cov¬
S&L's net worth will remain - $50 million. ered by federal deposit insurance, which guar¬
However, there is another option, alterna¬ antees their deposits will be made good with
tive 2, which is riskier than alternative 1. This al¬ federal dollars if the savings and loan is insol¬
ternative pays $50 million when things go well vent.
(50 percent of the time) but loses $100 million Thus, neither the manager, shareholders,
when things go badly (also 50 percent). The av¬ nor even depositors have a reason to object to
erage return of this alternative is Continued on p. 163
Conclusion
Continued from p. 162 to prevent the savings and loan from taking on
high risk in the hope that a lucky occurrence will
the go-for-broke strategy of alternative 2. In restore it to solvency.
fact, if alternative 2 fails and things go badly, Although purely hypothetical, this example
this will encourage an even riskier investment in reflects the underlying incentives bank managers
the future. The problem is that no one except faced during the early 1980s. As a result of
the Federal Deposit Insurance Corporation (and these go-for-broke decisions, numerous S&Ls
those ultimately responsible for paying for this failed in the 1980s, leading to the multibillion-
insurance, the taxpayers) will have any incentive dollar S&L bailout by the federal government.
Banks also reduce the likelihood that used-car buyers will default by re¬
quiring larger down payments on used cars. With a small down payment, a
borrower will tend to default even if he or she learns the car has only minor
quirks. But with a large down payment, the used-car buyer will have to learn
the car is truly a lemon before he or she has an incentive to default.
Conclusion
In this chapter, we looked at how banks behave in the loan market. The
structure of the loan market, be it pure competition, monopoly, or oligopoly,
influences this behavior. We stressed the importance of the strategic elements
of actions by banks in oligopoly by modeling this behavior as a game
between banks. Finally, we examined the impact of asymmetric information
between banks and borrowers in the loan market.
In the next chapter, we look at the behavior of banks in the deposit
market and how that behavior affects the structure of the deposit market.
We also examine the important issues of deposit insurance and bank failure
in connection with these topics.
Key Terms
pure competition oligopoly bank

interest elasticity of demand for loans collusion
marginal cost imperfect information
marginal revenue symmetric information
monopoly banks asymmetric information
economies of scale adverse selection
location advantage collateral
1. Who demands bank loans? Who supplies (a) . What is the bank’s average cost of is¬
them? suing $100,000 loans?
(b) . How does one interpret this number?
2. Suppose the market interest rate on bank
loans is 10 percent. How many loans will 7. Explain why a bank may have market
a purely competitive bank issue if it power.
charges an interest rate on loans of 10.5
8. ‘ ‘If a bank with market power raises its
percent? Explain.
interest rate above those charged by other
3. What is the impact on the costs to a bank banksr no borrowers will seek funds from
if an increase occurs in (a) the interest rate that bank.” Is this statement true or false?
paid to depositors for funds, (b) the fre¬ Explain.
quency with which borrowers default, and
9. Suppose a bank with market power expe¬
(c) the wage paid to loan processors?
riences a decrease in the marginal cost of
4. Suppose the market demand for bank issuing loans.
loans increases in a purely competitive (a) . Describe the factors that could lead to
market for loans. Using words and a dia¬ a decrease in the bank’s marginal cost.
gram, explain the impact of this on (b) . What is the impact of the decrease in
(a) . The market interest rate marginal cost on the bank’s profit-
(b) . The demand for loans issued by an maximizing interest rate on loans and on
individual bank the corresponding level of loans issued?
(c) . The number of loans issued by an in¬ Use a graph in your answer.
dividual bank (c) . Do bank profits necessarily increase
(d) . The profits of an individual bank if the marginal cost of issuing loans de¬
creases? Explain.
5. Explain why a purely competitive bank is¬
sues a quantity of loans such that the in¬ .
10 Why are there laws against collusion in
terest rate on loans equals the marginal banking?
cost of issuing another dollar in loans.
Answer questions 11-14 based on the fol¬
6. Suppose the cost to a bank of issuing lowing profit matrix for two duopoly
$100,000 in loans is $8,000. banks:
Bank Two riod if the interest rate used in their pres¬

Interest ent value calculations is
Rate 8% 15% (a) . 10 percent
Bank (b) . 20 percent
One 8% 0,0 25, -5 (c) . 30 percent
15% -5, 25 5, 5 (d) . 40 percent
15. Explain the difference between symmetric
11. What are Bank One’s profits if it charges and asymmetric information.
an interest rate of 8 percent when Bank
16. In the presence of symmetric information
Two charges an interest rate of (a) 8 per¬
about the ability of borrowers to repay
cent and (b) 15 percent?
loans, would you expect banks to charge a
single interest rate to all borrowers? Why
12. What are Bank Two’s profits if it charges or why not?
an interest rate of 15 percent when Bank
One charges an interest rate of (a) 8 per¬ 17. In a recent television advertisement, a
cent and (b) 15 percent? bank offered to issue a credit card to any¬
one regardless of his or her credit history.
13. Suppose the banks do not compete repeat¬ What would you expect the interest rate to
edly. What are their equilibrium profits? be on credit card loans at this bank com¬
Why? pared to those at other banks? Why?
18. Suppose you are the president of the bank

14. Suppose the banks compete repeatedly in question 17. Given the bank’s policy,
over time. Determine whether (and if yes, what can you do to reduce or eliminate
how) they can earn profits of $5 each pe¬ the problem of adverse selection?
Selections for Further Reading
Edmister, R. O., and H. E. Merriken. “Pricing Effi¬ Banking.” Quarterly Journal of Business and Eco¬
ciency in the Mortgage Market.” American Real nomics, 30 (Autumn 1991) 82-107.
Estate and Urban Economics Association Journal, Ostas, J. R. “Effects of Usury Ceilings in the Mort¬
16 (Spring 1988), 50-62. gage Market.” Journal of Finance, 31 (June
Gropper, D. M. “An Empirical Investigation of 1976), 821-834.
Changes in Scale Economies for the Commercial Peterson, R. L., and D. A. Black. “Consumer Credit
Banking Firm, 1979-1986.” Journal of Money, Search: A Note.” Journal of Money, Credit, and
Credit, and Banking, 23 (November 1991), 718— Banking, 16 (November 1984), 527-535.
727. Shaaf, M., and P. A. Smith. “The Role of the Re¬
Knudsen, J. W. “Consumer Spending and Economic gional Economy in Oklahoma Bank Failures.”
Activity.” Federal Reserve Bank of Kansas City Journal of Economics, 15 (1989), 86-91.
Monthly Review, (February 1971), 3-10.
Kolari, J., and A. Zardkoohi. “Further Evidence on
Economies of Scale and Scope in Commercial
CHAPTER
The Bank as a Firm: Deposits
ur study of loan decisions in the previous chapter may have left you
wondering how deposits fit into this picture. After all, individual banks not
only supply loans to borrowers but also attract deposits from savers, and in
fact pay interest to those who have funds deposited in the banks. How do
banks determine how many deposits they want to attract and how much
interest to pay on deposits?
In this chapter, we address these issues by applying the theory of pro¬
duction to banking firms. Basically deposits are an input into the production
of loans, and our task here is to examine how banks determine the interest
rate and the amount of deposits that will maximize profits. As we will see,
the answers to these questions depend on the degree of market power banks
enjoy in both the loan market and the deposit market. In the final part of
this chapter, we look at problems that arise due to uncertainty about when
deposits will be withdrawn and discuss several strategies banks use to alle¬
viate such difficulties.
Deposits as an Input in Producing Loans

Banks use many inputs to produce loans. Like any other firm, a bank uses
labor (e.g., tellers, accountants, loan officers), buildings, computers, and
utilities. It also issues commercial paper, stocks, and bonds to obtain financial
and working capital. Unlike with other firms, however, the primary source
of the funds a bank uses to produce its output is the reserves created when
the bank accepts deposits. As we saw in Chapter 1, at the very minute you
deposit $1,000 into your checking or savings account, the reserves at your
bank increase by $1,000. By law, the bank must keep a fraction (say, 9
percent) of your deposit as required reserves. It may also hold an additional
1 percent as excess reserves against withdrawals. The remaining 90 percent
will be used to supply loans in the loanable funds market, as discussed in
Chapters 4 and 5, until the bank ends up with a balance sheet like that in
Table 6.1.
Thus, the primary reason your bank wants your deposits is that they
create reserves the bank can use to produce loans on which it earns interest
income. Therefore, the individual bank’s demand for deposits (and indeed
the demand for any input into the production process) is called a derived
166
D E POSITS AS AN IN PUT IN PRODUCING LOANS 167
This bank has $1,000 Table 6.1

worth of deposits (a lia¬ Balance Sheet for First Hypothetical Bank
bility), of which 10
percent is kept as re¬
serves (an asset). The
Assets Liabilities
bank loans out the re¬
maining $900 in as¬
sets in the loanable Reserves $100 Deposits $1,000
funds market.
Loans $900 Net Worth $0
demand—demand derived from the demand for loans that banks try to
satisfy. To highlight the role deposits play in producing loans, we look at
their productivity in producing loans by taking the levels of all other inputs,
such as accountants, tellers, and loan officers, as given.
Total and Marginal Product

The demand for deposits depends in part on their productivity. Total product
refers to the total level of output produced with a given quantity of an input.
For example, if $100,000 in deposits permits a bank to issue $80,000 in
loans, the total product of $100,000 in deposits is $80,000 in loans.
Marginal product is the change in total product (or total output) that
results from a one-unit change in the usage of an input, holding the quantities
of other inputs constant. For instance, if a $1 increase in deposits increases
the amount of loans issued by 80 cents, the marginal product of deposits is
80 cents in loans.
In most production processes, initial increases in the use of an input will
lead to successively larger increases in total output. In a hairstyling salon,
for example, the first hairdresser hired increases the number of haircuts from
zero to, say, two haircuts per hour. The marginal product of the first hair¬
dresser is thus two haircuts. The second hairdresser hired increases the
number of haircuts per hour from two to five. The marginal product of the
second hairdresser is three haircuts, which is greater than the marginal prod¬
uct of the first hairdresser. The reason for the increase in the marginal product
of the second hairdresser may be the fact that when the phone rings, one
hairdresser can continue to cut hair while the other answers the phone. When
there is only one worker, production stops when the phone rings.
For given levels of other inputs, as more and more of a specific input is
used, its marginal product eventually begins to decline; that is, total product
continues to increase, but at a decreasing rate. For example, hiring a third
hairdresser may increase output from five to seven haircuts per hour. In this
instance, the marginal product of the third hairdresser is two haircuts, which
is less than the marginal product of the second hairdresser. Furthermore, as
more and more hairdressers are packed into the salon, the marginal product
168 Chapter 6 The Bank as, a Firm: Deposits
of each additional hairdresser gets smaller and smaller. This phenomenon is

known as the law of diminishing marginal returns, which states that if the
amount of one input increases and the amounts of all other inputs remain
constant, the marginal product of the input eventually begins to decline.
Applied to banking, the law of diminishing marginal returns says that
for given quantities of other inputs, increasing the amount of deposits will
eventually lead to a declining marginal product of deposits. Banks must hire
labor and capital equipment (such as computers) to monitor loan payments,
staff teller windows, collect bad debts, process loan applications, and keep
the books on each deposit and each loan account. For given levels of these
inputs, as the bank attracts more and more depositors, the desks of loan
officers and accountants pile up with new-account forms and loan applica¬
tions. As a result the amount of new loans the bank can issue eventually
diminishes.
Part a of Figure 6.1 shows the total product curve for loans—the rela¬
tionship between loans produced by a bank and the amount of deposits in
the bank, holding other inputs constant. As deposits increase from $0 to
$100 million, the amount of loans the bank issues rises at an increasing rate.
For example, the first $50 million in deposits yields $40 million in loans,
while the second $50 million increases loans by $45 million. The marginal
product of the first $50 million in deposits is $40 million in loans, and the
marginal product of the second $50 million in deposits is $45 million in
loans. The “extra” $5 million in loans is reflected in the increasing slope
of the total product curve as deposits rise from $0 to $100 million.
Part b graphs the marginal product of deposits (MPD), which is less
when deposits are $50 million than when deposits are $100 million. Note
the relation between the marginal product of deposits graphed in part b and
the slope of the total product curve in part a: As the total product curve gets
steeper, its slope increases and the marginal product of deposits also rises.
When the total product curve gets flatter, its slope declines and the marginal
product of deposits falls.
Notice that as deposits increase beyond $100 million in part a of Figure
6.1, the total amount of loans increases, but at a decreasing rate. In particular,
increasing deposits from $100 million to $200 million increases loans by
only $45 million, $40 million less than the $85 million in loans that resulted
from increasing deposits from $0 to $100 million. The slope of the total
product curve declines as deposits increase beyond $100 million, indicating
that the marginal product of each additional deposit declines over this range.
This is consistent with the declining marginal product curve in part b for
deposits beyond $100 million. In part b, the law of diminishing marginal
returns applies when deposits exceed $100 million.
A Simple Technology for Producing Loans

We can more clearly see the link between bank loans and bank deposits by
looking at a simple loan technology that builds on what we learned about
D E POSITS AS AN INPUT IN PRODUCING LOANS 169
Figure 6.1
Total and Marginal Products of Deposits
. ,±;
This figure shows the production function for loans as a function of deposits, holding other
banking inputs constant. As deposits increase from $0 to $200 million, the bank's loans (its
total product, or output) increase from $0 to $130 million. This is shown in part a, where
loans are graphed on the vertical axis and deposits on the horizontal axis. Part b graphs the
marginal product of deposits—the slope of the production function in part a. As deposits
increase, so do loans, but the rate of increase changes as the size of deposits changes. The
first $100 million in deposits yields $85 million in loans, while the next $100 million in
deposits yields only $45 million in additional loans. This is due to the law of diminishing
marginal returns, which is illustrated by the declining marginal product of deposits when
deposits exceed $100 million in part b.
Deposits (Millions $)
fractional reserve banking in Chapter 1. Suppose the required reserve ratio

is rr, meaning the bank is required by law to keep a constant fraction, rr, of
its deposits on reserve. Thus if we let D denote deposits, the bank can legally
lend up to (1 — rr) X D of its deposits. However, as deposits increase, the
bank may not be able to immediately issue the quantity of loans allowed by
law due to diminishing marginal returns. This is most likely to be the case
at very small banks, where loan officers sometimes are unable to process
sufficient loan applications to keep pace with rising deposits.
170 Chapter 6 The Bank as a Firm: Deposits
In this case, the production of loans is a function of the amount of

deposits the bank can legally loan out:
L = F([l - rr\ X D).
For instance, if the required reserve ratio is .09 and deposits are $1 million,
the bank can legally issue up to $910,000 in loans. How much it will actually
produce depends on its production function. Perhaps the bank will only issue
$900,000 in loans. When deposits grow to $2 million, the bank will be
legally able to issue up to $1,820,000 in loans, since [1 — .09] X $2,000,000
= $1,820,000. However, because of diminishing marginal returns, the bank
actually might transform deposits into only $1,750,000 in loans. Like the
total product curve graphed in part a of Figure 6.1, these numbers illustrate
that a doubling of bank deposits will not double the amount of loans when
diminishing marginal returns to deposits exist. We will look more closely at
how changes in the required reserve ratio affect the entire banking industry
and the overall economy in Chapters 7, 13, and 14.
Suppose a large bank can produce loans according to the relation
L — (1 — rr) X D,
where L is the amount of loans, rr is the constant required reserve ratio, and
D is the level of deposits, (a) Graph this bank’s total product curve as a
function of deposits when the required reserve ratio is . 1. (b) Graph the total
product curve when the required reserve ratio is .2. What can you conclude
about the impact of a change in the required reserve ratio on the production
of loans? (c) Graph the bank’s marginal product curve when the required
reserve ratio is .1 and .2. What happens to marginal product when the
required reserve ratio increases? (d) Does this bank’s production of loans
satisfy the law of diminishing marginal returns?
(a) When the required reserve ratio is .1, the formula for the total
product curve is L = .9 X D, which is graphed as the linear relation in part
a. Notice that $100,000 in deposits produces $90,000 in loans, (b) When the
required reserve ratio increases to .2, the total product curve becomes L =
.8 X D, which is also plotted in part a. Given the higher required reserve
ratio, only $80,000 in loans can be produced with $100,000 in deposits. We
conclude that an increase in the required reserve ratio decreases the total
product curve for loans, (c) When the required reserve ratio is .1, each
additional dollar in deposits increases loans by 90 cents, so the marginal
product of deposits is .9. This is graphed in part b as the horizontal line MPD
= .9. When the required reserve ratio increases to .2, each additional dollar
in deposits increases loans by 80 cents, so the marginal product of deposits
is .8. This is graphed in part b as the horizontal line MPD = .8. Thus, an
increase in the required reserve ratio decreases the marginal product of
The Demand for Deposits by Purely Competitive Banks 171
deposits, (d) Since the marginal product curve is horizontal, the production
process for this bank does not exhibit diminishing marginal returns. Here a
doubling of bank deposits leads to a doubling of loans, since the bank is
able to maintain the same level of excess reserves (zero) regardless of the
size of its deposits.
Marginal
Product
MPd= .9
MPd= .8
Deposits (Thousands $)
The Demand for Deposits by Purely Competitive Banks

Now that we understand the production relationship between a bank’s output
(loans) and input (deposits), we look at how banks determine the interest
rate on deposits and the level of deposits needed to produce loans. These
figures vary with the market structure of the deposit market. Similar to what
we saw in the loan market, purely competitive banks and banks with market
power behave differently. We begin by looking at a bank in a purely com¬
petitive market.
Market Supply and the Supply of

Deposits to an Individual Bank
In a purely competitive deposit market, individual banks are small relative
to the entire market for deposits and must pay their depositors the market
rate on deposits. Depositors would not choose to deposit funds in a bank
that offered a lower rate on deposits than the rate other banks offered.
When the market for bank deposits is purely competitive, the intersection
of the market demand and supply of deposits determines the market interest
rate on deposits. All banks demand deposits to use as an input in producing
loans. Households and businesses supply these deposits. Part a of Figure 6.2
illustrates equilibrium in the deposit market; the market interest rate on
172 Chapter 6 The Bank as,a Firm: Deposits
deposits—determined by the intersection of the market demand and supply

curves—is 4 percent.
The supply of deposits to an individual bank is horizontal at the market
interest rate on deposits, which is 4 percent in part b of Figure 6.2, because
the individual bank must offer that rate. In contrast, the supply of deposits
to all banks in the deposit market is upward sloping in Figure 6.2. The reason
for the difference is that at the individual bank level, depositors are very
sensitive to changes in the bank’s interest rate. If a purely competitive bank
reduced the rate it paid on deposits below the market rate its competitors
paid, the bank would lose its depositors to other banks. Thus, the supply of
deposits to an individual purely competitive bank is perfectly elastic at the
market rate on deposits.
The Value Marginal Product of Deposits

Purely competitive banks must pay the market interest rate on deposits (iD)
to attract deposits. As we learned in the previous chapter, they also must
charge the market interest rate on loans (iL) if they wish to issue any loans.
(Box 6.1 shows how bank profits are partly determined by the difference in
these rates.) While purely competitive banks have no power over the interest
rates charged for loans or paid on deposits, they can control the number of
loans they issue at those rates. In Chapter 5, we saw how purely competitive
banks determine the level of loans that maximizes profits. All that remains
to complete our analysis of purely competitive banks is to see how banks
determine the amount of deposits they need to generate these loans.
Let MPd represent the marginal product of deposits—the additional
amount of loans a bank can issue if it acquires an additional $1 in deposits.
Since the bank loans out money at the market interest rate of iL, the value
marginal product of deposits (VMP/)) to the bank—the additional revenue
it gets from an additional $1 of deposits—is
VMPd = iL X MPd.
The value marginal product of an additional $1 of deposits simply reflects

the value of the interest income that will be generated by obtaining an
additional $1 of deposits and converting the resulting reserves into loans.
To maximize profits, a purely competitive bank continues to attract deposits
up to the point where the value marginal product of deposits equals the
bank’s cost of acquiring an additional dollar of deposits:
VMPd = iD.
To see why, suppose a $1 increase in deposits increases loans by 80

cents so that the marginal product of the last deposit (the increase in loans)
is .80. If the interest rate on loans is 10 percent, the value of the marginal
product of the deposit is
VMPd = .1 X .80 = .08.
In other words, the last dollar in deposits earned the bank 8 cents in interest
income. If the market interest rate the bank must pay on deposits is 4 percent,
the cost to the bank of one more dollar in deposits is only $1 X .04, or 4
cents. It costs the bank less (4 cents) to attract another dollar in deposits
than it will receive by converting the deposit into a loan (8 cents). Clearly
it pays for the bank to obtain additional deposits and convert them into loans.
But as the bank acquires more deposits, the law of diminishing marginal
returns implies that the marginal product of deposits falls, which reduces the
value marginal product. As the bank attracts additional deposits, VMPD falls
until it eventually equals the interest rate on deposits, iD. The bank will not
actively seek additional deposits, because doing so would reduce the value
\
Loan and Deposit Interest Rates
Banks make profits on the difference between the rate charged on loans and the rate paid on
the interest rate charged on loans and the inter¬ deposits represents the bank's spread—the
est rate paid on deposits. One key interest rate profit margin it earns on its funds.
on loans is the prime interest rate—the rate The accompanying figure plots the deposit
banks charge to creditworthy commercial cus¬ interest rate as measured by the rate on three-
tomers. As such, it is a benchmark of the rate at month certificates of deposit, along with the
which such firms can borrow. Less creditworthy loan interest rate charged to commercial bor¬
firms or individuals usually pay a rate in excess rowers (the prime loan rate) for the period
of the prime rate. On the deposit side, banks 1972-1992. With a few exceptions in the early
also pay a host of different rates, depending on 1970s, the loan rate exceeded the deposit rate,
the type of deposit. Some checking accounts re¬ revealing a generally positive spread between
ceive no interest at all, while other deposits re¬ the loan and deposit rates. The spread changed
ceive various rates depending mostly on how over time but has remained moderately steady
long a depositor is willing to commit to keeping since the early 1980s. At the end of
the funds in the bank. The difference between Continued on p. 175
Continued from p. 174 are not subject to reserve requirements) at a 6

1992, the prime rate was 6 percent and the percent annual rate to a commercial customer
three-month CD rate was 3.44 percent. Ignoring such as AT&T. In doing so, the bank could not
other costs, in 1992 a bank could take in only link borrowers and lenders but earn 2.56
$1,000 in a CD and pay a 3.44 percent annual percent for its efforts as a financial intermediary
rate to the depositor while lending out the en¬
tire amount if it so desired (because CD deposits Source: Citibase electronic database.
marginal product to a level below the interest rate on deposits. In that case,
the bank would pay more for the last deposit than it could earn by converting
it into a loan. Thus, to maximize profits, a purely competitive bank will
agressively seek deposits up to the point where VMPD = iD.
Figure 6.3 illustrates the profit-maximizing level of deposits for a purely
competitive bank. The vertical axis measures the interest rate on deposits,
and the horizontal axis measures the dollar amount of deposits. The supply
of deposits to the individual bank is horizontal at the market interest rate of
4 percent. The point at which the value marginal product of deposits (VMPD)
equals the interest rate on deposits denotes the profit-maximizing level of
deposits, D*. Since the VMPD curve determines the bank’s quantity of
To maximize profits, Figure 6.3

the bank will attract Profit-Maximizing Amount of Deposits in Pure Competition
enough deposits to
equate the value mar-
ginal product of de- Interest Rate
posits to the interest on Deposits
rate on deposits. The (%)
value marginal prod-
uct of deposits, given
by the downward- \
sloping curve VMP0/ is \
the marginal product \
of an additional $1 of
Supply of
deposits times the in- 4
terest rate the bank
;\ Individual Bank
charges on loans. The ; \
VMPd curve is a purely \
competitive bank's de- \
mand curve for depos- \ VMPn
its
D*
Deposits at an Individual Bank (Millions $)
176 C HAPTER 6 The Bank as a Firm: Deposits
deposits demanded at each interest rate on deposits, it is the individual purely

competitive bank’s demand curve for deposits. The demand for deposits by
a purely competitive bank is downward sloping due to the law of diminishing
marginal returns.
Summary Graphically illustrate the effect on a purely competitive bank’s profit-

Exercise 6.2 maximizing demand for deposits if (a) the interest rate on loans increases
and (b) the required reserve ratio on deposits increases.
Answer: (a) An increase in the interest rate on loans {if) increases the
value marginal product of deposits, since VMPD = iL X MPD. In part a, if
the initial interest rate on loans is if the initial value marginal product curve
for deposits is VMP?) = i°L X MPD. Given the market interest rate on deposits
(if), the bank attracts D0 in deposits. An increase in the interest rate on loans
to i[ increases the value marginal product of deposits to VMP^ = ilL X
MPd, which is a rightward shift in the demand for deposits in the figure. As
shown, this increases the equilibrium quantity of deposits the bank desires
to Dj. (b) An increase in the required reserve ratio reduces the marginal
product of deposits from MP?, to MP]), since a greater fraction of deposits
must be kept in reserve than before. Graphically, this shifts the demand for
deposits to the left in part b, reducing the equilibrium amount of deposits
from D0 to Dx.
Interest Rate Interest Rate

on Deposits on Deposits
Deposits at an Individual Bank (Millions $) Deposits at an Individual Bank (Millions $)
(a) (b)
The Demand for Deposits by Banks with Market Power
The preceding analysis is relevant only for very small banks that are unable
to influence interest rates. In contrast, larger banks typically have power that
enables them to raise loan rates or lower deposit rates without losing all of
their customers. To determine the profit-maximizing level of deposits for
banks with market power, we distinguish between two possible cases. In the
first, the bank has market power in issuing loans but no market power in
obtaining deposits. In the second, the bank has market power in both the
market for loans and the market for deposits.
Market Power in the Loan Market Only

We first consider a bank that has market power in the loan market but not
in the market for deposits. Thanks to its market power in the loan market,
the bank faces a downward-sloping demand for loans, as we saw in the
previous chapter. This means the bank must lower the interest rate on loans
to issue additional loans. However, since the bank operates in a purely
competitive deposit market, it can readily obtain deposits at the market
interest rate on deposits.
Figure 6.4 illustrates this situation. In part a, demand for the individual
bank’s loans is downward sloping, which reflects the bank’s market power
in the loan market. In part b, the supply of deposits to the bank is perfectly
If a bank has market Figure 6.4

power in the market Demand for Loans at a Bank with Market Power and Supply
for loans, it must of Deposits to Individual Banks
lower the interest rate
it charges for loans to
make more loans. This
Interest Rate
is illustrated in part a
on Loans
by a downward-slop¬
ing demand curve for
loans. Since this bank
operates in a purely (a)
competitive deposit
market, it faces a hori¬
Demand for Loans at
zontal deposit supply
an Individual Bank
curve in part b. The
bank need not change
Individual Bank's Amount of Loans (Millions $)
the interest rate on
deposits to attract
more deposits. Interest Rate
on Deposits
(b)
Market / Supply of Deposits to
Interest an Individual Bank
Rate
Deposits at Individual Bank (Millions $)

elastic at the market-determined interest rate, reflecting the fact that the bank
obtains deposits in a purely competitive deposit market.
Each additional dollar in deposits the bank obtains increases loans by
the marginal product of deposits, MP#. Since the bank has market power in
the loan market, to issue additional loans it must lower its loan interest rate.
The additional revenue the bank receives when it issues another dollar in
loans is the marginal revenue of a loan, MRL. If we multiply the additional
revenue received from a loan by the change in loans resulting from the
receipt of an additional dollar in deposits, we obtain the marginal revenue
product of deposits (MRP#):
MRP# = MRl X mp#.
This tells us the additional interest income the bank will earn if it accepts
another dollar in deposits and converts the corresponding reserves into loans.
How many deposits must a bank with market power in the loan market
attract to maximize profits? When the interest rate on deposits is market
determined and given by /#, the bank attracts deposits until the marginal
revenue of loans multiplied by the marginal product of deposits equals the
interest rate on deposits:
MRl x MP# = /#,
or
MRP# = iD.
Let’s look at the intuition behind this rule. MP# is the additional amount of
loans the bank can make with the reserves generated from an additional
dollar of deposits. MRL tells us how much the increase in loans will increase
the bank’s revenue. Multiplying marginal product by marginal revenue gives
us the additional revenue from another dollar of deposits (since the marginal
product tells us the additional loans, and multiplying by marginal revenue
tells us the additional revenue from these additional loans). For profit max¬
imization, MRP# should equal the cost of an additional dollar of deposits,
which is the interest rate on deposits (/#). Thus, when the profit-maximizing
level of deposits is achieved, the additional revenues from attracting the last
dollar of deposits just equals the additional cost of the last dollar of deposits.
Table 6.2 illustrates this principle. Deposits vary from $0 to $200 mil¬
lion, and the associated level of loans varies from $0 to $130 million. The
marginal product of deposits is calculated as the change in loans (AL) divided
by the change in deposits (AD). For example, when deposits rise from $0 to
$50 million, loans increase from $0 to $40 million, so the marginal product
of deposits is $40/$50 = .8. When deposits increase from $50 million to
$100 million, loans increase from $40 million to $85 million, and the mar¬
ginal product of deposits is $45/$50 = .9.
Columns 2 and 4 in Table 6.2 show the demand for loans at this bank.
When the interest rate in column 4 is high, such as 15 percent, the quantity
of loans demanded in column 2 is $0. As the interest rate falls to 11 percent,

the quantity of loans demanded rises to $40 million in accord with the law
of demand. Multiplying each entry in column 2 by the corresponding entry
in column 4 gives us column 5, the revenue (interest income) the bank earns
from issuing different amounts of loans: R = iL X L. For instance, if the
bank charges 11 percent, it will be able to issue $40 million in loans to earn
revenues of. 11 X $40 million, or $4.4 million. If the bank lowers its interest
rate to 6.5 percent, the amount of loans it issues increases to $85 million
and revenues increase to $5,525 million.
Column 6 summarizes the marginal revenue to the bank of issuing
loans—the change in bank revenues divided by the change in the amount of
loans. For instance, if the bank increases its loans from $0 to $40 million,
the marginal revenue is ($4.4 — $0)/($40 — $0) = .11, or 11 percent. Thus,
column 6 contains an entry of 11 percent when the bank issues $40 million
in loans. When loans are $85 million, we calculate marginal revenue as
($5,525 — $4.4)/($85 — $40) = .025, or 2.5 percent. Notice that as the
amount of loans increases from $40 million to $85 million, marginal revenue
declines from 11 to 2.5 percent. Furthermore, if the bank increases its loans
to $115 million, marginal revenue becomes negative. This indicates that
after some point, as the bank issues more loans, its total revenues fall because
the additional revenue from issuing more loans is more than offset by the
revenue lost from the reduction in the interest rate the bank needed to issue
more loans.
Finally, column 7 of Table 6.2 calculates the marginal revenue product
for deposits as the marginal revenue of loans (column 6) multiplied by the
marginal product of deposits (column 3). The bank maximizes profits by
continuing to attract deposits up to the point where the marginal revenue
product of deposits equals the interest rate on deposits. For example, if this
bank can obtain all the deposits it wants at an interest rate of 2.25 percent,
it will choose to attract $100 million in deposits to maximize profits. When
deposits are $100 million, the interest rate on deposits of 2.25 percent will
just equal the marginal revenue product of deposits. The bank will issue
loans of $85 million at a loan interest rate of iL — 6.5 percent. It will have
revenue (interest income) of $85 million X .065 = $5,525 million, while
its interest costs on deposits are $100 million X .0225 = $2,225 million.
Figure 6.5 graphs the tables’ results. The MRPD curve is the demand for
deposits by this bank, which has market power only in the loan market. The
vertical axis shows the interest rate on deposits (which is determined in the
purely competitive deposit market), while the horizontal axis shows the
amount of deposits. When the market interest rate on deposits is 2.25 percent,
the profit-maximizing level of deposits is $100 million, where iD = MRPD.
There are two reasons the demand for deposits slopes downward in the
case of a bank with market power in only the loan market. First, if dimin¬
ishing marginal returns exist, the marginal product declines as additional
deposits are obtained, causing MRPD to decline. Second, because the mar-
. .
• . .: • Table 6.2
-
Calculating the Marginal Revenue Product of Deposits Sf?r
The marginal revenue product of deposits in column 7 is the marginal product of deposits (column 3) times the mar-
ginal revenue of issuing loans (col umn 6). The marginal revenue product of deposits declines as the bank receives more
deposits.
Deposits Loans Marginal Interest Revenue Marginal Marginal

(Millions of (Millions of Product Rate on (Millions of Revenue Revenue
Dollars) Dollars) of Deposits Loans Dollars) Product of
Deposits
AL AR MRPd =
D L MPD = — R = iL X L
1 AD
lL
MR‘ = JL MRl X MPd
(!) (2) (3) (4) (5) (6) (7)
0 0 — 15% 0 — —
50 40 .8 11% 4.4 11% 8.8%

100 85 .9 6.5% 5.525 2.5% 2.25%
150 115 .6 3.5% 4.025 -5% -3%
200 130 .3 2% 2.60 -9.5% -2.85%
ginal revenue of loans decreases as more loans are issued, MRPD declines
as deposits increase. The second factor is not present in the case of a purely
competitive bank, since the bank need not lower the interest rate to issue
more loans. Thus, a bank with market power in the loan market will have a
downward-sloping demand for deposits even if there are not diminishing
marginal returns to deposits.
Summary I Suppose the market for deposits is purely competitive. Illustrate the impact
Exercise 6.3 of an increase in the market supply of deposits on the profit-maximizing
level of deposits at a bank that has market power in the loan market.
Answer: The initial equilibrium in the deposit market is at point A in part

a, where the equilibrium interest rate on deposits is i°D. Given this interest
rate, the individual bank obtains D° deposits in part b, where the marginal
revenue product of deposits equals the market interest rate on deposits. An
increase in the market supply of deposits shifts the market supply curve
from S° to Sl in part a, resulting in a lower interest rate on deposits of if.
The individual bank now needs more deposits, Dl, to maximize profits.
Interest Rate Interest Rate

on Deposits on Deposits
Market Amount of Deposits (Millions $) Deposits at an Individual Bank (Millions $)
(a) (b)
Market Power in Both the Loan and Deposit Markets

Large banks often have market power in both the loan market and the deposit
market, and this tends to lower the interest rate they pay depositors. Let us
see why.
Any profit-maximizing bank compares the benefits of obtaining an ad¬
ditional dollar in deposits with the corresponding costs. If a bank enjoys
market power in the loan market, the relevant measure of the benefits to the
bank of obtaining an additional dollar in deposits is the marginal revenue
product of deposits (MRPD):
MRPd = MRl X MPd.
However, the cost of obtaining an additional dollar in deposits differs when

the bank has market power in the deposit market, since it faces an upward-
sloping supply curve for deposits such as the one labeled S in Figure 6.6.
This affects the bank’s cost of obtaining additional deposits. For instance, if
the bank currently obtains D0 in deposits and wishes to increase deposits to
Dx, it must increase the interest rate from i°D to i]D. In this situation, the cost
to the bank of obtaining an additional dollar in deposits is not the current
interest rate on deposits but the marginal resource cost of deposits
(MRCd), the change in the cost of deposits due to a $1 change in the level
of deposits.
Table 6.3, on page 184, shows how to calculate the marginal resource
cost. The first two columns summarize the supply of deposits to this bank.
As the interest rate on deposits rises, so does the quantity supplied. For
example, if the bank pays .5 percent interest, it attracts $25 million in
deposits, which costs the bank $0,125 million. If the bank wants to increase
its deposits to $50 million, it must increase the interest rate on deposits to 1
percent. When it does so, its costs increase to $0.5 million. The change in
cost divided by the change in deposits is the marginal resource cost. In this
case, the marginal resource cost is $.375/$25 = 0.015, or 1.5 percent.
Figure 6.5
Profit-Maximizing Amount of Deposits in a Bank
with Loan Market Power
Since this bank has market power in the loan market, the interest rate it receives on loans
depends on the amount of loans it makes. The additional revenue the bank receives from
making an additional loan is the marginal revenue of the loan, MR,. To find the demand
for deposits by the bank, we multiply the marginal revenue of the additional loan by the
marginal product of an additional deposit to obtain the marginal revenue product (MRPD
= MRl x MPd). With a market interest rate on deposits of iD = 2.25 percent, the bank
will set MRPd equal to 2.25 percent and obtain $100 million in deposits.
The bank’s MRCD is greater than 1 percent, the interest rate on deposits,
because the bank must raise the interest rate from .5 to 1 percent not just on
the additional $25 million in deposits but also on the initial $25 million.
Imagine what would happen if the bank announced it was going to pay 1
percent only on new deposits but existing deposits would still earn .5 percent!
The existing deposits would likely be withdrawn and redeposited as new
deposits. Thus, the concept of marginal resource cost takes into account the
fact that a bank with market power can raise additional deposits only by
increasing the market interest rate on all deposits, and this raises costs by
more than just the amount paid on the additional deposits. In terms of Figure
6.6, the marginal resource cost curve, MRCD, lies above the supply curve
This bank has market Figure 6.6

power in both the Marginal Resource Cost of Deposits
loan and deposit mar¬
kets. Consequently, it
must lower the inter¬
est rate on loans to
make additional loans
and must raise the in¬
terest rate on deposits
to obtain additional
deposits. The need to
raise deposit interest
rates to obtain more
deposits is shown by
the upward-sloping
supply curve of depos¬
its, 5. The curve la¬
beled MRCd is the
marginal resource cost
of deposits. To in¬
crease deposits from D0
to D1f the bank must
raise the interest rate
it pays depositors from
/'d to /?.
of deposits, because an increase in deposits raises resource costs by more

than iD.
Now that you understand the relevant benefits and costs to a bank of
obtaining additional deposits, we can easily determine the level of deposits
that maximizes bank profits. A bank with market power in both the loan and
deposit markets will maximize profit by obtaining deposits up to the point
where the additional costs of increasing deposits, MRCD, just equals the
additional revenue from increasing deposits, MRPD, or
MRPd = MRCd.
For example, in Figure 6.7, the intersection of MRPD and MRCD occurs at
point A, so the profit-maximizing quantity of deposits is D*.
What interest rate on deposits will the bank pay to attract D* deposits?
Since the bank in Figure 6.7 has market power in the deposit market, it will
pay the lowest possible interest rate that will generate D* in deposits. This
interest rate is determined by point B on the supply curve, which corresponds
to an interest rate of i%. In other words, if the bank offers depositors an
interest rate of i%, the quantity of deposits supplied will be D*, which is the
184 Chapter 6 The Bank.as a Firm: Deposits
Table 6.3
Calculating the Marginal Resource Cost of Deposits
v
The marginal resource cost of deposits is the change in deposit interest required to obtain
an additional dollar in deposits. The marginal resource cost of deposits increases as addi¬
tional deposits are received. This is because the only way a bank with market power in the
deposit market can attract more deposits is to raise the interest rate paid to depositors.
Quantity Interest Rate Interest Cost Marginal

Supplied on Deposits Resource Cost
of Deposits of Deposits
(Millions of (Millions of
Dollars) Dollars)
AC
D h) C = iD X D MRCD = —
1 AD
25 0.5% .125 —
50 1% .5 1.5%
75 1.5% 1.125 2.5%
100 2% 2.0 3.5%
125 2.5% 3.125 4.5%
150 3% 4.5 5.5%
profit-maximizing level. If the bank offered a lower interest rate, it would

attract fewer deposits than £>* and therefore would not maximize profits.
Summary Suppose that due to a reduction in the required reserve ratio, the marginal
Exercise 6.4 product of deposits increases. Illustrate the impact of this change on the
interest rate paid on deposits and the quantity of deposits obtained by a bank
with market power in both the loan and deposit markets.
Answer: The initial level of deposits is given by D°, where MRP^ =

MRCd at point A. The bank pays the lowest interest rate that will attract this
level of deposits, which is i°D. The increase in the marginal product of deposits
shifts the MRPD curve to the right to MRP|>. The new profit-maximizing
level of deposits is D\ which corresponds with point B. The bank pays the
lowest interest rate that will attract D1 in deposits, or ilD. Thus, the increase
in the marginal product of deposits induces the bank to obtain more deposits
by offering a higher interest rate on deposits.
Uncertainty and Bank Deposits

The final step in our analysis of deposits as an input into producing loans is
to show how banks deal with uncertainty concerning when deposits will be
withdrawn. We have seen that to maximize profits, a bank must convert the
reserves generated by its optimal level of deposits into loans. In doing so,
however, the bank converts liquid assets (reserves) into less liquid assets
(loans). This raises the possibility that the bank will not have enough liquid
assets to meet its depositors’ demands. In this section, we examine how
banks solve this problem and also how the failure to solve it can lead to
bank panics.
Bank Withdrawals and the Law of Large Numbers

How can a bank dare to convert into a loan the liquid reserves created when
you deposit money in your account? If you deposit $100 and the bank is
certain you will not withdraw the funds for one year, it can loan out the
reserves your deposit creates for one year without losing any sleep. In reality,
however, banks do not know how long each depositor will leave their de¬
posits in the bank before withdrawing funds. How do banks deal with this
uncertainty?
The answer lies in what statisticians call the law of large numbers. When
applied to banking, the law of large numbers says that if individual with¬
drawal decisions are independent and the number of depositors is large, the
bank can determine very precisely how much it can afford to lend out and
still cover the withdrawals of its many depositors. One of the best ways to
grasp the law of large numbers is to illustrate what it implies about something
we all understand: flipping a coin.
When you flip a fair coin, there is a 50-50 chance it will come up heads.
If you flip a coin only once, you cannot be certain whether it will turn up
ItflSS mm * jpt mem
Figure 6.7
Profit-Maximizing Amount of Deposits: The Case of Market Power
Both in the Loan and Deposit Markets
To determine the profit-maximizing level of deposits, a bank with market power in both
the loan and deposit markets will equate the cost of obtaining an additional dollar of
deposits with the benefits of obtaining an additional dollar of deposits. The benefit of an
additional dollar of deposits is the marginal revenue product of deposits (MRPD), and the
cost is the marginal resource cost of deposits (MRCD). These are equal at point A, which
yields D* worth of deposits. The interest rate on D* deposits is given by the minimum
interest rate depositors are willing to accept to provide that quantity of deposits. This is
given by point B on the supply curve, 5, or i^.
heads or tails. However, if you flip a coin repeatedly, you will find that the
ratio of the number of heads to the total number of flips gets closer and
closer to the probability that any one toss will result in heads, which is 1/2
For instance, if you flip a coin 10 times, you may find that 4/10 of the flips
are heads. If you flip a coin 100 times, maybe 47/100 of them will be heads
If you flip a coin 1,000 times, you can be very confident that close to 1/2 oi
the flips will turn up heads. Even though the outcome of any one coin toss
is purely random, when the outcomes are averaged over a large number oi
tosses, the result can be predicted with great certainty.
U NCE RTAINTY AND BANK DEPOSITS 187
Similarly as the number of depositors at a bank increases, the bank can

predict with increasing accuracy the amount of withdrawals depositors will
make in a given period of time. To be sure, the amount withdrawn by a
single individual is uncertain, just as the outcome of a coin toss is. But when
the bank averages over a large number of depositors, this randomness van¬
ishes. The bank can then predict with great certainty the amount of reserves
it needs to cover withdrawals.
To see why, suppose there are N depositors in the bank, each with an
equal-size account. Thus, if the total deposits at the bank is D, any individual,
i, has Dj = D/N of these deposits in his or her name. If individual i withdraws
IT, dollars of deposits, the fraction of deposits withdrawn, denoted dh is
d = Wi=Wj_
' A D/N
Notice that different individuals withdraw different fractions of their depos¬

its; that is, df varies across different individuals in much the same way that
different tosses of a coin result in different outcomes.
If we divide both sides of the above expression by the total number of
Summary depositors (AO, we get
Exercise 6.1
_ 24 2 IT,
d = — = -1
N D
In other words, the average fraction of deposits withdrawn by all depositors

(id) equals the total withdrawals by all individuals (2IT,) divided by total
deposits at the bank (D).
If withdrawal decisions are independent of one another and the number
of depositors is large, the law of large numbers says that total withdrawals
as a fraction of total deposits is very close to the probability that one depos¬
itor will withdraw her or his deposits. For example, if the probability that
any one depositor will withdraw funds from the bank is p, then when N is
very large,
I Answer: 2 IT,-
D
This means that when a bank has a large number of depositors, it knows
total withdrawals as a fraction of total deposits will be very close to the
probability, p, that a single depositor will withdraw his or her funds from
the bank. This is true even though the bank does not know whether any
individual depositor will withdraw funds. For instance, if p = .5, there is a
50-50 chance that you will withdraw all of your deposits. Still, by keeping
one-half of all deposits as reserves and loaning out the rest, a bank with
many depositors can be reasonably sure of having enough reserves to cover
withdrawals by you and other depositors (see Box 6.2).
188 Chapter 6 The Bankas a Firm: Deposits
The Law of Large Numbers and

Deposit Withdrawals
A simple illustration will show how the law of deposits and the second will withdraw his is also
large numbers helps banks accurately predict .90 X .10 = 9 percent. Finally, the probability
withdrawals. We will consider the impact of that both will withdraw their deposits is .10 x
four different scenarios on the bank's uncer¬ .10=1 percent. Thus, the bank faces a situa¬
tainty about needed reserves. In the first, the tion in which it will have $1 million in deposits
bank has only a single depositor with $1 million. with 81 percent probability and $500,000 in de¬
In the second, the bank has two depositors, posits with 18 pecent probability. The bank will
each with $500,000. The third has four deposi¬ end up with no deposits only 1 percent of the
tors, each with $250,000. Finally, the bank has time. Thus, by merely having two depositors in¬
1,000 depositors, each with $1,000. In all four stead of one, the bank reduces the probability
cases, total deposits at the bank are $1 million of having no deposits from 10 to 1 percent.
and there is a 90 percent chance that each indi¬ With four depositors—the third scenario—
vidual depositor will leave deposits in the bank the number of possibilities rises. All four deposi¬
and a 10 percent chance that each will with¬ tors may keep their money in the bank, in which
draw all deposits. The withdrawal decisions of case the bank has $1 million in deposits. The
one depositor are assumed to be independent odds of this happening are 65.61 percent. The
of the decisions of other depositors. odds that three depositors will keep their money
What is the bank's reserve situation with in the bank and one will not, leaving deposits of
only one depositor? In this case, the bank $750,000, are 29.16 percent. The odds that two
knows there is a 90 percent chance that it will depositors will keep their money in the bank
have $1 million in deposits, but also a 10 per¬ and two will withdraw theirs, so that deposits
cent chance that it will have no deposits at all. total $500,000, are 4.86 percent. The odds that
This is a very risky situation for the bank. It is only one depositor will keep his or her money in
risky to make loans, since chances are 1 out of the bank, leaving deposits of $250,000, are
10 that the single depositor will withdraw all of 0.36 percent, and the odds of all four depositors
its deposits. withdrawing their deposits are 0.01 percent.
With two depositors, the bank is a bit bet¬ Thus, by having four depositors instead of one,
ter off. Now the first depositor will keep her de¬ the odds that the bank will have no deposits
posits in the bank with 90 percent probability, have shrunk from 1 in 10 (10 percent) to 1 in
as will the second, so the probability that both 10,000 (.01 percent). Furthermore, the odds of
will keep their deposits in the bank is .90 x .90 the bank having less than $500,000 in deposits
= 81 percent. The probability that the first will has fallen to .37 percent. Thus, the bank is fairly
withdraw her deposits and the second will not sure of having $500,000 or more in deposits.
withdraw his is .10 X .90 = 9 percent, and the
probability that the first will not withdraw her Continued on p. 189
Continued from p. 188 $850,000 in deposits, with only a small chance

Finally, what if there are 1,000 depositors, of having less than $875,000 in deposits. This is
each with $1,000 in deposits? In this case, the the impact of the law of large numbers on this
odds that the bank will end up with less than bank.
$500,000 in deposits are, for all practical pur¬ Of course, this example has ignored many
poses, zero. In fact, the odds that the bank will complicated real-world issues, such as the fact
end up with less than $800,000 in deposits are that different depositors have different probabil¬
also zero. The odds that the bank will end up ities of withdrawal and different sizes of depos¬
with less than $850,000 in deposits is a little less its and that few withdraw their entire accounts.
than 1 in 10 million! The odds that the bank will Still, the main point is that having a large num¬
end up with less than $875,000 in deposits is ber of depositors is a great aid to banks in pre¬
about 4 in 1,000. Thus, with 1,000 depositors, dicting the amount of reserves they need to
the bank is almost certain it will have more than hold.
Differences in Withdrawal Rates Among Households. Some

depositors are more likely to withdraw funds than others. For example, most
students end up with a zero balance in their checking accounts virtually
every month, while others tend to maintain large, positive balances. What
impact does this have on banks?
First, note that if banks have better information about the likelihood of
withdrawals, they can loan out more money. For example, the rate of with¬
drawals from savings accounts tends to be lower than the rate of withdrawals
from checking accounts. This means banks can lend out a larger fraction of
savings deposits relative to checking deposits and still have sufficient funds
in reserve to accommodate withdrawals.
To manage the size of its reserves efficiently, a bank must have infor¬
mation about how often depositors will withdraw funds. One of the more
common methods of obtaining this information is by sorting depositors.
Flow is this done?
Banks tend to offer depositors a variety of options. For those depositing
money in checking accounts, banks offer “no-fee” checking accounts, as
well as “no-minimum-balance” accounts. With a no-fee account, there is
no monthly fee, provided a minimum balance of, say, $1,000 is kept in the
account. If the balance falls below $1,000, however, the bank might impose
a hefty service charge ($10). In contrast, no-minimum-balance accounts
typically charge depositors a fee (say, $5 per month), regardless of the
account balance.
Given these options, individuals who usually end up with a zero balance
have an incentive to open a no-minimum-balance account, while those with
large balances have an incentive to open a “no-fee” account. Thus, individ¬
uals tend to sort themselves into different types of checking accounts ac¬
cording to the frequency with which they will drain their accounts. Banks
know that no-minimum-balance accounts are preferred by individuals who

tend to withdraw most of their deposits each month, and therefore they hold
a greater fraction of such deposits in reserve.
Similar sorting occurs with savings deposits. Banks offer regular savings
accounts, which pay a low interest rate but have no restrictions on the number
of withdrawals permitted per statement period. They offer a moderate interest
rate on accounts that limit the number of withdrawals or charge a penalty if
the savings balance falls below some specified amount. The highest interest
rates are paid on certificates of deposit (CDs). When a bank issues a certif¬
icate of deposit, it promises to pay a relatively high interest rate if the account
is kept for the specified period of time (usually between six months and five
years). However, it imposes a substantial penalty for early withdrawal.
Savings depositors sort themselves by selecting different types of savings
accounts. Depositors who deplete their savings accounts during a given time
period tend to deposit money in regular savings accounts to avoid the penalty
for early withdrawal they would end up paying if they deposited funds in a
CD and then withdrew them. Those who tend not to withdraw funds opt for
CDs. The bank thus can lend out a greater fraction of CD deposits than it
can regular savings deposits.
Recessions and Reserve Management. In applying the law of

large numbers to deposits, we assumed individuals’ withdrawals were in¬
dependent of one another. For instance, Mr. Jones’s decision to make a
withdrawal is independent of Ms. Smith’s decision.
Unfortunately, changes in the economic environment can lead to a sit¬
uation where many depositors decide to withdraw funds at the same time.
In a recession, for example, many depositors deplete their savings accounts
as they attempt to make ends meet. In severely depressed areas, this can
drain a bank of its reserves and, in some instances, leave the bank illiquid.
A bank is illiquid when it lacks enough liquid assets to meet the immediate
demands of its creditors or depositors.
How does a bank respond when its reserves are too low? First, the bank
may liquidate some of its assets. For example, it can sell some of its loans
or securities (such as Treasury bills) to another bank or to an individual
investor in return for cash. Since it takes time to sell assets, this practice is
of limited help to the bank when an individual depositor wants cash today.
Thus, banks continually monitor their reserves to determine whether they
need to begin selling off assets to generate additional reserves. In addition,
banks can borrow funds from other banks at what is known as the federal
funds rate, to obtain funds on a short-term (overnight) basis. Banks can also
borrow funds from the central bank, the Federal Reserve System. We cover
these latter two sources of funds in more detail in Chapter 14.
Bank Panics
A bank panic occurs when the failure of one bank to honor its deposits
leads the general public to fear that other banks will be unable to honor their
deposits. When this occurs, depositors attempt to withdraw their deposits

before their bank fails and in so doing may place it and other banks in
jeopardy.
Imagine that a bank did not have enough cash on hand to satisfy with¬
drawals made by depositors on a given day. Telling a depositor “Sorry, we
don’t have enough cash for you to withdraw your funds today” could well
set off a bank panic. Depositors will quickly line up at the bank, hoping to
be the first in line to withdraw their deposits on the next day. This, of course,
worsens the problem. As more and more depositors begin to withdraw funds,
the bank must sell more and more assets to obtain cash. In extreme cases, a
bank may be forced to sell assets at prices below their book value, leading
to insolvency. A bank is insolvent if its total liabilities exceed the value of
its assets. Even if the bank sold all its assets, it would not generate enough
funds to pay all those to whom it owed money, including depositors.
Why Banks Are Vulnerable to Panics. Two features of commer¬

cial banking make even otherwise healthy banks vulnerable to bank panics.
First, as we learned in Chapter 1, the banking system operates on a fractional
reserve system in which each bank maintains only a fraction of its deposits
in reserves. Thus, banks literally lack enough cash in the vault to immediately
cover all requests for withdrawals by depositors. Second, a large portion of
bank liabilities is in the form of deposits payable on demand. Thus, banks
face a potential calamity during a bank panic: They have promised their
depositors payment on demand, but if too many depositors take them up on
it, there is no way the banks can actually honor the promise. This condition
snowballs into a bank panic, since depositors of otherwise healthy banks
know of this danger and will want to be the first to withdraw their deposits
before the bank has depleted its reserves.
Of course, the bank has sources of funds other than vault cash. It can
call on its reserves on deposit at the Federal Reserve Bank. It may be able
to sell off some of its portfolio of liquid assets, such as government bonds,
to obtain the needed funds quickly. But in a banking panic many banks
scramble for funds, and this will likely drive down the market prices of
government bonds and other assets, further contributing to their problems.
This scramble for funds across the banking system and the resulting decline
in the prices of liquid securities may cause otherwise healthy banks to fail.
This further exacerbates the banking panic, as depositors at healthy banks
have reason to fear for the health of their banks after the decline in security
prices.
Why Bank Panics Pose an Economic Problem. The problem is

not the failure of a single insolvent bank. After all, thousands of poorly run
businesses go bankrupt each year. The real problem is that the failure of one
insolvent bank spreads to other banks in the banking system, and the scram¬
ble for funds among healthy banks may set off failures that would not
otherwise occur. Thus, bank panics are a problem because of their effect on
The Bank Panic of 1907
In October and November of 1907, the United new management. But this action was not suffi¬
States suffered a banking panic. The political re¬ cient to stop the bank panic. Instead depositors
sponse to this event led to the establishment of began withdrawing funds from other banks as
the Federal Reserve System, which began oper¬ the panic spread. By November 1907, banks in
ations in 1914. In late 1906 and early 1907, New York City had suspended cash payments
European investors who had previously invested on deposits. This suspension spread to other cit¬
heavily in the United States began withdrawing ies and did not end until early 1908. However,
funds, causing prices of U.S. stocks to fall the recession deepened during this time, and
sharply in the first half of 1907. By June 1907, bankruptcies of nonfinancial businesses in¬
the U.S. economy was in recession. creased. Both of these occurrences were blamed
The panic of 1907 began when the Mer¬ on the panic of 1907.
cantile National Bank in New York City suffered
Source: R. Alton Gilbert and Geoffrey E. Wood, "Coping
large losses and depositors began withdrawing
with Bank Failures: Some Lessons from the United States
funds. In October, other New York City banks and the United Kingdom," Federal Reserve Bank of
cooperated to help the Mercantile National Bank St. Louis Review, V. 68, December 1986, p. 5-14. Copy¬
by supplying it with funds after it was put under right © 1986. Used with permission.
otherwise healthy banks. They have an element of self-fulfilling prophecy,

since a mere rumor that certain banks are in ill health can lead to withdrawal
of deposits, thus actually causing the ill health of these and other banks.
Dealing with Bank Panics. The time-honored method for dealing

with bank panics is the suspension of cash payments of deposits (see Box
6.3). During a suspension of cash payments, banks do not pay cash for
deposits, but otherwise they conduct business as usual, clearing checks,
making loans, and accepting deposits. Tellers are allowed to resume pay¬
ments after the panic subsides. Suspending cash payments deals with one of
the two problems that lead to bank panics: Not having to pay cash on demand
frees banks from the scramble to obtain funds.
Another method of dealing with bank panics is for the central bank, the
Federal Reserve System in the United States, to act as the lender of last
resort. The lender of last resort provides cash loans to banks during a bank
panic. A bank can secure these loans with otherwise illiquid assets. An
otherwise healthy bank can thus get the cash it needs to satisfy its depositors’
U NCERTAINTY AND BANK DEPOSITS 193
demands. The main idea is that because the central bank stands willing to
lend cash to its banks if necessary, it assures the public that no healthy bank
will fail because of a panic begun by the failure of an unhealthy bank.
The Role of the FDIC. Today the existence of federal deposit insur¬
ance, which guarantees the full payment (up to $100,000) to owners of
certain deposit accounts at failed banks, greatly reduces the likelihood of a
bank panic. With deposit insurance, there is no reason to hurry even to an
unhealthy bank to withdraw your deposits (unless your account contains
more than $100,000), since the federal deposit insurance guarantees you will
recover the amount of your deposit even from a failed bank. Indeed, even
during the recent banking crisis, in which hundreds of banks and thrifts
failed in 1989 alone, no bank panic occurred. The reason is that federal
deposit insurance gives depositors little incentive to withdraw funds from
insolvent banks or savings and loans. One of the authors of this text calmly
watched as his insolvent savings and loan institution in a small Texas city
changed hands three or four times in as many years, and countless others
around the nation acted likewise.
Ironically, such calmness among depositors in the face of insolvency is
a cause for concern among policymakers. Without deposit insurance, depos¬
itors would shift funds away from banks and savings and loans that appeared
to be in danger of failure, and the fear of this happening would serve to
discipline the actions of banks and S&Ls. Moreover, when these banks or
savings and loans did get in trouble, the response of depositors would provide
for the shrinkage or even quick failure of the institutions, and thereby limit
the losses.
With deposit insurance, however, depositors are less likely to flee an
insolvent bank or savings and loan, and the institution keeps operating,
sometimes at ever growing losses. Then, by the time regulators get around
to closing or otherwise dealing with the insolvent institution, the costs of so
doing have skyrocketed. During the 1980s, many insolvent institutions of¬
fered the highest interest rates on deposits and, as we would expect, saw
their deposits increase. This was an attempt to grow out of their problems,
an effort that was made possible only by deposit insurance and by the slow
movement to close insolvent institutions. As often as not, such attempts led
to even greater losses and eventually to greater expense to the regulators and
to those ultimately responsible—the taxpayers.
Moral Hazard
The presence of deposit insurance reduces the likelihood of bank panics and
therefore lowers the risk to bank managers of massive withdrawals. Unfor¬
tunately, as we inferred earlier, it leads to another problem: moral hazard.
Moral hazard arises when one party in a transaction is subjected to the risk
(or hazard) that the other will take actions that are undesirable (immoral)
from the first party’s viewpoint. For instance, when a rental car company
sells you insurance on a rental car, it is subjected to moral hazard because
there is a chance you are the kind of customer that likes to “hot rod” in
rental cars—an action the company views as undesirable.
Flow does deposit insurance lead to moral hazard? Since depositors are
insured against the insolvency of a given bank, they are likely to spend less
time scrutinizing the bank’s health before opening a deposit account. This
means banks that make excessive risky loans may attract as many deposits
as those that make prudent loan decisions. As a result, the presence of federal
deposit insurance creates the hazard that people will deposit in banks that
make loans to poor credit risks. This can ultimately increase the number of
insolvent banks and, consequently, the insurance payments made by the
FDIC to depositors at those banks.
During the 1980s, many savings and loans—and some banks—were
allowed to continue operating even though they were insolvent. Regulators
failed to close these insolvent institutions for a number of reasons, including
intercessions on behalf of troubled institutions by some members of Con¬
gress. Of course, in the absence of deposit insurance, the depositors of
insolvent institutions would have withdrawn their funds and forced the in¬
stitutions to shut down. But armed with the security of deposit insurance,
depositors did not withdraw their funds and the insolvent institutions con¬
tinued to operate. As we first saw in Box 5.3 in the previous chapter, this
created perverse incentives. Insolvent institutions were tempted to invest in
ways that had the potential to generate large returns, even if those invest¬
ments were also very risky. For an insolvent institution, there is little down¬
side risk. The worst that can happen if an investment fails is that the insti¬
tution becomes “more insolvent” than before. As long as no laws are
violated, the managers and owners are not personally responsible for any
further losses, and, since the institution is already insolvent, the owners have
already lost their invested funds. The limited liability of corporations protects
these owners from losing more than their invested funds. Moreover, if the
risky investment succeeds, the institution may become solvent again, and
the owners may recover at least some of their investment. The only parties
standing to lose from this behavior are the deposit insurance funds and,
ultimately, the taxpayers. If the investment doesn’t succeed, the deposit
insurance fund will have to provide even more funds to reimburse depositors
when the institution finally closes.
To be concrete, suppose an insolvent institution has a negative net worth
of $100 million and is allowed to continue operating. It faces two investment
choices: a safe investment and a risky investment. The safe investment is
certain to net the bank $20 million, while the risky investment will net the
bank $200 million in profits or $160 million in losses, with equal probability.
On average, the risky investment will net the bank $20 million, the same
amount the safe investment will. But the bank is just as likely to lose $160
million as it is to strike it rich and earn $200 million.
Since the bank has a negative net worth of $100 million, the safe in¬
vestment will not earn the bank enough to pull it out of the hole. In contrast,
the risky investment offers a 50 percent chance of earning $200 million,
more than enough to put the bank in the black. But it is equally likely to put
the bank deeper in the whole: Its net worth will decline by $160 million if
the investment turns sour.
Given these alternatives, the insolvent institution has an incentive to
choose the risky investment. If this investment succeeds in earning the $200
million, the insolvent institution will be restored to solvency and will no
longer be a problem for the insurance fund. If the investment proves unsuc¬
cessful, the insolvent institution goes from a $100 million to a $260 million
liability to the insurance fund. Moreover, a real asymmetry exists here, since
the insurance fund will pay for the losses if things turn out bad but will not
receive an offsetting amount if things go well. In other words, the insurance
fund, on average, loses if this institution chooses the risky investment over
the safe one. The presence of deposit insurance creates moral hazard; it
allows savings and loans to gamble with other people’s money and keep the
winnings if they succeed. Investment decisions such as the one in this
example exacerbated the S&L crisis in the 1980s and caused deposit insur¬
ance funds to more carefully scrutinize investments made by insured finan¬
cial institutions. Box 6.4 shows the rise in the thrift and bank failures that
occurred during the 1980s.
Summary Suppose a bank has two types of depositors: A large depositor who has $10
Exercise 6.5 million at the bank and many small depositors with deposits totaling $10
million. Each depositor withdraws all funds from his or her account with a
probability of 1/4 and leaves all money in the account with a probability of
3/4. (a) What is the minimum amount of reserves the bank must keep to be
certain of having enough cash on hand to cover withdrawals? (b) What
strategy can the bank use to enable it to keep fewer reserves?
Answer: (a) Before we apply the law of large numbers, note that in this
case, one depositor contributes $10 million of the bank’s $20 million in
deposits, but many small depositors together have $ 10 million deposited in
the bank. Since not all depositors have equal-size deposits, we must deter¬
mine the reserves needed for each type of depositor.
There is a 25 percent chance that the large depositor will withdraw all
funds from the account. The only way to ensure having enough money to
cover such a withdrawal is to keep the entire $10 million deposit in reserve.
In contrast, we can apply the law of large numbers to the small depositors.
The law of large numbers implies that the total withdrawals of the small
depositors as a fraction of the deposits of all small depositors equals 1/4.
Thus, the bank must keep $10,000,000 X .25 = $2,500,000 in reserve to
be certain of having enough cash to cover withdrawals by small depositors.
Failures of Insured Thrifts

and Banks, 1934-1989
In 1933, the United States established the Fed¬ of insured savings and loans that failed reached
eral Deposit Insurance Corporation (FDIC) to in¬ 13 in 1941, when 2,343 insured savings and
sure bank deposits and the Federal Savings and loans existed.
Loan Insurance Corporation (FSLIC) to insure Compare this to the experience of the
savings and loan deposits. The accompanying 1980s. In 1988, 205 out of a total of 2,949 in¬
graph shows the number of failures of insured sured savings and loans failed. The assets of
banks and savings and loans from 1934 until these failed institutions totaled $10,660,000. A
1989, when the FSLIC was eliminated and the similar number of bank failures occurred—206
FDIC began administering the insurance fund for out of 12,712 insured banks. Assets of these
both banks and savings and loans. Notice that failed banks totaled $29,168,596.
the number of failures of insured banks during
the 1930s reached a peak of 77 in 1937, when Source: Congressional Budget Office, Reforming Federal
there were 13,797 insured banks. The number Deposit Insurance (September 1990), Tables C-1, C-2.
Key Terms 197
In total, then, the bank must keep $10,000,000 + $2,500,000 = $12,500,000

in reserves.
(b) The lion’s share of the reserves is due to the large depositor. One
strategy for reducing reserves is to induce the large depositor to purchase a
certificate of deposit. For example, suppose the bank offered the large de¬
positor a favorable interest rate on a one-year CD, with the stipulation that
the money can be withdrawn early only by forfeiting 10 percent of the initial
deposit. Then, even if the large depositor withdrew funds early, the bank
would have to give the depositor only 90 percent of the deposit. Persuading
the large depositor to do this would considerably reduce the bank’s necessary
reserves.
Conclusion
In this chapter, we learned how banks decide how many deposits to attract,
and how market forces determine the interest rate banks pay on deposit.
Combined with Chapter 5, in which we saw how banks decide how many
loans to make and at what rate, we have now analyzed the behavior of banks
in both the loan market and the deposit market. We also investigated in this
chapter several related issues concerning the management of reserves, the
importance of the law of large numbers for predicting withdrawals, the
problem of bank panics, and the role of deposit insurance as a solution.
Finally, we discussed the moral hazard problem implicit in the provision of
deposit insurance and gave an example showing how deposit insurance and
insolvency combined to create a severe moral hazard problem in the savings
and loan industry in the 1980s.
Our emphasis in this chapter was on the behavior of individual banks.
In the next chapter, we will look at the banking industry in more depth.
There we will see more historical details about this industry, including the
rules and regulations governing banks.
KEY TERMS
derived demand marginal revenue product of deposits (MRPD)
total product marginal resource cost of deposits (MRCD)
marginal product law of large numbers
law of diminishing marginal returns illiquid
required reserve ratio bank panic
excess reserve ratio insolvent
purely competitive deposit market federal deposit insurance
value marginal product of deposits (VMPD) moral hazard
198 Chapter 6 The Bank as'a Firm: Deposits

1. Why is the demand for deposits by a bank terest rate it pays on deposits? What
called a derived demand? would happen to the quantity of deposits
it attracts? Use a diagram to illustrate your
2. Tenth National Bank is a small bank in a
answer.
large city. How do you think Tenth Na¬
tional determines the interest rate to pay .
7 Stockholders at Bank One are concerned
depositors and to charge for loans? Ex¬ that management is not running the bank
plain. properly. The reason for their concern is
that deposits at the bank have doubled in
3. Is it possible for a bank to have market
recent years, but the amount of loans is¬
power in the deposit market but no market
sued has increased by only 30 percent.
power in the loan market? Justify your an¬
Are the stockholders’ concerns justified?
swer.
Explain.
4. Town Bank obtains deposits and issues 8. Any Bank obtains deposits and issues
loans in purely competitive markets where loans in purely competitive markets where
rD = 5 percent and rL = 10 percent. The rD = 5 percent and rL = 10 percent.
bank manager has learned from the bank’s (a) . Use a graph to illustrate how Any
research department that the marginal Bank determines the profit-maximizing
product of deposits is .2. level of deposits.
(a) . Explain what it means for the mar¬ (b) . Use a graph to show what would
ginal product of deposits to be .2. happen to Any Bank’s profit-maximizing
(b) . Based on these numbers, does an ad¬ level of deposits if the interest rate on de¬
ditional $1 in deposits increase loans by posits increased to 7 percent.
$1? Why or why not? (c) . Use a graph to show what would hap¬
(c) . Is Town Bank attracting the profit- pen to Any Bank’s profit-maximizing
maximizing amount of deposits? Explain level of deposits if the market interest rate
carefully. on loans increased to 13 percent.
.
5 Suppose the required reserve ratio is 15 .
9 Bertha Bank is the sole user of deposits
percent. Can the marginal product of de¬ and issuer of loans in its market.
posits ever be greater than 85 percent? (a) . Explain how the incentive to maxi¬
Explain. mize profits leads Bertha Bank to pay a
6. Major Bank is the sole supplier of loans positive interest rate on deposits, even
in its market area but competes in a na¬ though it is the only bank in the market.
tional market for deposits. (b) . “If the marginal productivity of de¬
(a) . Why might Major Bank be a monop¬ posits rises (due to a decline in the re¬
oly in the (local) loan market but a purely quired reserve ratio), Bertha Bank will
competitive bank in the deposit market? lower the interest rate it pays on depos¬
(b) . Use a diagram to show how Major its.” Is this statement true or false?
Bank determines the interest rate it pays 10. Explain why the marginal resource cost of
on deposits to attract the profit-maximizing deposits is greater than the interest rate
level of deposits. paid to depositors. (Assume, of course,
(c) . If Major Bank’s required reserve ratio that the bank has market power in the de¬
decreased, what would happen to the in¬ posit market.)
Q U ESTIONS AND PROBLEMS 199
11. Determine the likely impact of the follow¬ 16. Central Bank offers 8 different types of
ing on an individual bank’s value mar¬ checking accounts and 24 different types
ginal product of deposits. of savings accounts. Why doesn’t Central
(a). An increase in the market supply of Bank reduce its bureaucracy by offering
deposits only one type of account?
(b>. An increase in the market demand for 17. Suppose a bank has market power in the
deposits deposit market. The law of supply reveals
(c) . A reduction in the bank’s required re¬ that if the bank reduces the interest rate it
serve ratio pays on deposits, it will attract fewer de¬
posits. What does the economic principle
(d) . A reduction in the fraction of bor¬
of adverse selection suggest about which
rowers who default
depositors will continue to leave their
12. Suppose a bank has market power in the money in the bank? Explain carefully.
loan market. The bank can produce loans
18. Some banks do not pay interest on depos¬
according to a total product curve, L =
its and, in fact, charge depositors a
(1 — rr) D, where rr is the required re¬
monthly maintenance fee, plus 10 cents
serve ratio.
for each check written. Does this policy
(a) . Does this bank’s marginal product contradict the analysis in this chapter?
curve obey the law of diminishing mar¬ Explain.
ginal returns?
19. Does the presence of federal deposit in¬
(b) . Graph the bank’s marginal revenue
surance lead to moral hazard in the bank¬
product curve for deposits, and explain
ing industry? Explain.
why it looks as it does.
20. Suzie has never written a bad check in her
13. Can a solvent bank be so short on cash
30 years as a customer at Friendly Bank.
that it cannot accommodate withdrawals?
The bank manager recently called Suzie to
Explain.
inform her that she has been given “auto¬
14. Carefully explain why banks need not matic overdraft privileges.” This is a
keep all deposits in reserve to cover with¬ fancy way of saying that the bank will auto¬
drawals by depositors. matically transfer funds from Suzie’s sav¬
ings account to her checking account to
15. A bank with market power in both the de¬
cover any checking account overdrafts.
posit and loan markets is experiencing
losses. The bank plans to reduce the rate it (a) , “Automatic overdraft privileges are
pays to depositors while increasing the worthless to Suzie, since she never writes
rate it charges on loans. bad checks.” Is this statement true or
(a) . Ignoring the consequences of “im¬ false? Explain.
perfect information” in the loan and de¬ (b) . Do you think automatic overdraft
posit markets, will the bank’s profits nec¬ privileges lead to an increase in the num¬
essarily increase if it adopts this action? ber of checks bounced at a bank? (Re¬
Explain carefully. member: For a check to bounce with auto¬
(b) . How would the presence of imperfect matic overdraft privileges, it must exceed
information alter your answer in part? Ex¬ the amount in both the savings and check¬
plain carefully. ing accounts.) Explain.
Selections for Further Reading .
Boughton, J. M., and E. R. Wicker. “The Behavior of Froyen, R. T., and K. J. Kopecky. “A Note on Re¬
the Currency Deposit Ratio during the Great De¬ serve Requirement and Monetary Control with a
pression.” Journal of Money, Credit, and Banking, Flexible Deposit Rate.” Journal of Banking and
11 (November 1979), 405-418. Finance, 7 (March 1983), 101-109.
Day, A. E. “Reserve Ratios: A Proposal for McCulloch, J. H. “Bank Regulation and Deposit In¬
Change.” Journal of Macroeconomics, 8 (Fall surance.” Journal of Business, 59 (January 1986),
1986), 479-484. 79-85.
Edmister, R. O. “Margin Analysis for Consumer De¬ Steindl, F. G., and M. D. Weinrobe. “Natural Haz¬
posit Interest Rate Policy.” Journal of Bank Re¬ ards and Deposit Behavior at Financial Institutions:
search, 13 (Autumn 1982), 179-184. A Note.” Journal of Banking and Finance, 7
Edmister, R. O., and H. E. Merriken. “Measuring In¬ (March 1983), 111-118.
terest Rate Sensitivity of Consumer Depositors.” Wells, D. R. “Security Reserve Requirements—A
Journal of Financial Services Research, 2 (June Possible Substitute for Deposit Insurance?” Jour¬
1989), 133-145. nal of Economics, 14 (1988), 76-83.
CHAPTER
The Banking Industry
n the previous two chapters, we analyzed the economic decisions made

by individual banks. In this chapter we broadly survey the banking industry,
which is made up of numerous individual banks as well as savings and loans
and credit unions.
In the United States and other industrialized nations, the government
actively regulates the banking sector. In particular, the banking industry is
subject to rules and regulations imposed by various state and federal regu¬
latory agencies, including the Federal Reserve System, state banking com¬
missions, the comptroller of the currency, and the Federal Deposit Insurance
Corporation. State and federal legislative bodies also affect the banking
industry.
In the first part of this chapter, we examine the evolution of the banking
industry. We then turn to an economic analysis of the many rules and
regulations supervisory agencies impose on the banking industry. As we will
see, some past regulations had surprisingly adverse effects on the industry,
giving regulators and banks powerful incentives to evolve and innovate to
overcome them.
Dual Banking
We learned in Chapter 1 that money and banks existed in very early societies.
While banks initially provided a safe place to store gold, bankers soon
realized that most of the gold in their vaults was never withdrawn. Eventually
banks began to issue bank notes, which were an early form of paper money.
Prior to 1863, each bank issued its own bank notes, which could be
redeemed on demand for gold. Although banks were regulated by the states
in which they operated, it was relatively easy for unscrupulous individuals
to defraud banks and merchants by counterfeiting bank notes. Because each
bank’s bank notes differed in appearance from those of other banks, it was
difficult to determine whether a particular bank note was genuine or coun¬
terfeit. Some banks issued bank notes that exceeded the value of the assets
the banks held, while others defaulted on the bonds backing their bank notes.
As a result of these and other problems, many state-chartered banks failed.
The National Banking Act of 1863 was passed in an attempt to eliminate
the problems associated with the state banking system and individual banks’
201
202 Chapter 7 The Banking Industry
bank notes. Under the new banking system this act created, the federal
government issued charters to certain banks. These banks, known as national
banks, still exist today under the regulatory supervision of the U.S. Treasury
(more specifically, the comptroller of the currency). The intent of the act
was to entirely eliminate state banks by imposing heavy taxes on the bank
notes they issued. The bank notes of national banks, however, could be used
as currency without having to pay such taxes, thus putting state banks at a
competitive disadvantage. But the state banks quickly countered by creating
a close substitute for currency, called demand deposits, which were the
forerunners of the checks you use today. This innovation not only allowed
state banks to survive but led national banks to adopt demand deposits as
well.
As a consequence, today we have a dual banking system wherein state
and national banks coexist. The comptroller of the currency, an office of the
Treasury, grants charters to national banks. State banking commissions in
each state grant charters to state banks. Currently about two-thirds of banks
have state charters and the remainder have national charters.
Acquiring a Bank Charter
Opening a bank involves considerable red tape and financial capital. As we

saw in Chapter 5, this condition restricts free entry and may reduce com¬
petition in the banking industry. To form a bank, the parties involved must
obtain approval from the relevant regulatory body (the comptroller of the
currency for a national bank and the state banking commission for a state
bank). The process of obtaining such approval is known as acquiring a bank
charter.
To obtain a charter, the potential owners of the bank must prove they
have not only the required financial capital but also the qualifications needed
to operate the bank. In addition, they must show that the area in which they
propose to locate both needs and can support a new bank. If the chartering
agency believes the proposed bank has insufficient capital, the applicants are
not qualified, or the area does not need another bank, it will not approve the
charter. Showing need is very subjective. One time-honored approach is to
identify a geographic area or an interest group that is not being adequately
served by existing banks and promising to serve that area’s or group’s
banking needs. Box 7.1 discusses the criteria used by the comptroller of the
currency in considering applications for national bank charters.
Supervision and Examination of Banks
Once a bank charter is granted, it is supervised and examined from time to

time by various agencies, including the Federal Reserve System, the Federal
Requirements for Obtaining

a Bank Charter
To illustrate the reasoning behind decisions by The second decision is a denial of a pro¬
the comptroller of the currency to approve or posed charter in Arkansas due to intense com¬
deny a bank charter, we reproduce here three petition already present in the relevant market
decisions made by that office as reported in the and an inadequate organizational plan:
Quarterly Journal. These decisions highlight the
On August 19, 1988, the Office denied a
importance of demonstrating a need for the
charter application in Arkansas because of a
proposed bank, an adequate organizational
weak and unrealistic operating plan and a
plan, and evidence of the qualifications of the
weak organizing group. The organizers did
organizers and proposed bank officers.
not demonstrate that they possessed the
The first decision is an approval of a bank
qualifications to succeed in the market area
charter in California. The approval is conditional
where the economy was weak and compe¬
on a few items; it explicitly states that two of
tition was intense (page 15).
the proposed executive officers are not accept¬
able: The final decision explicitly mentions that
the bank will service a particular interest group
On August 4, 1988, the Office approved a (recall this is one way to establish a need for a
new bank charter in California on the con¬ new bank):
dition that two of the organizers be prohib¬
United Citizens Bank, National Association,
ited from serving as executive officers of the
Los Angeles, California was approved by the
proposed institution. The individuals were Office on June 30, 1988. The new bank
involved with a company that had recently
was established in order to serve the Ko¬
emerged from Chapter XI bankruptcy and it
rean community in Los Angeles. The Office
was the opinion of the Office that manage¬
noted that the proposed chief executive of¬
ment of the company would require an in¬
ficer had strong credentials, the organizers
ordinate amount of the individuals' time. As had very strong ties to the targeted com¬
such, they would be unable to devote the
munity, and that similar new banks had
necessary amount of time to the proposed been successful in the recent past (page
bank to ensure that it would have a reason¬ 29).
able likelihood of success. The remaining
organizers were considered excellent and
Source: Quarterly Journal of the Comptroller of the Cur¬
the proposed chief executive officer pos¬ rency, Administrator of National Banks, (December 1988),
sessed good credentials (page 15). pp. 15, 29.
Deposit Insurance Corporation (FDIC), the comptroller of the currency, and

state banking authorities. Bank examinations typically focus on the status of
the bank’s loan portfolio and the bank’s capitalization.
Status of the Loan Portfolio

One important purpose of supervision and examination is to reduce the
hazards of asymmetric information between bank managers and the public
regarding the riskiness of the bank’s loan portfolio and thus assure depositors
that the bank is a safe place to deposit funds. To this end, bank regulations
require the examining agency to evaluate the quality of the bank’s loan
portfolio and verify that the bank has sufficient assets and reserves to cover
its deposits. This monitoring helps alleviate depositors’ concern about such
things as the quality of loans issued by the bank.
Quality loans are loans with a high probability of being repaid. For
example, suppose your bank claims to have $100 million in assets. It would
be difficult (and costly) for you to verify this information, but a bank ex¬
aminer does this for you. He or she examines the portfolio to determine
whether the assets really are worth $100 million. If the assets consist solely
of idle cash, the true value of the bank’s assets is clear. However, since
banks use reserves to issue loans, most of the bank’s assets will be in the
form of loans rather than cash.
In fact, two-thirds of the average bank’s assets consist of loans. The true
value of these loans depends on whether the borrowers are making their
interest and principal payments on time (that is, on whether borrowers are
paying off their debt). For instance, the value of a car loan on the books
might be $10,000 even if the loan is in default. If the bank is forced to
repossess the car and can sell it for only $6,000, the true value of the loan
is $6,000, and the bank has overstated the loan’s value by $4,000. A bank
examiner, however, can require the bank to value such assets properly.
When a loan is in default, the bank examiner may require the bank to
either write off the loan—that is, not include it as an asset—or liquidate the
asset. Doing so reduces the bank’s net worth, or capital. If this happens too
often, the bank’s capital may fail to satisfy minimum capital regulations. A
bank in this situation (and any bank that is caught manipulating the books
to hide information from the examiner) is classified as a “problem bank.”
Such a bank will be very heavily scrutinized in the future, and its officers
may even face penalties and legal actions.
Bank Capitalization
As noted earlier, sufficient capital is required to obtain a charter and to
continue operating a bank. Bank capital is the equity of stockholders; it
serves as a buffer against bank insolvency when losses on loans or other
securities occur. Prior to the 1980s no official capital requirements existed,
but the various supervisory and regulatory agencies used informal means to
encourage banks to maintain what they deemed a reasonable level of capi¬

talization. Bank examiners began noticing that capitalization rates were fall¬
ing in the late 1970s and early 1980s, in part due to bank losses on inter¬
national loans to developing countries. The International Lending
Supervisory Act of 1983 authorized the Federal Reserve System, the FDIC,
and the comptroller of the currency to set and enforce minimum capital
requirements in an effort to stem the international debt crisis.
Under this act, regulatory capital limits were first put in place in 1985.
The results were disappointing, however, and bank capital to asset ratios
were still low by the late 1980s. In response, a whole new structure of capital
requirements was imposed. These regulations applied to both core capital
(stockholder equity in the bank) and supplemental capital (loan loss reserves
up to 1.25 percent of the assets and subordinated debt issued by the bank).
Subordinated debt is debt that is sold subject to the condition that, in the
event of insolvency, all other liability holders have precedence. Thus, sub¬
ordinated debt is in many ways similar to shares of stock, except the holder
receives a stated interest payment instead of a share of residual profits.
In addition to imposing regulations on core and total (core plus sup¬
plemental) capital to asset ratios, the new regulations required examiners
to weight assets by risk. Cash, U.S. government securities, and GNMA
mortgage-backed securities were considered safe assets, and no risk adjust¬
ment occurred. Interbank deposits, general-obligation municipal securities,
and FNMA and FHLMC mortgage-backed securities were considered to
have a small default risk and were assigned a risk weight of 20 percent.
First-home mortgages and municipal revenue bonds were assigned a risk
weight of 50 percent. Finally, all other bank securities and loans were as¬
signed a risk weight of 100 percent. Currently the risk-adjusted ratio of total
capital to assets must be at least 8 percent.
The response of many banks to these requirements was to increase their
capitalization by issuing more equity shares and more subordinated debt.
This increased these banks’ stake in their own fate, reducing to some extent
their incentive to gamble by making risky loans. Unfortunately for stock¬
holders, the large number of new issues of equity by banks may be respon¬
sible for the declines in prices of bank stocks in the early 1990s.
Overlapping Responsibilities
We have mentioned a host of agencies involved in chartering banks, super¬
vising banks’ operations, and examining banks’ loan portfolios and capital¬
ization. There is considerable overlap of responsibilities, as banks are subject
to multiple layers of supervision and examination among agencies. In prac¬
tice, however, these agencies divide their duties as follows. The comptroller
of the currency charters, supervises, and examines all national banks. State
banking commissions charter state banks, and the Federal Reserve System
supervises and examines state banks that are members of the Fed. Similarly,
the FDIC supervises and examines state banks that are members of the FDIC
but are not part of the Federal Reserve System. State banking commissions
supervise and examine banks that are neither Federal Reserve System nor
FDIC members. Table 7.1 summarizes these responsibilities. In the next two
sections, we look at how and why the supervisory powers of the Federal
Reserve System and the FDIC have evolved over time.
The Federal Reserve System

Congress established the Federal Reserve System (the Fed) when it passed
the Federal Reserve Act in 1913. The Fed was given the task of promoting
stability in the banking industry and serving as the sole supplier of U.S.
currency. This currency is now known as Federal Reserve notes—the dollar
bills in your pocket or purse. Every national bank was required to be a
member of the Federal Reserve System and to honor the Fed’s rules and
regulations. One benefit of membership was Fed services such as check
clearing. State banks, on the other hand, could join the Federal Reserve
System and enjoy similar services if they paid the required membership fee,
but they were not obligated to do so.
Over time, shrinking membership reduced the Fed’s influence on the
banking system. Between the early 1950s and the 1980s, assets of member
banks declined from 85 percent to about 50 percent of total banking system
assets. Much of this decline occurred during the 1970s—when interest rates
M Billl
Table 7.1
Responsibilities of Commercial Bank Regulatory Bodies
This table summarizes the responsibilities of the four bodies that regulate commercial banks in the
United States.
Supervised and
Type of Bank Chartered by Examined by Insured by
National banks Comptroller of the currency Comptroller of the currency FDIC

State banks that are members State banking commissions Federal Reserve System FDIC
of the Fed
State banks insured by the State banking commissions FDIC FDIC
FDIC (nonmembers of the
Fed)
State banks not insured by State banking commissions State banking commissions State authorities
the FDIC
The FDIC 207
were high—because banks actually profited by avoiding or giving up Fed

membership. The Federal Reserve System not only charged a membership
fee but imposed tougher reserve requirements and other regulations than did
the various state banking agencies. For instance, the Fed did not pay interest
on member banks’ reserves, whereas many states allowed state-chartered
banks to hold reserves in interest-earning securities. Consequently banks
found it profitable to obtain state charters and opt against Fed membership.
This ultimately reduced the Fed’s control over the banking system.
This situation changed in 1980 with the passage of the Depository
Institution Deregulation and Monetary Control Act (DIDMCA). While
DIDMCA contained other provisions that we will discuss later in this chap¬
ter, a key provision enhanced the Fed’s ability to stabilize the banking
system. This provision created uniform reserve requirements that essen¬
tially required all depository institutions to meet Federal Reserve System
reserve requirements in exchange for enjoying the privileges of membership.
Specifically the act stated that all depository institutions were subject to the
same set of reserve requirements, which would not earn interest, and gave
the Federal Reserve System authority to set supplemental reserve require¬
ments, which would pay interest. In exchange, all depository institutions
were permitted to borrow from the Federal Reserve System (called using the
discount window), and all could avail themselves of Fed services such as
check clearing and wire transfers. However, these services, which had been
free, would now be priced by the Federal Reserve System at cost. As a result
of DIDMCA, the distinction between members and nonmembers has blurred,
since both are now subject to many of the same regulations. In Chapter 13,
we discuss the Fed in more detail.
The FDIC
The Federal Deposit Insurance Corporation (FDIC) emerged as a result
of the Fed’s inability (or unwillingness) to stabilize the banking system
during the Great Depression. When banks began getting into trouble, the
Fed failed to serve as a lender of last resort and in fact raised the rate it
charged banks for emergency funds. Ultimately one-third of all U.S. banks
failed, and many depositors were left penniless. This crisis culminated with
the Bank Holiday in March 1933, when President Roosevelt closed all U.S.
banks for one week as a temporary measure to calm the panic. Longer-term
measures were instituted when Congress passed the Banking Act of 1933
(better known as the Glass-Steagall Act). Among other things, this legislation
established the FDIC to insure depositors against bank failures. Today every
national and state bank that is a member of the Federal Reserve System is
required to purchase FDIC insurance to cover depositors. State banks that
are not members of the Fed may choose not to purchase FDIC insurance,
but membership in the FDIC is almost universal. Figure 7.1 illustrates the
extent to which banks have a choice in purchasing FDIC insurance depending

on their charter and membership in the Fed.
Today the FDIC insures depositors up to $100,000 against loss due to
bank failure. In return for this insurance, a bank is required to pay an
insurance premium and to adhere to rules and regulations on deposits and
loans established by the FDIC. The FDIC currently charges .23 percent of
the dollar value of bank deposits. Thus, if you open up an account at your
bank for $100, your bank will pay 23 cents to the FDIC to insure your
deposit.
In addition to insuring depositors against loss, deposit insurance reduces
the likelihood of bank panics. To see why, suppose depositors at a bank
without deposit insurance heard a rumor that their bank was about to fail.
Some depositors would rush to the bank to withdraw their deposits before
the bank ran out of funds. Since the bank keeps only a small fraction of
deposits as excess reserves, these reserves would become depleted as more
and more depositors withdrew funds. Eventually the bank would lack enough
liquid assets to cover any further withdrawals. This situation could result in
a bank panic, in which virtually every depositor would try to withdraw
TheFDIC 209
deposits. This, of course, would exacerbate the problem, because the bank
would have to sell off illiquid assets at a very low price to obtain cash.
Deposit insurance reduces the likelihood of such an occurrence. In par¬
ticular, since depositors know they will receive their deposits even if the
bank fails (thanks to the insurance), they have a greatly reduced incentive
to withdraw their funds if they hear a rumor that the bank is on the verge of
failure. The FDIC will pay them the full value of their deposits in such an
event. Consequently, bank panics are much less likely to occur when banks
have deposit insurance.
During the past decade, however, many banks that were members of the
FDIC failed because of the moral hazard problems created by deposit insur¬
ance (see Box 6.4 in the previous chapter). In 1989, at the height of the
trend, more than 200 banks failed in the United States. In an attempt to cover
the costs of the claims paid to depositors who otherwise would have lost
their deposits, new proposals were pending in 1994 that would increase the
cost to banks of FDIC insurance. Under the new plan, the cost to a bank of
deposit insurance would rise to between 25 and 30 cents per $100 in deposits
depending on the bank’s financial health.
In the same way the FDIC insures depositors at commercial banks, the
Federal Savings and Loan Insurance Corporation (FSLIC) insured depositors
at savings and loans, but it went out of existence in 1989 as a result of the
S&L crisis. Today the FDIC is the insurance agency for both commercial
banks and savings institutions.1 In the next section, we briefly examine the
crisis that caused the FDIC to take over the responsibility for insuring S&Ls
and discuss some recent legislation directed at improving the FDIC.
The S&L Crisis and Changes in the FDIC

The S&L crisis began with large losses in the thrift industry during the 1970s
and 1980s. These losses were due in part to the high nominal interest rates
at the time, which left many S&Ls paying more on deposits than they were
receiving on (relatively old) mortgages. People who bought homes in the
late 1960s were still paying 7 percent on their mortgages when their deposits
at S&Ls were earning more than 10 percent. This created perverse incentives
for S&Ls to gamble by making risky investments in the hope of big payoffs
(see Box 5.3 in Chapter 5, “Going for Broke in the Savings and Loan
Industry”). But the big payoffs never came. Rather, the S&Ls incurred big
losses, and in the end the volume of failed S&Ls left the FSLIC with too
few funds to cover their deposits. The result was the S&L “bailout,” in
which Congress picked up the tab at an estimated cost at the time of $680
for every person in the United States.
A different agency, the National Credit Union Administration (NCUA), administers an insurance
fund for credit unions. This fund is the National Credit Union Shareholders Insurance Fund
(NCUSIF), and insures credit union depositors (called shareholders) up to $100,000.
FIRREA. Congress abolished the bankrupt FSLIC and restructured the

FDIC to insure depositors at both commercial banks and savings institutions
when it passed the Financial Institutions Reform, Recovery, and Enforcement
Act of 1989 (FIRREA). The FDIC now administers separate funds for the
two types of institution. The fund for banks is the Bank Insurance Fund
(BIF), while the Savings Association Insurance Fund (SAIF) insures depos¬
itors at S&Ls. FIRREA also abolished the Federal Home Loan Bank Board
(FHLBB), which had regulated the savings and loan industry. To replace
the FHLBB, this act created the Office of Thrift Supervision (OTS) at the
U.S. Treasury.
FDICIA. More recently, Congress took steps to reduce the moral hazard
problems that bankrupted the LSLIC by passing the Federal Deposit Insur¬
ance Corporation Improvement Act (FDICIA) in 1991. This legislation re¬
quired the LDIC to initiate proposals for risk-based insurance premiums by
1994, meaning that institutions with relatively risky loan portfolios will have
to pay higher rates for deposit insurance than those with safer ones. This act
will give banks stronger incentives to manage risk to avoid higher premiums.
The act also mandated the LDIC to resolve problems at troubled institutions
at the least possible long-term cost to the insurance fund. To do this, the
LDIC must itself monitor banks and bank examiners. These incentive effects
on both banks and the FDIC will reduce the potential for asymmetric infor¬
mation and moral hazard problems to bankrupt the FDIC.
Summary You are considering opening an account in either First National Bank or
Exercise 7.1 First State Bank in your local city. Briefly explain the differences between
these two banks. What specific information should you request before de¬
positing funds in First State Bank? Explain.
Answer: Lirst National, being a national bank, was chartered by the U.S.
Treasury (the office of the comptroller of the currency). The regulations that
guided its creation were established at the federal government level. In
contrast, rules established by the state banking commission guided the char¬
tering of Lirst State Bank. Lirst National is necessarily a member of both
the Lederal Reserve System and the LDIC. Your deposit (up to $100,000)
in Lirst National would be insured against loss by bank failure (or by theft).
In contrast, Lirst State Bank is not necessarily a member of the Lederal
Reserve, nor does it necessarily have FDIC deposit insurance. You should
ask First State Bank whether it is a member of the Federal Reserve System.
If the answer is yes, it is also a member of the FDIC, and your deposit in
the bank will be insured. If it is not a member of the Federal Reserve, it
may still be a member of the FDIC. By all means, find out. Otherwise you
might end up like the grandmother of one of the authors: She lost $25,000
at a depository institution that didn’t have FDIC coverage.
An Overview of Banking Regulation 211
An Overview of Banking Regulation

We have seen that a host of regulatory bodies oversee the operations of the
more than 11,000 commercial banks in the United States. The large number
of banks, coupled with the numerous regulatory agencies and the many
changes made in banking regulations each year, makes for a very confusing
regulatory system. Not only is it difficult for the average citizen to understand
the operation of the banking system, but even banks struggle to fully under¬
stand the nature and implications of regulations on their operations. A report
by the Federal Financial Institutions Examination Council (which consists
of regulatory agencies that include the Federal Reserve System, the comp¬
troller of the currency, and the Federal Deposit Insurance Corporation) es¬
timates that complying with banking regulations costs commercial banks
and thrift institutions (savings and loans, credit unions, and mutual savings
banks) anywhere from $7.5 billion to $17 billion each year. This includes
the costs of paperwork, legal fees, and other costs of conforming to govern¬
ment mandates. As a result of the report, as of 1994 the Clinton administra¬
tion was considering proposals to ease the regulatory burden on the banking
industry.
Regulation of the banking industry aims to achieve several goals. The
first is maintaining bank solvency or, alternatively, limiting bank failures.
As we saw in Chapters 5 and 6, the presence of asymmetric information
makes it difficult for the average depositor to know how safe his or her
deposits are at a bank, and the mere rumor that a bank is on the verge of
failure could cause a bank panic. Proponents of regulation argue that gov¬
ernment’s presence to regulate and monitor banks bolsters depositor confi¬
dence, thus enhancing the stability of the banking system.
Another goal of regulation is to ensure liquidity of the banking industry,
making sure that banks can honor their promise to redeem demand deposits
for currency on demand. As we saw in the previous chapter, the fact that the
Fed is the lender of last resort and the FDIC insures depositors helps banks
honor their promises to depositors.
The third goal is to promote economic efficiency, which includes pro¬
viding deposit and loan services to people in local communities at compet¬
itive interest rates. Regulators have enacted myriad regulations designed to
foster such competition by keeping banks from growing too big and gaining
monopoly power in loan and deposit markets.
The banking system we know today has been shaped directly and indi¬
rectly by legislation designed to achieve one or more of these goals of
regulation. As we will see in the next few sections, however, some pieces
of legislation did not have their intended effect, and as a result multiple
layers of new legislation have been added over the years to deal with the
problems they created. Furthermore, we will see that banks have successfully
gotten around many of the regulations through financial innovations and
loopholes in the laws.
How Branching Restrictions Shaped the Banking Industry

A distinguishing feature of the U.S. banking system is the large number of
small, local banks. This is the result of a long history of regulations that
restricted banks from freely opening branches within states and across the
nation. Imagine a law that allowed a shoe manufacturer to sell shoes only
to consumers who lived in the same city as its plant! This is, in effect, what
branching restrictions in many states required banks to do.
Branching restrictions grew out of fear that in their absence, banks would
grow progressively larger and acquire excessive market power, making them
less sympathetic to the needs of people and businesses in local communities.
Consequently, numerous banks were formed across the states to service the
needs of local communities. Many states (primarily those in the central
United States) limited the ability of state-chartered banks to branch within
their states’ boundaries. This practice, called limited branching, restricted
the number of branches a bank could have within a state. Some states took
this to an extreme, enacting unit banking laws that allowed a bank to serve
only a single location in its state. However, not all states limited the ability
of their banks to open branches within their boundaries. Primarily along the
West Coast, some states allowed statewide branching, which permitted
banks to open branches throughout the state.
Initially national banks were not subject to the branching restrictions
imposed by the states, and as a result they had an advantage over state banks.
To level the playing field, Congress passed the McFadden Act of 1927,
which made national banks subject to the same branching restrictions im¬
posed by state regulators on state-chartered banks. Since most states had
restrictions against branching, neither state nor national banks could freely
establish branches across state lines. As a result, numerous small banks
emerged across the country as individual states enacted laws that prevented
entry by out-of-state banks. Box 7.2 contrasts this trend with the current
move toward free trade within the boundaries of the European Community.
Over time, however, two financial innovations occurred that allowed
banks to get around these laws to better serve their customers: bank holding
companies and electronic banking. These innovations, along with recent
changes in state laws that grew partly out of the S&L crisis of the 1980s,
are discussed next.
Bank Holding Companies

Bank holding companies, such as Citicorp (which owns Citibank) and
Mellon Bank Corp. (which owns Mellon Bank), are corporations that own
the controlling interest in one or more banks. A bank holding company may
also own other companies that are in businesses related to banking, such as
credit card services and life insurance. Some bank holding companies, known
as nonbank banks, do not provide the full array of services offered by full-
service banks. For example, a nonbank bank might have an office that accepts
How Branching Restrictions Shaped the Banking Industry 213
I
Banking in the European Community and
the Principle of Mutual Recognition
Recently Gerald P. O'Driscoll, Jr., economist and add additional regulations. Under the principle
vice president at the Federal Reserve Bank of of mutual recognition, though, a bank chartered
Dallas, wrote about the lack of free trade within in Texas would be free to operate in New York
the United States. This is particularly a problem or California while obeying the state regulations
in services and financial markets, including of Texas even if those regulations differed from
banking. O'Driscoll contrasted this situation with those of New York or California and even if they
the move toward free trade within the bounda¬ gave the Texas bank an "unfair" advantage.
ries of the European Community (EC). What are the implications of adopting this
A guiding principle in Europe is that of mu¬ principle of mutual recognition? First, it will cer¬
tual recognition. This principle stipulates that tainly create a single market, with free entry
each member state will recognize the rules and across the 50 states. No longer will states be
regulations of other member states. The Euro¬ able to keep out banks chartered in other states
pean Community will establish some minimum or even to regulate them. Second, state banking
criteria that all member states must adopt. rules and regulations will tend to evolve toward
Other than those, each member state will make those of the most liberal state, subject only to
its own rules and regulations, and these will be the federal minimum standards. This will occur
recognized as valid by all other member states. as banks naturally seek charters from the state
What exactly does this mean? It means that that provides the most favorable regulations.
a bank established under the rule of the Nether¬ This is exactly what happened in the United
lands is free to operate in Germany under the States for corporate charters. Delaware is the
rules that apply in the Netherlands as long as it home of a disproportionate number of corpora¬
obeys the overall EC rules. Moreover, this ap¬ tions because it provided more favorable treat¬
plies even if the German banks are disadvan¬ ment of corporations.
taged by more favorable banking rules in the
Netherlands.
Source: O'Driscoll, Jr., Gerald P., What About Free Trade
In the United States, federal banking guide¬
Within the United States?, from Southwest Economy,
lines apply to banks in all states. However, after January/February 1992, 1-5. Copyright© 1992. Used
meeting these requirements, states are free to with permission.
HI
deposits but does not make loans and another office that makes loans but
does not accept deposits.
Bank holding companies evolved as a financial innovation to enable
banks to circumvent branching restrictions. By acquiring the controlling
interest of banks located in different regions of the country, a bank holding
company could effectively offer services to customers in different regions
or states without violating branching laws. Congress responded to this in¬

novation by passing the Bank Holding Company Act of1956, which restricted
bank holding companies from acquiring banks in other states without the
consent of those states. Further innovations occurred due to a loophole in
the law that allowed one-bank holding companies to cross state lines. Con¬
gress closed this loophole when it passed the Bank Holding Company Act
of 1970, which extended restrictions on bank holding companies to one-
bank holding companies. Despite these attempts to limit bank holding com¬
pany acquisitions, bank holding companies today indirectly hold (through
their subsidiaries) more than 90 percent of all bank deposits.
Electronic Banking
Automatic teller machines (ATMs) now make it possible for individuals
located far from their banks to withdraw funds from their bank accounts via
machines located in other cities, states, and even countries. However, most
of these transactions take place within the same states as the banks holding
the deposits. Many automatic teller machines are not owned by a particular
bank but are paid for on a per-transaction basis. Consequently they techni¬
cally are not branch banks and are not controlled by branching regulations.
However, the regulatory decision not to count an ATM as a bank branch
was made only after much discussion and negotiation. Today banks are free
to place these machines where they please. Electronic banking thus repre¬
sents a financial innovation that circumvents branching restrictions and al¬
lows depositors ready access to their deposits in areas far from their banks.
Moreover, as Box 7.3 suggests, this innovation has not adversely affected
employment in the banking system.
Current Status
During the 1980s, states gradually began relaxing their restrictions on inter¬
state banking. This resulted in part from a need to allow out-of-state banks
to acquire failed banks and S&Ls and to accommodate the needs of corpo¬
rations that have offices in different regions. This trend has continued through
the 1990s and has led to a dramatic change in the structure of banking. Most
states now permit branching within their borders, and a few have reciprocal
privileges that allow branches from other states to enter their states. As of
July 1994, every state except Hawaii allows some form of interstate banking.
How Glass-Steagall Shaped the Banking Industry

Perhaps the piece of banking legislation that had the most profound impact
on the evolution of the banking system into its present form was the Glass-
Steagall Act of 1933. Although not all elements of this act are still in force,
Employment in Banking and

ATM Machines
Many employees worry about the effect of auto¬ Employment

mation on employment, and indeed automation in Banking
can eliminate some jobs while creating others. as a Percentage
The accompanying table shows the number of Number of Nonagricultural
ATM machines, as well as employment in the Year of ATMs1 Employment2
banking industry as a percentage of total non-
1980 19,858 1.74%
agricultural employment, during the early years
1983 53,332 1.84
of ATM growth. The 300 percent increase in the
1986 81,510 1.74
number of ATM machines between 1980 and
1986 did not reduce employment in banking 1 Magazine of Bank Administration, May 1987.
2 Supplement to Employment and Earnings, Bureau of
relative to total employment. In fact, from 1980
Labor Statistics, 1983-1989.
to 1983, employment in banking increased
slightly, as banks adjusted to the change in
technology by hiring fewer bank tellers but
more accountants and high-tech employees.
Glass-Steagall led to fundamental changes in the structure of the banking

industry that persist today.
As we already noted, Glass-Steagall was passed in response to the col¬
lapse of the financial system during the Great Depression, and one of its
provisions created the FDIC in the hope of limiting incentives for depositors
to engage in bank runs. In addition, Glass-Steagall imposed a number of
restrictions on commercial banks, including restrictions against investment
banking and limitations on deposit interest paid by banks. We now examine
why these restrictions were enacted and how banks, courts, and even legis¬
lators responded to some of the problems they created.
Restrictions Against investment Banking

The Glass-Steagall Act banned commercial banks from engaging in invest¬
ment banking. An investment bank handles the administration, promotion,
and distribution of securities and guarantees the issuer a certain price. It then
sells the issue in the primary market and earns a profit on the difference.
Prior to Glass-Steagall, banks freely engaged in these activities. Unfortu¬
nately, some banks experienced losses when they were unable to sell the
securities in the primary market after the stock market crash of 1929. This
created losses for some banks that exacerbated the banking crisis of the time.
In an effort to prevent such problems from recurring, the Glass-Steagall
Act tied the hands of commercial banks, prohibiting them from engaging in
investment banking activities. Its passage also stemmed fears that such ac¬
tivities by commercial banks could lead to conflicts of interest if the banks
compromised the safety of their portfolios of assets to accommodate their
own investment banking activities. Consequently, our banking system today
differs substantially from those of many other countries. For instance, Ger¬
many, France, Switzerland, and the United Kingdom allow unified banking;
that is, their banks serve as both commercial banks and investment banks.
In recent years, however, financial innovation in the U.S. banking in¬
dustry has led to a wider array of services at both commercial and investment
banks. Through bank holding companies, banks set up mutual funds, bro¬
kerage services, life insurance, and a host of other services. During the 1980s,
investment banks acquired some of the failed S&Ls, and firms like Merrill
Lynch offered money market funds with check-writing privileges. Further¬
more, in 1988 the Supreme Court ruled that the Fed could allow banks to
underwrite some types of bonds. As a result of financial innovation and court
rulings, commercial and investment banks today can offer a wider array of
services than originally allowed under the Glass-Steagall Act, although they
are still not entirely free to do as they please.
Interest Rate Restrictions

The Glass-Steagall Act also prohibited banks from paying interest on demand
deposits (i.e., checking accounts) and allowed the Federal Reserve System
to regulate interest rates on saving and time deposits at commercial banks.
The authors of Glass-Steagall reasoned that one cause of bank failures during
the early 1930s was excessive competition for deposits by banks, which
eroded their profits. The purpose of interest rate ceilings on deposits, known
as Regulation Q, was to reduce the cost to banks of obtaining funds and
thereby enhance their profits and the stability of the banking system. For
many years, the Fed used its newly given authority to impose interest rate
ceilings on saving and time deposits at commercial banks through Regulation
Q, until the caps were phased out during the 1980s (see Box 7.4). The
discussion that follows describes the events that led to the elimination of
those ceilings.
Interest Rate Adjustment Act. An unintended consequence of

Glass-Steagall’s restrictions on bank interest payments was that banks were
put at a competitive disadvantage relative to S&Ls. Under Glass-Steagall,
interest rate ceilings applied only to deposits at commercial banks. Savings
and loans were free to compete for deposits by paying higher interest rates
on saving and time deposits (but not on demand deposits; at that time, savings
The Long Phaseout of Regulation Q
The accompanying table shows the steps in the eliminated except the restriction that no interest
phaseout of Regulation Q. Note that by March could be paid on demand deposits. (This restric¬
31, 1986, all interest rate ceilings had been tion is no longer in force.)
Effective Date
of Change Nature of Change
June 1, 1978 MMCs established, with minimum denomination of $10,000 and maturities of 26 weeks. The
floating ceiling rates for each week were set at the discount yield on six-month Treasury bills
at S&Ls and MSBs, 25 basis points less at CBs.
November 1, 1978 CBs authorized to offer ATS accounts, allowing funds to be transferred automatically from
savings to checking accounts as needed to avoid overdrafts. The ceiling rate on ATS
accounts was set at 5.25 percent, the same as the ceiling rate on regular savings accounts
at CBs.
July 1, 1979 SSCs established with no minimum denomination, maturity of 30 months or more and floating
ceiling rates based on the yield on 272-year Treasury securities, but 25 basis points higher at
S&Ls and MSBs. Maximums of 11.75 percent at CBs and 12 percent at S&Ls and MSBs.
June 2, 1980 The floating ceiling rates on SSCs raised 50 basis points relative to the yield on 2V2-year
Treasury securities at S&Ls and MSBs and at CBs. The maximum ceiling rates set in June
1979 were retained.
June 5, 1980 New floating ceiling rates on MMCs. All depository institutions may pay the discount yield on
6-month Treasury bills plus 25 basis points when the bill rate is 8.75 percent or higher. The
ceiling rate will be no lower than 7.75 percent. A rate differential of up to 25 basis points
favors S&Ls and MSBs if the bill rate is between 7.75 percent and 8.75 percent
December 31, 1980 NOW accounts permitted nationwide at all depository institutions. Ceiling rates on NOW and
ATS accounts set at 5.25 percent.
August 1, 1981 Caps on SSCs of 11.75 percent at CBs and 12 percent at S&Ls and MSBs eliminated. Ceiling
rates float with the yield on 2V2-year Treasury securities.
October 1, 1981 Adopted rules for the All Savers Certificates specified in the Economic Recovery Act of 1981.
November 1, 1981 Floating ceiling rates on MMCs each week changed to the higher of the 6-month Treasury bill
rate in the previous week or the average over the previous four weeks.
December 1, 1981 New category of IRA/Keogh accounts created with minimum maturity of 1V2 years, no
regulated interest rate ceiling and no minimum denomination.
May 1, 1982 New time deposit created with no interest rate ceiling, no minimum denomination and an
initial minimum maturity of 3V2 years.
New short-term deposit instrument created with $7,500 minimum denomination and 91-day
maturity. The floating ceiling rate is equal to the discount yield on 91 -day Treasury bills for
S&Ls and MSBs, 25 basis points less for CBs.
Maturity range of SSCs adjusted to 30-42 months.
September 1, 1982 New deposit account created with a minimum denomination of $20,000 and maturity of 7 to
31 days. The floating ceiling rate is equal to the discount yield on 91 -day Treasury bills for
S&Ls and MSBs, 25 basis points less for CBs. These ceiling rates are suspended if the 91-day
Treasury bill rate falls below 9 percent for four consecutive Treasury bill auctions.
Continued on p. 218

December 14, 1982 MMDAs authorized with minimum balance of not less than $2,500, no interest ceiling, no
minimum maturity, up to six transfers per month (no more than three by draft), and
unlimited withdrawals by mail, messenger or in person.
January 5, 1983 Super NOW accounts authorized with same features as the MMDAs, except that unlimited
transfers are permitted.
Interest rate ceiling eliminated and minimum denomination reduced to $2,500 on 7- to 31-day
accounts.
Minimum denomination reduced to $2,500 on 91-day accounts and MMCs of less than
$100,000.
April 1, 1983 Minimum maturity on SSCs reduced to 18 months.

October 1, 1983 All interest rate ceilings eliminated except those on passbook savings and regular NOW
accounts. Minimum denomination of $2,500 established for time deposits with maturities of
31 days or less (below this minimum, passbook savings rates apply).
January 1, 1984 Rate differential between commercial banks and thrifts on passbook savings accounts and 7-
to 31-day time deposits of less than $2,500 eliminated. All depository institutions may pay a
maximum of 5.50 percent.
January 1, 1985 Minimum denominations on MMDAs, Super NOWs and 7- to 31 -day ceiling-free time deposits
reduced to $1,000.
January 1, 1986 Minimum denominations on MMDAs, Super NOWs and 7- to 31 -day ceiling-free time deposits
eliminated.
March 31, 1986 All interest rate ceiings eliminated, except for the requirement that no interest be paid on
demand deposits.
Terms:
S&Ls—savings and loan associations SSCs—small saver certificates
MSBs—mutual savings banks ATS—automatic transfer service accounts
CBs—commercial banks NOW accounts—negotiable order of withdrawal accounts
MMCs—money market certificates MMDAs—money market deposit accounts
Source: Gilbert, R. Alton, Requiem for Regulation Q: Reserve Bank of St. Louis Review February 1986, pp. 22-
What It Did and Why It Passed Away from Federal 37. Copyright © 1986. Used with permission.
and loans did not offer checkable accounts). Over time many savers switched
saving deposits from banks into thrift institutions, which paid higher interest
rates. To level the playing field, Congress passed the Interest Rate Adjustment
Act of 1961, which extended the Glass-Steagall interest rate limits to thrifts.
The Interest Rate Adjustment Act reduced the flow of deposits from
banks to thrift institutions. However, the soaring interest rates of the 1970s
ultimately led to a problem called disintermediation: The ceilings on de¬
posit rates at banks and thrifts led depositors to pull funds out of banks and
savings and loans and place those funds in alternative financial instruments
that paid market interest rates, such as money market mutual funds. The
impact of this drain of funds from banks and thrifts to money market mutual
funds is difficult to overemphasize. Money market mutual funds grew from
about $3 billion in assets in 1977 to nearly $190 billion in assets in 1981!
This represents a very large volume of funds that would otherwise have been
available to the banking industry had it been allowed to offer market rates
on deposits.
DIDMCA. During the late 1970s, disintermediation was rapidly draining

banks and thrifts of their deposits and reserves, reducing their ability to issue
new loans that would have earned higher market interest rates. Furthermore,
thrifts were saddled with long-term loans (such as mortgages) that were
earning low, fixed interest rates. This problem was rapidly driving the entire
savings and loan industry into insolvency.
Congress’s response to these problems was to pass the Depository In¬
stitution Deregulation and Monetary Control Act (DIDMCA) in 1980.
DIDMCA phased out interest rate ceilings on deposits and permitted nation¬
wide NOW (negotiable order of withdrawal) accounts, which were essen¬
tially checking accounts at thrift institutions. This allowed banks and thrifts
to attract new deposits by offering interest rates that were more in line with
the market rates paid by money market funds. To allow thrifts to convert
these new deposits into new loans at the relatively high market interest rates,
DIDMCA permitted S&Ls to make some consumer loans, credit unions to
make real estate loans, and thrifts to issue credit cards. As mentioned earlier
in the chapter, DIDMCA also enacted uniform reserve requirements, which
gave back to the Fed some of the power it had lost as a result of its declining
membership.
Garn-St. Germain. DIDMCA blurred the distinction between thrifts

and commercial banks, and Congress further obscured it when it passed the
Garn-St. Germain Act of 1982. This act, which was directed at dealing with
S&L failures that occurred between 1980 and 1982, contained a provision
allowing the FDIC and FSLIC to arrange interstate mergers of troubled
institutions. More important, it gave thrifts more freedom to issue loans and
compete for deposits. For instance, thrifts were allowed to issue up to 10
percent commercial loans and 30 percent consumer loans, and depository
institutions were authorized to offer money market deposit accounts
(MMDAs) to compete with money market mutual funds. These MMDA
accounts could offer competitive interest rates and, unlike money market
mutual funds, were insured by federal deposit insurance. Today, as a result
of all of these events since Glass-Steagall that ultimately culminated in the
S&L crisis, commercial banks and thrifts are very similar in terms of both
the types of loans they can make and the regulations to which they are
subject.
Summary (a) List some of the financial innovations that resulted from banks’ attempts
Exercise 7.2 to circumvent banking restrictions, (b) List some of the legislative changes
that blurred the distinction between commercial banks and S&Ls.
Answer: (a) Your list should include checkable deposits, bank holding
companies, and electronic banking, among others. Make sure you understand
the environment that led to each of these innovations, (b) Among other
things, your list should include the following: (i) The FDIC now insures
S&Ls and banks, (ii) all depository institutions are subject to uniform reserve
requirements set by the Fed, and (iii) S&Ls and banks are now allowed to
make similar types of loans. Make sure you know the legislation and chain
of events that led to each of these changes.
The Economic Impact of Banking Regulations

As we have seen, legislative and regulatory bodies have imposed a variety
of restrictions and regulations that shaped the banking industry into what it
is today. While the primary aim of such regulations was to enhance the
stability of the banking industry and protect depositors, some of the restric¬
tions ultimately had the opposite effect. We have already seen that over the
years, banks have come up with financial innovations that circumvent many
of these laws. In this section, we examine some of the economic implications
of various regulatory policies for the markets for loans, deposits, and credit.
We will see that some regulations have had adverse effects on consumers
and banks.
Bank Charters
The requirement that banks be chartered before they can operate distin¬
guishes the banking industry from many other industries in the United States.
As we have seen, the strict capital requirements and lengthy time period
required for approval of a bank charter limits entry by new banks. This
means that if existing banks are earning sizable economic profits, it is dif¬
ficult for a new bank to quickly emerge and compete away some of those
profits. We can use the model of bank behavior developed in Chapter 5 to
see the implications of this condition.
Part a of Figure 7.2 shows the market supply and demand for loans in a
purely competitive banking industry. The equilibrium interest rate—10 per¬
cent in this example—is determined by the intersection of the supply (5°)
and demand (D°) curves for loans at point A in part a of Figure 7.2.
Part b illustrates the situation for an individual bank. An individual bank
in a purely competitive market can issue loans at the market interest rate of
10 percent, so the demand for an individual bank’s loans is horizontal and
given by D% Since the interest rate is also the individual bank’s marginal
revenue curve, the bank maximizes profits by issuing the quantity of loans
where the interest rate equals the marginal cost of loans, which is at point a
in part b of Figure 7.2. This corresponds to L°f loans issued by the individual
bank, at an interest rate of 10 percent. The economic profits earned by this
bank are given by the shaded rectangle.
If entry into the banking industry were unrestricted and existing banks
were earning profits, new banks would enter the banking industry to reap
some of the profits. The entry of new banks would shift the market supply
curve in part a of Figure 7.2 to the right to S], resulting in a new equilibrium
at point B. Notice that as a result of the entry of new banks into the industry,
the interest rate on loans falls to 8 percent. Borrowers now obtain loans at a
lower interest rate due to increased competition in the banking industry.
The decline in the market interest rate shifts the demand for an individual
bank’s loans down to DF in part b of Figure 7.2. Marginal revenue now
equals marginal cost at point b. Furthermore, at point b the interest rate on
loans exactly equals the bank’s average cost of issuing loans; the bank now
earns zero economic profits. Thus, with free entry into the banking industry,
new banks will enter the industry until the economic profits earned by
existing banks are zero. Banks receive just enough interest payments on
loans to cover the opportunity cost of the resources required to issue the
loans. At this point, there is no further incentive for additional banks to form.
As you can see, free entry into the banking industry would ensure that
banks charge an interest rate just high enough to cover the opportunity cost
of the loaned funds. However, there is not free entry into the banking
industry. Even when entry is allowed, it can take considerable time for a
new bank to go through the process of obtaining a charter; in the meantime,
interest rates and bank profits will remain at a high level.
In fact, limiting entry tends to keep bank profits and loan interest rates
at a higher level than they would be otherwise. For example, in part a of
Figure 7.2, the interest rate is 10 percent if new banks are not allowed to
enter but falls to 8 percent if entry is allowed due to the increased supply of
loans. Similarly, the shaded rectangle in part b shows profits if entry is not
allowed, which fall to zero if entry is permitted.
Proponents of restrictions on entry into banking argue that the negative
aspects of restricted entry (the higher interest rates to borrowers) are more
than offset by the positive aspects. First, they argue that restricting entry of
new banks means fewer banks will fail, since banks will earn greater profits.
Second, by carefully screening applications for a bank charter, banking
authorities can prevent the entry of banks that are likely to fail in the first
place. Together these factors reduce the likelihood of bank failures and also
decrease the size of the insurance claims that must be paid on those banks
that do fail.
Branching Restrictions
Branching restrictions decrease the accessibility of bank deposits and loans
outside the area serviced by a bank, which reduces competition among banks
and inconveniences customers. Individuals keep fewer funds on deposit,
since they need larger stashes of cash for unexpected trips outside of bank
service areas. This decreases the market supply of deposits from S° to S1 in
Figure 7.3 and increases the interest rate on deposits from i°D to i lD. (Remem¬
ber, banks use deposits to produce loans; therefore, they are the demanders
in the deposit market, and households are the suppliers.) Thus, because of
branching restrictions, banks actually have to offer higher interest rates on
deposits to attract customers than they would otherwise. These higher rates
partially compensate depositors for the inconvenience of keeping deposits
at banks. But since an increase in the interest rate paid on deposits increases
a bank’s marginal cost of issuing loans, interest rates on bank loans will rise
as well. This rise is further exacerbated by lessened competition among
banks for loan customers.
As we noted earlier, however, banks today are less adversely affected
by branching restrictions than in the previous decade due to the emergence
of automated teller machines. As a result, depositors are now willing to
maintain larger balances in their accounts, knowing they can readily with¬
draw funds from ATMs in distant locations. The emergence of ATMs shifts
the supply of loanable funds from S] back to S'0 in Figure 7.3, lowering the
cost to banks of obtaining funds.
A subtler but more important aspect of branching restrictions is their
effect on the riskiness of a bank’s loan portfolio. When a bank is restricted
to making loans and obtaining deposits in a small geographic area, it is more

sensitive to fluctuations in local market conditions than it would be if it were
allowed to engage in banking activities across a larger geographic region.
Consider a bank in a city on the Gulf Coast that is restricted from opening
branches outside the city. If a hurricane hits and destroys the city, depositors
are likely to withdraw sizable amounts of their savings from the bank.
Moreover, loans to borrowers in the city may not be repaid. This can have
disastrous effects on the bank. In contrast, if the bank is allowed to open
branches throughout the United States, the local hurricane will not likely
affect withdrawals in other regions of the country, and loans issued elsewhere
will not be affected. Consequently, the bank will be diversified; the unantic¬
ipated withdrawals in the local area will likely be offset by unexpected
deposits in other regions. In short, branching restrictions actually increase
the likelihood of bank failures.
Bank holding companies, as we saw earlier, allow banks to circumvent
these negative aspects of branching restrictions. A bank holding company
that owns the controlling interest of many banks across the country owns a
diversified portfolio of banks, just as an investor who owns shares of stock
in many different companies is diversified. Largely for this reason, the 1990s
have seen a movement toward allowing banks and bank holding companies
more freedom in crossing regional boundaries as well as in the type of
activities they can conduct.
Interest Rate Regulations

In addition to agencies specifically designed to supervise the banking indus¬
try, state and federal legislative bodies also pass laws that have an impact
on the banking industry. This section examines laws and regulations that
affect the interest rate banks may charge for loans and pay on deposits.
Usury Laws. Many states have usury laws, which are ceilings that
restrict the interest rate lenders can charge on loans and credit card balances.
Figure 7.4 shows the impact of a usury law on the banking industry. Without
a usury law, market forces determine the interest rate where the supply of
and demand for loanable funds intersect at point A. At this point, the market
interest rate is i£, and the banking industry issues a total of L° loans. The
usury law makes it illegal to charge an interest rate above ilL for a loan. At
this regulated interest rate, borrowers desire Ld loans, but the quantity sup¬
plied of loans is only L\ The usury law results in a shortage of Ld — Ls
loans, and fewer loans are issued (Ls) than would be in the absence of the
usury law (L°).
By reducing the interest rate banks can legally charge on loans, usury
laws result in lower profits for banks. In extreme instances, where the cost
to a bank of obtaining funds exceeds the interest rate the bank can legally
charge for loans, the bank may choose to issue no loans at all. In less extreme
circumstances, such as those depicted in Figure 7.4, the bank issues loans,
but to fewer borrowers than want them at the ceiling interest rate.
Since usury laws result in a shortage of loanable funds, banks respond
by loaning funds to only the most creditworthy borrowers; in essence, they
ration credit. The interest rate required to compensate a bank for the risk of
loaning to a noncreditworthy individual (who is less likely to repay a loan)
may exceed the rate allowed by law. Credit rationing can thus negatively
affect borrowers whom banks view as bad credit risks. This is an unfortunate
paradox, since the most common reason given for passing usury laws is to
protect those borrowers who most need funds from having to pay high
interest rates. Usury laws often prevent such individuals from borrowing any
funds at all, at least from a bank.
Regulation Q. Regulation Q restricted the maximum interest rate banks

could pay on saving and time deposits. Fortunately for small savers, the
gradual elimination of these restrictions began with DIDMCA in 1980.
Earlier in this chapter, we examined Congress’s rationale for including in¬
terest rate restrictions in the Glass-Steagall Act. We now look at how they
affected borrowers and lenders.
Part a of Figure 7.5 shows the market for deposits, where households
supply deposits to banks and banks demand them. In the absence of Regu¬
lation Q, equilibrium is at point A, the interest rate on deposits is 8 percent,
and the resulting level of deposits is D°. For simplicity, suppose banks
convert all deposits to loans; thus, the supply of loans is given by the vertical
line labeled S = D° in part b of Figure 7.5. In the absence of an interest
rate ceiling on deposits, the equilibrium in the loan market is at point B and
the interest rate on loans is 10 percent.
Now suppose an interest rate ceiling of 5 percent is imposed in the
deposit market, as it was during the 1970s under Regulation Q. Let us use
supply and demand analysis to illustrate the problem of disintermediation
we discussed earlier in the chapter. At a rate of 5 percent, banks desire Dd
in deposits, but households deposit only Ds in funds; the result is a shortage
of deposits. Because of this shortage, banks can now issue fewer loans. In
particular, since they can issue only Ds in loans, the supply curve of loans
in part b of Figure 7.5 shifts to the left to S = D\ The effect of this decrease
in the supply of loans, due to the availability of fewer deposits, is to increase
the interest rate charged on loans from 10 to 12 percent. Thus, interest rate
ceilings on deposits—designed to keep bank costs low—actually increase
the rate banks charge on loans. In fact, the law harms borrowers, lenders,
Figure 7.5
Regulation Q and Loan Interest Rates
Part a depicts the deposit market, where the market-determined interest rate on deposits is 8 percent
and D° in deposits is provided to banks. Assuming for simplicity that the supply of loans equals the
quantity of deposits, D° worth of deposits is converted into D° worth of loans by banks in part b. This is
given by the vertical supply of loans curve 5 = 0° and results in an interest rate charged on loans of 10
percent. If there is a restriction on the interest rate paid on deposits (as under Regulation Q) of 5 percent,
the quantity of deposits falls to 0s in part a. This implies that the quantity of loans banks may offer also
falls to 5 = D5, and the interest rate on loans rises to 12 percent in part b. Thus, restrictions on the
interest rate paid on deposits raises the interest rate charged on loans.
Deposit Interest
Rate (%)
and banks. Those depositing money in banks receive a lower interest rate (5
percent), while those borrowing money from banks pay a higher rate (12
percent). Banks make fewer loans. Of course, DIDMCA phased out ceilings
on deposit rates, which had the opposite effect of lowering interest rates on
loans and raising rates on bank deposits.
Uniform Reserve Requirements

Prior to DIDMCA, banks that were not members of the Fed were subject to
lower reserve requirements than member banks. As a consequence of
DIDMCA, uniform reserve requirements were instituted across all depository
institutions. These uniform reserve requirements gave the Fed considerably
more control over the money supply and the banking industry than was the
case prior to DIDMCA.
To see this, look at Figure 7.6, which shows the market for loanable
funds before uniform reserve requirements were in place. If states and other
Figure 7.6
Decreases in Reserve Requirements and Loan Interest Rates
Prior to DIDMCA, changes in reserve requirements by depository authorities other

than the Fed could affect interest rates. Here a cut in the reserve requirement at
depository institutions (such as thrifts or state banks) that are not members of the
Fed increases the supply of loanable funds to 5\ resulting in a lower interest rate.
DIDMCA changed this possibility by allowing the Fed to set uniform reserve require¬
ments across all depository institutions, thus giving the Fed greater control over the
banking system. The Fed used this power in the early 1990s when it reduced re¬
serve requirements, resulting in lower interest rates as shown in this graph.
authorities controlling reserve requirements at nonmember banks and thrifts

set reserve requirements at 15 percent, the supply of loanable funds is given
by S°. The equilibrium interest rate is i{[, and the quantity of loans issued by
all banks is L°. If authorities lower reserve requirements to 10 percent, their
institutions can then issue more loans at any given interest rate, since a
smaller fraction of deposits must be kept in reserve. Consequently the supply
curve shifts to S\ resulting in a lower interest rate (/}) and more loans (L1).
Thus, prior to DIDMCA, changes in reserve requirements by authorities
other than the Fed affected interest rates and the equilibrium quantity of
loanable funds. This example illustrates that the Fed was not the only driver
at the wheel of the banking system prior to DIDMCA.
Today the Fed controls reserve requirements at all depository institu¬
tions, and it alone has the authority to change them. For instance, between
1990 and 1992, the Fed decreased its average reserve requirement from
10.73 to 7.28 percent of total checkable deposits. Similar reductions occurred
on April 2, 1992, when it lowered reserve requirements on transaction ac¬
counts from 12 to 10 percent. This had effects similar to those shown in
Figure 7.6, resulting in lower interest rates.
We point out, however, that historically the Fed has been reluctant to
change reserve requirements, and as a result they have tended to remain
stable over time. There are two reasons for this stability. First, the Fed can
use a host of other policies to affect the financial system; we will look at
these policies in Chapter 13. Second, prior to DIDMCA the Fed was reluctant
to change its reserve requirements because setting them below those of
nonmember institutions would place those institutions at a disadvantage
relative to member banks. While it remains to be seen whether the Fed will
change reserve requirements more frequently now that uniform reserve re¬
quirements are in place, most economists believe it will not.
Deposit insurance
Deposit insurance insulates depositors from losses due to a bank failure. As
a consequence, even if you hear a rumor that your bank is about to fail, you
have little reason to be concerned or withdraw funds from your account.
This enhances the safety of the banking system by reducing the likelihood
of a bank panic. Unfortunately, as we learned in Chapter 6, deposit insurance
leads to moral hazard.
To be concrete, suppose you are considering depositing money in two
banks. One bank offers a high interest rate on savings deposits, and the other
bank offers a lower interest rate. One bank is more likely to fail than the
other, but the only way for you to find out which one is to spend considerable
time and money researching each bank’s financial situation. In which bank
would you choose to deposit money?
If the banks did not have deposit insurance, you would probably spend
lots of time trying to find out how likely each bank is to fail before making
a decision. But with deposit insurance, you need not spend time collecting
this information; you can simply shop around for the highest rate on deposits.
You would correctly reason, “So what if the bank fails? The FDIC will pay
back every penny I have deposited.” This is moral hazard: You can take
action to avoid losing your deposits, but since you are insured you fail to
do so.
What if the bank offering the higher interest rate does so because it
knows it is more likely to fail than its rival? It reasons that, being on the
brink of bankruptcy, its only hope is to attract additional deposits by offering
higher interest rates than other banks so that it can increase the number of
loans it issues. But to cover the higher cost of obtaining deposits, the bank
will have to charge a higher interest rate on loans than other banks do. We
saw in Chapter 5 that this can lead to adverse selection: The only people
willing to pay the higher interest rate for a loan will be those who are unable
to obtain funds at lower rates from other banks because they are bad credit
risks. As the bank issues more and more loans to bad credit risks, the
likelihood that it will fail actually increases.
Thus, we see that deposit insurance leads to both moral hazard (people
don’t carefully check out the financial status of a bank before depositing
money and banks make risky loans) and adverse selection (the higher interest
rate on loans tends to attract borrowers who know they are bad credit risks).
Together these factors tend to increase the probability of bank failures and
the cost to the FDIC of covering depositor losses.
Bank Examination as a Solution to

Moral Hazard and Adverse Selection
To circumvent the problems of moral hazard and adverse selection, regula¬
tory agencies frequently monitor the loan portfolios of individual banks. To
be most effective, these examinations should be done on a random basis so
that the banks cannot prepare for them or attempt to cover up any probems.
Examining banks randomly over time instead of on prespecified dates en¬
hances the effectiveness of bank examinations in reducing the asymmetric
information that can lead to bank failures.
Summary The city council of a major city is concerned about the high interest rates
Exercise 73 local banks charge on automobile loans. The banks have argued that the high
rates are necessary to cover their cost of obtaining loanable funds. In re¬
sponse, two proposals are before the city council: (a) an ordinance that would
limit the interest local banks could legally pay on deposits and (b) an ordi¬
nance that would limit the rate local banks could charge on car loans. How
would each ordinance affect the local market for car loans? Use a graph of
the local deposit and loan market to answer this question.
Answer: (a) Suppose the initial equilibrium interest rate in the local de¬
posit market is at point A in part a. The interest rate on deposits is 6 percent,
and $12 million is deposited in the local banks. The first proposal is an
interest rate ceiling on deposits of, say, 3 percent. This ceiling creates a
shortage of $15 million in deposits. Given the prospect of earning a lower
rate on deposits at local banks, large depositors would choose to deposit
funds at banks in other cities or use their funds for investments not affected
by the local ordinance. Given the ordinance, local banks collect only $5
million in deposits. Assuming for simplicity that banks use all of their
deposits for car loans, the decrease in deposits leads to a decrease in the
local supply of car loans from S° to S] in part b. The result is an increase in
the interest rate local banks charge for car loans from 10 to 13 percent. Thus,
the first proposal would lead to higher rather than lower interest rates on car
loans. It might also lead to adverse selection: Customers with good credit
ratings could obtain loans at the free market rate (10 percent) from lending
institutions located outside the jurisdiction of their city council. This would
leave local banks with a pool of riskier borrowers who would be willing to
pay the higher local rate.
(b) The second ordinance is an interest rate ceiling in the local market
for automobile loans. As in Figure 7.4, the result of such a ceiling would be
a shortage of car loans. More car loans would be desired at the regulated
interest rate than there would be funds available. Consequently the local
banks would be forced to ration credit. People with average or marginal
credit histories would be unlikely to qualify for a car loan. If the city
regulated the interest rate on car loans at a level below that paid by banks
to acquire deposits, the banks would refuse to issue car loans at all. It would
make no sense for them to lend money at a rate lower than their marginal
cost of obtaining funds. If the city council is not careful, it may create
something worse than a shortage and credit rationing: A situation might
result in which the city’s residents cannot borrow funds locally to purchase
cars.
(a) Local Deposits (b) Local Loans
CONCLUSION
In this chapter we detailed the regulation and supervision of the banking

industry, tracing its evolution and highlighting the financial innovations and
legislative changes that shaped today’s banking system. Table 7.2 sum¬
marizes some of the important legislation. We also provided economic anal¬
yses of the impact of these rules and regulations on the banking industry.
The material in this chapter, coupled with that in the previous two
chapters, has provided you with an understanding of how the banking in-
Conclusion 231
Table 7.2
Chronology of Important Banking Legislation
This table summarizes the chronology and key features of banking regulations in the
United States between 1863 and 1991.
Year Name Legislation
1863 National Banking Act Created national banks

1913 Federal Reserve Act Created the Federal Reserve
System
1927 McFadden Act Required each bank to operate
under the branching restrictions
imposed by the state in which it
operated; essentially allowed
each state to restrict the opera¬
tion (by restricting branching)
of banks chartered nationally or
in other states
1933 Glass-Steagall Act Created the FDIC
Prohibited commercial banks
from engaging in investment
banking (eventually weakened)
Imposed various interest rate reg¬
ulations (no longer in force)
1956 Bank Holding Company Act Prohibited interstate acquisitions
unless permitted by state in
which acquired bank was
located
1961 Interest Rate Adjustment Extended interest rate regulations
Act to thrifts (no longer in force)
1970 Bank Holding Company Act Extended restrictions on bank
holding companies to include
one-bank holding companies
1980 Depository Institution De¬ Phased out interest rate ceiling
regulation and Monetary Extended Federal Reserve System
Control Act (DIDMCA) reserve requirements to all
institutions regardless of
membership
Made Federal Reserve System
services available to all banks
and thrifts; services must be
priced to cover costs
Continued on p. 232

Expanded powers of thrifts
Increased dollar amount of depos¬
its covered by FDIC insurance
1982 Garn-St. Germain Act Allowed money market deposit
accounts
Further increased powers of thrifts
Allowed interstate mergers of
troubled institutions
1983 International Lending Super¬ Authorized national bank super¬
visory Act visory and regulatory agencies
to set minimum capital require¬
ments for banks
1989 Financial Institutions Re¬ Abolished the Federal Home Loan
form, Recovery, and En¬ Bank Board and the FSLIC
forcement Act (FIRREA) Restructured the FDIC to have
bank insurance fund (BIF) and
savings association insurance
fund (SAIF)
Increased deposit insurance
premiums
1991 Federal Deposit Insurance Requires the FDIC to monitor ex¬
Corporation Improvement aminers and minimize long¬
Act (FDICIA) term cost to insurance fund of
troubled banks
Requires the FDIC to institute
risk-based insurance premiums
dustry and individual banks operate. In the next chapter, we focus on finan¬
cial markets and look at how the prices and yields of financial instruments
such as stocks and bonds are determined.
KEY TERMS
national bank unit banking

dual banking system statewide branching
state bank bank holding company
bank charter investment bank
International Lending Supervisory Act of 1983 unified banking
Federal Reserve System Regulation Q
Key Terms continued
uniform reserve requirements disintermediation

Federal Deposit Insurance Corporation (FDIC) usury law
limited branching

1. What does the term dual banking system 8. Why was the McFadden Act of 1927 en¬
mean? Why does the United States oper¬ acted?
ate under such a system?
9. What is a bank holding company? Why
2. Who is responsible for granting charters are more than 90 percent of all banks in
to national banks? To state banks? the United States owned by holding com¬
panies?
3. Can state banks become members of the
Federal Reserve System? Can they join 10. Explain the pros and cons of bank charter
the FDIC? restrictions.
4. Can a national bank choose not to become 11. Suppose the government made bank hold¬
a member of the Federal Reserve System? ing companies illegal. What would you
Can it refuse to become a member of the expect to happen to the number of bank
FDIC? failures? Why?
5. A current proposal would increase the 12. Do usury laws benefit or harm (a) well-to-
premium banks pay for FDIC insurance do borrowers or (b) borrowers who are
from 25 cents to 31 cents per $100 in de¬ down on their luck?
posits. What impact, if any, would this
have on (a) a national bank’s decision to 13. What impact did Regulation Q have on
obtain FDIC insurance, (b) a state bank’s the interest rates (a) earned on deposits
decision to renew its FDIC insurance, and and (b) charged on bank loans?
(c) the relative number of new banks that
14. Explain how deposit insurance can actu¬
would open as state banks instead of na¬
ally increase the likelihood that a troubled
tional banks?
bank will fail.
6. When was the FDIC established? Why?
15. What practices do banking supervisory
7. Why are there more than 11,000 banks in agencies undertake to mitigate the situa¬
the United States? tion in problem 14?

Elyasiani, E., and S. Mehdian. “Efficiency in the tier Approach.” Applied Economics, 22 (April
Commercial Banking Industry: A Production Fron- 1990), 539-551.
Fenstermacker, J. V., R. P. Malone, and S. R. Stan- Hester, D. D. “Instructional Simulation of a Commer¬

sell. “An Analysis of Commercial Bank Common cial Banking System.” Journal of Economic Edu¬
Stock Returns: 1802-97.” Applied Economics, 20 cation, 22 (Spring 1991), 111-143.
(June 1988), 813-841. Ladenson, M. L., and K. J. Bombara. “Entry in Com¬
Fraser, D. R., and J. W. Kolari. “The 1982 Deposi¬ mercial Banking: 1962-78.” Journal of Money,
tory Institutions Act and Security Returns in the Credit, and Banking, 16 (May 1984), 165-174.
Savings and Loan Industry.” Journal of Financial Lobue, M. “Categorical Bank Acquisitions.” Journal
Research, 13 (Winter 1990), 339-347. of Bank Research, 14 (Winter 1984), 274-282.
Graddy, D. B., and A. S. Kama. “Net Interest Margin Reidenbach, R. E., D. L. Moak, and R. E. Pitts. “The
Sensitivity among Banks of Different Sizes.” Impact of Marketing Operations on Bank Perform¬
Journal of Bank Research, 14 (Winter 1984), 283- ance: A Structural Investigation.” Journal of Bank
290. Research, 17 (Spring 1986), 18-27.
Graddy, D. B., and A. S. Kama. “Dividend Policy Wells, D. R., and L. S. Cruggs. “Historical Insights
and the Return of Bank Holding Company Stock.” into the Deregulation of Money and Banking.”
Quarterly Journal of Business and Economics, 25 Cato Journal, 5 (Winter 1986), 899-910.
(Spring 1986), 3-21. Witte, W. E. “Dynamic Adjustment in an Open
Graddy, D. B., and R. Kyle III. “The Simultaneity of Economy with Flexible Exchange Rates.” South¬
Bank Decision-making, Market Structure, and ern Economic Journal, 45 (April 1979), 1072-
Bank Performance.” Journal of Finance, 34 1090.
(March 1979), 1-18.
PART THREE
The Valuation of Financial Assets

CHAPTERS
8
The Time Value of Money and Asset Pricing
9
Risk, Uncertainty, and Portfolio Choice
10
The Term Structure of Interest Rates
11
Foreign Exchange Markets
12
Rational Expectations and Efficient Markets
CHAPTER
The Time Value of Money and

Asset Pricing
ow do investors on Wall Street determine the value of a share of IBM

stock or an AT&T bond? How do changes in taxes, inflation, risk, or other
economic variables affect the price of NationsBank’s stock or a Treasury
bill in the firm’s portfolio? This chapter and the next two chapters will
answer these questions. In this chapter, we show that various financial in¬
struments will have different prices due to differences in payment schedules
(when and how much money is paid). In Chapter 9, we extend the analysis
to situations where the interest rate depends on the riskiness of the asset. We
will see that riskier financial assets tend to sell at lower prices than less risky
ones. In Chapter 10, we further extend the analysis to situations where the
interest rate varies based on the length of the loan.
In this chapter, we focus on financial assets such as Treasury bills,
Treasury bonds, corporate bonds, and common stock that are bought and
sold in well-established secondary markets at a dollar price (as opposed to
an interest rate). You can think of the prices of these financial instruments
as being analogous to the prices paid for various types of automobiles or
refrigerators. In this chapter, we examine how these prices are determined
and why they fluctuate with economic conditions.
The price of any asset, including financial instruments such as stocks
and bonds, reflects the present value of the benefits enjoyed by owning the
asset. For this reason, we begin with a review of present value analysis,
introduced in Chapter 3. We then adapt this analysis to determine the value
of bonds, stocks, and other financial instruments. Finally, we see how the
basic supply and demand analysis developed in Chapter 4 can be used to
analyze the impact of changes in such things as taxes and inflationary ex¬
pectations on the prices of different financial instruments. This material is
the building block for the next two chapters, which show how risk and the
term structure of interest rates affect the analysis.
The Time Value of Money
As we saw in Chapter 3, the time value of money reflects the fact that $1
received in the future is worth less than $1 today. By waiting until a future
236
date to receive a given amount of money, you forgo the immediate use of
the money, including the interest you would earn if you saved the money.
Economists use present value analysis to determine the value today (in the
present) of a future sum of money. The present value of an amount received
in the future is the amount you would have to invest today, at the prevailing
interest rate, to end up with the future amount.
In a sense, present value analysis is the reverse of future value analysis.
If you deposit $100 today in a bank account that pays 5 percent interest per
year, at the end of the year you will have $105. Thus, $105 is the future
value of $100 today. Stated in reverse, $100 is the present value of $105
received in the future. The present value of a future sum of money depends
on (1) the interest rate, (2) the length of time until you will receive the future
amount, and (3) the size of the future amount.
To be more concrete, suppose you deposit $2,000 in the bank today and
the bank agrees to pay you an interest rate of 4 percent per year. After one
year, you will have
(1 + .04) X $2,000 = $2,080.
The 1 in parentheses reflects the fact that you get your principal back, and
the .04 is the interest rate on the principal. At the end of two years, you will
have
[(1 + .04) X $2,000] X (1 + .04) = (1 + .04)2 X $2,000 = $2,163.20.
The fact that (1 + .04) is raised to the power of 2 reflects the fact that the
original amount is invested for two years; you earn interest in the second
year on the interest received in the first year.
More generally, suppose a single market interest rate, i, is available to
all borrowers and lenders (we relax this assumption in the next two chapters,
where we allow interest rates to vary depending on the riskiness and length
of the loan). If the present amount to be invested at the interest rate is PV,
the future value (FV) of the present amount in T years is
FV = PV X (1 + if. (8.1)
This equation suggests that present value and future value are closely related.
In particular, if we solve equation 8.1 for PV in terms of FV, we obtain
FV
PV (8.2)
(1 + if
Equation 8.2 tells us the present value (PV) of an amount (FV) received T
years in the future when the interest rate is i percent per year. For example,
if FV = $1,050, T = 1, and i = .05, you will receive $1,050 one year in
the future, and the prevailing interest rate is 5 percent. In this case, the
present value of the future amount is
FV _ $1,050
PV $1,000.
(1 + if ~ (1.05)1
238 Chapter 8 The Time Value of Money and Asset Pricing
In other words, if the interest rate is 5 percent, you will need to invest $1,000
today (the present) to receive $1,050 a year from now.
Equation 8.2 illustrates the time value of money. The farther into the
future a given dollar amount is to be received (the greater T is), the lower
its present value. In other words, the present value of a future amount is
inversely related to the time in which that amount is received. Moreover,
note that the present value of an amount to be received at a given point in
the future is inversely related to the interest rate. As the interest rate rises,
the present value of a given future amount falls. Conversely, the lower the
interest rate, the greater the present value.
We can easily apply the basic formula for computing the present value
of a future amount to calculate the present value of a series of future values.
Suppose varying amounts are to be received over time; FVj will be received
at the end of the first year, FV2 at the end of two years, and so on for T
years. The present value of this stream of future payments is
FVi FV2 FV3 FVr

PV 7 + 7 + ' 7 + • • • + 7.
(1 + 0 (1 + if (l + if (l + if
The first term on the right-hand side of the equality is the present value of
the first future payment. To this we add the present value of the payment
received two years in the future, three years in the future, and so on. By
adding together the present value of each future payment, we end up with
the present value of the series of future payments.
Summary Your uncle plans to purchase a $50,000 Mercedes Benz when you graduate
Exercise 8.1 three years from now. If the interest rate is 5 percent, how much will he
have to deposit in the bank today to be able to purchase the car when you
graduate?
He must deposit the present value of the $50,000 needed in three

years. Given the 5 percent interest rate, this amount is
$50,000
PV $43,191.88.
(1.05)3
The Valuation of Debt Instruments

Fet us now see how present value analysis determines the market price of
debt instruments held by banks and other investors. Short-term debt instru¬
ments such as U.S. Treasury bills, commercial paper, and repurchase agree¬
ments are bought and sold at prices determined in the money market—the
market for short-term debt. They comprise an important component of the
portfolios of banks and other financial intermediaries such as pension funds

and insurance companies. In contrast, debt instruments that are long term in
nature, such as U.S. Treasury bonds and corporate bonds, are traded at prices
determined in the capital market—the market for long-term debt. These debt
instruments are held primarily by pension funds, insurance companies, mu¬
tual funds, and individual investors.
A debt instrument is simply a written promise by a borrower to pay
the owner of the instrument a specified amount (or amounts) in the future.
When a debt instrument is first issued by a borrower and sold to a lender in
the primary market, the owner of the instrument is the original lender.
However, debt instruments can often be resold in a secondary market, so the
current owner is not necessarily the original lender.
Debt instruments take various forms. In some instances, a single amount
will be paid in the future; in other cases, several future payments may be
promised. In the following subsections we see how the payment schedule
affects the valuation of a given type of debt instrument.
Treasury Bills and Other Debt

instruments Sold on a Discount Basis
Many debt instruments—including Treasury bills, banker’s acceptances, and
some forms of corporate bonds (called discount bonds)—do not pay interest
directly. Instead they are sold on a discount basis, that is, at a price that is
less than their face value. At maturity, the owner presents these bonds to the
borrower, who pays the face value. For instance, the face value of a Treasury
bill is $10,000; the U.S. Treasury promises to pay $10,000 to the owner of
the Treasury bill when it matures—say, in a year. Treasury bills are sold in
an auction at a price that is less than their face value, for example, $9,500.
Thus, for these types of debt instruments, the return to the owner is the
difference between the face value collected at maturity and the purchase
price.
Discount Yield and Yield to Maturity. The yield on a Treasury

bill or a discount bond is a measure of the percentage annual return earned
by the owner of the instrument. There are two methods of measuring the
yield of debt instruments sold on a discount basis: the discount yield method
and the yield to maturity.
The discount yield, which is frequently used to measure yields on
Treasury bills, is defined as
FV - PTB 360
Discount yield = - X -,
FV Days to maturity
where FV is the future value of the Treasury bill, or the face value—the
amount paid when the bill matures—and PTB is the price paid for the Treas-
ury bill. For instance, a $10,000 Treasury bill that matures in 365 days and
was purchased for $9,500 would have a discount yield of
$10,000 - $9,500 360

Discount yield = .049,
$10,000 365
which is 4.9 percent.

Two aspects of the discount yield formula cause it to understate the
actual yield on a Treasury bill. First, note that the formula is based on a
360-day year, not on the standard 365-day year. Second, the discount yield
formula uses the face value of the Treasury bill rather than its actual purchase
price in the denominator. In other words, the term (FV — PTB)/W in the
discount yield formula does not represent the actual percentage return to the
investor; the actual return is (FV — PTb)!Ptb• Since PTB is less than FV, the
discount yield formula understates the actual yield on a Treasury bill.
A measure of yield that does not suffer from these two deficiencies is
the yield to maturity. The yield to maturity is the interest rate at which the
amount used to purchase the Treasury bill would have to be invested to grow
to the face value paid at maturity. Consider the one-year, $10,000 Treasury
bill purchased for $9,500. At what interest rate would one have to invest
$9,500 today to end up with $10,000 at the end of one year? Mathematically,
we must find the value of i that satisfies the following equation:
$10,000
$9,500
1 + i
Solving this equation for (1 + /) gives us
_10
1 + i
95'
so i = .053. Thus, the yield to maturity on the Treasury bill is 5.3 percent,
which is greater than the discount yield.
Box 8.1 describes how to interpret financial data for Treasury bills with
various maturities.
Valuing Debt Instruments Sold on a Discount Basis. How

much will a bank or other investor pay for a debt instrument that is sold on
a discount basis? The answer depends on the market interest rate. Consider
a Treasury bill with a face value of $10,000 that matures in one year. Since
Treasury bills are sold on a discount basis, the only thing the buyer will
receive is $10,000 one year from today. The amount an investor would be
willing to pay for this Treasury bill depends on the market interest rate.
When the interest rate is low, the investor is willing to pay a relatively high
price and earn a relatively low yield, since alternative investments also give
a low return. Conversely, when the interest rate is high, the investor will pay
a relatively low price and earn a relatively high return.
How to Read Information on Treasury Bills

in the Wall Street Journal
The Wall Street Journal reports information on bill maturing on February 9, 1995, the bid dis¬
Treasury bills daily. The information reported count did not change from the previous day.
here is from the Wednesday, February 16, 1994, The final column, Ask Yld., tells the yield to
edition and contains prices from Tuesday, Febru¬ maturity the investor would receive at the asked
ary 15, 1994. discount. For the Treasury bill, the yield to ma¬
The first column, labeled Maturity, gives the turity was 3.82 percent.
date on which the Treasury bill matures. The
second column, labeled Days to Mat., indicates
the days until maturity from the settlement date TREASURY BILLS
of the purchase until the maturity date. For in¬ Days
Ask
to
stance, the last T-bill quoted had 357 days to Maturity Mat. Bid Asked Chg. Yld.
Feb 17 '94 0 3.11 3.01 + 0.16 0.00
maturity from the settlement date. Feb 24 '94 7 2.50 2.40 — 0.13 2.43
Mar 03 '94 14 2.96 2.86 — 0.03 2.90
The third column, labeled Bid, reports the Mar 10'94 21 3.00 2.90 — 0.04 2.95
Mar 17'94 28 2.94 2.90 — 0.07 2.95
highest discount by buyers. It is reported as a Mar 24 '94 35 2.96 2.92 — 0.01 2.97
Mar 31 '94 42 3.01 2.97 — 0.02 3.02
percentage of the face value and then con¬ Apr 07 '94 49 3.10 3.06 — 0.02 3.12
Apr 14 '94 56 3.11 3.07 — 0.01 3.13
verted to an annual rate. Thus, for the last T-bill Apr 21 '94 63 3.19 3.17 3.23
Apr 28 '94 70 3.22 3.20 3.26
quoted, maturing on February 9, 1995, the May 05 '94 77 3.25 3.23 3.30
May 12 '94 84 3.25 3.23 — 0.01 3.30
highest discount bid by a buyer was a 3.68 per¬ May 19'94 91 3.27 3.25 3.32
May 26 '94 98 3.28 3.26 3.33
cent discount from the face value. What does Jun 02 '94 105 3.31 3.29 3.37
Jun 09 '94 112 3.31 3.29 — 0.01 3.37
this mean? A T-bill with a $10,000 face value Jun 16'94 119 3.33 3.31 3.39
Jun 23 '94 126 3.34 3.32 3.41
had a bid discount yield of .0368, which means Jun 30 '94 133 3.35 3.33 + 0.01 3.42
Jul 07 '94 140 3.36 3.34 3.43
the price bid satisfies ($10,000 - PTB)/$ 10,000 Jul 14 '94 147 3.37 3.35 + 0.01 3.44
Jul 21 '94 154 3.38 3.36 + 0.01 3.46
x (360/357) = .0368, or the bid price was Jul 28 '94 161 3.39 3.37 3.47
Aug 04 '94 168 3.40 3.38 3.48
$9,635.07. Aug 11 '94 175 3.40 3.38 — 0.01 3.48
Aug 18'94 182 3.41 3.39 — 0.02 3.50
The fourth column, labeled Asked, is the Aug 25 '94 189 3.44 3.42 + 0.01 3.53
Sep 22 '94 217 3.47 3.45 3.56
lowest discount asked by sellers. On the bill ma¬ Oct 20 '94 245 3.53 3.51 3.63
Nov 17'94 273 3.56 3.54 — 0.01 3.67
turing on February 9, 1995, the asked discount Dec 15'94 301 3.59 3.57 3.70
Jan 12 '95 329 3.64 3.62 3.76
yield of 3.66 percent means the seller offered to Feb 09'95 357 3.68 3.66 3.82
sell the $10,000 T-bill for a price satisfying

($10,000 - PTB/$ 10,000) X (360/357) =
.0366, or $9,637.05. Source: Reprinted by permission of the Wall Street Jour¬
The fifth column, labeled Chg., indicates the nal, © 1994 Dow Jones Company, Inc. All Rights Re¬
percentage change in the bid discount. For the served Worldwide.
ummmmammmm
242 Chapter8 The Time VaLue of Money and Asset Pricing
In the market for loanable funds, the interest rate is determined by the
interaction of demand and supply. Consider the supply and demand curves
for loanable funds given by S] and D] in Figure 8.1. The demanders of
loanable funds are borrowers, and the suppliers are lenders. In this case,
equilibrium is at point A and the market interest rate is 7 percent.
Given the scenario in Figure 8.1, we can determine the present value of
the Treasury bill—the amount a typical investor would be willing to give
up today to receive the promised $10,000 in one year. In particular, when
the interest rate is 7 percent, the price of the Treasury bill (Pjb) is the present
value of the $10,000 received one year in the future:
$10,000
$9,345.79.
1.07
Thus, the maximum price the seller of the Treasury bill could obtain given
the 7 percent market interest rate is PTB = $9,345.79. If the seller asked a
higher price, investors would not purchase it; they could earn a higher return
by investing their money in some alternative investment.
We learned in Chapter 4 that changes in determinants of demand or
supply—factors like wealth, inflationary expectations, and tax rates—can
shift demand or supply, resulting in changes in the interest rate. Table 8.1
When the supply of Figure 8.1

loanable funds in¬ An Increase in the Supply of Loanable Funds
creases from 51 to 52,
the interest rate falls
from 7 to 6 percent. Interest
Rate (%)
s'
i /pv
\
Table 8.1
Variables that Shift the Demand and/or Supply of Loanable Funds
This table summarizes some of the important variables that can shift the demand or supply of loanable
funds. A change in any of these variables will change the equilibrium interest rate and therefore affect
the dollar prices of financial assets such as Treasury bills. Tables 4.1 and 4.2 in Chapter 4 present a
complete list of determinants of the demand and supply of loanable funds.
Shifts the Shifts the

Demand Curve Supply Curve
for Loanable for Loanable
An increase in Funds Funds
Inflationary expectations -> <-

Tax deductibility of household interest payments -> No effect
Size of government budget deficit -> No effect
Tax rate on interest income No effect <-
Wealth No effect ->
summarizes a few of these determinants and their impact on demand and

supply (a more complete list appears in Tables 4.1 and 4.2 in Chapter 4).1
We can use this information to see how a change in a determinant of demand
or supply affects bond prices.
For instance, what do you think would happen to bond prices if wealth
increased? The increase in wealth would increase lenders’ willingness to
loan out funds, thus increasing the supply of loanable funds to S2 in Figure
8.1. In this case, equilibrium moves to point B, and the market interest rate
falls from 7 to 6 percent. Given this lower interest rate, the price of the
Treasury bill rises to
$10,000
$9,433.96.
1.06
The increase in wealth led to a decline in the interest rate, which in turn
increased the price of the Treasury bill.
This example illustrates a general principle: As the interest rate falls, the
price of a debt instrument sold on a discount basis rises. This inverse rela¬
tionship is shown in Figure 8.2, where each point on the horizontal axis
indicates the price (present value) of the one-year Treasury bill at various
1 Other determinants include risk, liquidity, and rates on substitute sources and uses of funds. These
variables are not listed in Table 8.1 because they are not important in our discussion of identical
risk-free loans. In the next chapter, we will see how allowances for risk change the analysis.
244 Chapter8 The Time Value of Money and Asset Pricing
The downward-slop¬ Figure 8.2

ing curve shows the Relationship Between Interest Rate and Price
inverse relationship of a Treasury Bill
between the market
interest rate and the
price of a debt instru¬
ment sold on a dis¬
count basis. As the in¬
terest rate falls from 7
to 6 percent, the price
of this one-year,
$10,000 Treasury bill
rises from $9,345.79
to $9,433.96.
interest rates. The points on this curve correspond with our earlier calcula¬
tions: When the interest rate is 7 percent, the price of the Treasury bill is
$9,345.79; when the interest rate falls to 6 percent, the price of the Treasury
bill rises to $9,433.96.
Coupon Bonds and Other

Interest-Bearing Debt instruments
Unlike discount bonds, some debt instruments—such as coupon bonds issued
by both the U.S. Treasury (Treasury bonds) and corporations (corporate
bonds)—provide regular interest payments in addition to paying their face
value at maturity. The face value of a coupon bond is similar to that of a
discount bond and a Treasury bill: It specifies the amount that will be paid
when the bond matures. In addition, however, the face value of the bond
indirectly determines how much the investor will receive in interest payments
each year up to and including the bond’s maturity date.
In particular, a coupon bond specifies a coupon rate of interest. This
coupon rate determines the percentage of the face value that will be paid
each year as interest. Consider a corporate bond with a face value of $1,000
and a coupon rate of 5 percent. If this bond matures in 20 years, the owner
of the bond will receive 20 coupon payments of .05 X $1,000 = $50 each.2
This annual payment of $50 is in effect an interest payment to the owner of
the bond. In addition to these coupon payments, the bond pays the owner
the face value of $1,000 at the end of 20 years. Note that the coupon rate of
interest is applied to the face value of the bond rather than to its purchase
price.
Box 8.2 explains how to interpret financial data on U.S. Treasury bonds
and notes.
Current Yield and Yield to Maturity. Because the coupon rate

applies to the face value of the bond, it generally does not reflect the actual
yield the owner will receive over the bond’s life. This is because the owner
may pay more or less for the bond than its face value, depending on the
relationship between the coupon rate and the market interest rate. The coupon
rate will exceed the yield on a bond when the price paid exceeds the bond’s
face value; the coupon rate will be less than the yield on a bond when the
price paid is less than the face value. To see why this is the case, we now
look at two measures of the yield on a coupon bond: current yield and yield
to maturity.
The current yield is simply the ratio of the annual coupon payment to
the price paid for the bond:
Annual coupon payment

Current yield = --,
P CB
where PCB is the price paid for the coupon bond. To illustrate, suppose you
pay $950 for a 6 percent coupon bond that has a face value of $1,000 and
matures in one year. Then you will receive an annual coupon payment of
.06 X $1,000 = $60. Since you paid $950 for the bond, the current yield
is
$60
Current yield = -- = .063,
J $950
which is 6.3 percent. The current yield is higher than the coupon rate stated
on the bond (6 percent) because you paid less than the bond’s face value.
Notice that the current yield does not take into account the fact that you
receive $1,000 in addition to the $60 interest payment when the bond ma¬
tures.
The yield to maturity on a coupon bond, like that on a discount bond or
Treasury bill, is the interest rate at which the price paid for the bond would
have to be invested to return the stream of payments generated by the bond.
2 Most coupon bonds actually make interest payments every six months instead of once each year.
Since this does not significantly affect the calculations, we ignore it for simplicity.
Inside Money
How to Read Information on Treasury
Bonds and Notes in the Wall Street Journal
The Wall Street Journal reports information on GOVT. BONOS & NOTES
Maturity Ask
Treasury bonds and notes daily. The information Rate Mo/Yr Bid Asked Chg. Yld.
53/8 Feb 94n 100:02 100:04 1.23
reported here appeared in the Wednesday, Feb¬ 53/4 Mar 94n 100:09 100:11 2.70
8'/2 Mar 94n 100:19 100:21 — i 2.71
ruary 16, 1994, edition and reports prices from 7 Apr 94n 100:18 100:20 2.92
53/8 Apr 94n 100:12 100:14 3.11
Tuesday, February 15, 1994. 7 May 94n 100:27 100:29 3.14
9V2 May 94n 101:14 101:16 3.13
The first column, labeled Rate, gives the 13’/8 May 94n 102:10 102:12 3.07
I
5Va May 94n 100:15 100:17 3.20
coupon rate on the bond or note. The second 5 Jun 94n 100:17 100:19 3.34
8V2 Jun 94n 101:26 101:28 3.30
column, Maturity Mo/Yr, gives the date of ma¬ 8 Ju! 94n 101:25 101:27 3.40
4>/4 Jul 94n 100:09 100:11 3.47
turity and indicates whether the instrument is a 67/8 Aug 94n 101:19 101:21 3.47
8Vs Aug 94n 102:14 102:16 3.48
note (denoted n) or a bond. The bid and asked 83/4 Aug 94 102:16 102:18 3.48
12Vs Aug 94n 104:13 104:15 3.43
prices are reported next. Prices are given in 41/4 Aug 94n 100:10 100:12 3.53
4 Sep 94n 100:06 100:08 3.59
number of dollars per $100 of face value, with 8’/2 Sep 94n 102:29 102:31 3.59
9V2 Oct 94n 103:22 103:24 3.68
"decimals" reporting 32nds of a dollar. The last 4'/4 Oct 94n 100:11 100:13 3.66
6 Nov 94n 101:20 101:22 + i 3.67
two columns report the change in the bid price 8'/4 Nov 94n 103:07 103:09 3.72
lOVe Nov 94 104:19 104:21 3.70
and the yield to maturity, annualized, calculated 115/8 Nov 94n 105:22 105:24 3.69
45/8 Nov 94n 100:21 100:23 + i 3.69
from the asked price. 4s/8 Dec 94n 100:21 100:23 3.78
75/8 Dec 94n 103:06 103:08 — i 3.78
As an example, consider the last note listed. 85/e Jan 95n 104:06 104:08 3.83
4V4 Jan 95n 100:11 100:13 3.81
The coupon rate is 10V2, or 10.5 percent, and 3 Feb 95
5V2 Feb 95n
99:07 100:07
101:17 101:19
+ i 2.78
3.85
the note matures in August 1995. The bid price 73/4 Feb 95n 103:23 103:25 3.84 s
IOV2 Feb 95 106:12 106:14 — i 3.84
was 109:03, or $1093/32 Per $100 of face value; 11 >/4Feb 95n 107:05 107:07 3.79
37/8 Feb 95n 99:30 100:00 3.88
that is, $1,090.94 was bid on each $1,000 in 37/8 Mar 95n 99:28 99:30 3.93
83/8 Apr 95n 104:28 104:30 + i 3.97
face value. The asked price was 109:05, so 37/8 Apr 95n 99:26 99:28 3.98
57/8 May 95n 102:05 102:07 4.03
$1,091.56 was asked on each $1,000 in face 8V2 May 95n 105:10 105:12 4.02
103/8 May 95 107:19 107:21 — i 3.99
value. The yield maturity on this note, annual¬ 10/4 May 95n 108:21 108:23 — 2 3.98
125/8 May 95 110:11 110:15 3.90
ized, is 4.12 percent. 4'/8 May 95n 100:02 100:04 + i 4.02
4Ve Jun 95n 100:01 100:03 + 1 4.05
87/8 Jul 95n 106:13 106:15 4.10
4'/4 Jul 95n 100:04 100:06 + i 4.12
Source: Reprinted by permission of the Wall Street Jour¬ 45/8 Aug 95n 100:20 100:22 + 1 4.15
8V2 Aug 95n 106:06 106:08 4.14 i
nal, © 1994 Dow Jones Company, Inc. All Rights Re- 10V2 Aug 95n 109:03 109:05 4.12
served Worldwide.
The yield to maturity on a coupon bond takes into account not only the face
value paid on maturity but also the annual coupon payments received during
its term. More concretely, having paid $950 today for the coupon bond, in
one year you will receive a $60 coupon payment, plus the $1,000 face value.
Thus, the yield to maturity on this coupon bond is the interest rate, i, that
satisfies the following condition:
^ $60 + $1,000
$950 = --—-.
1 + i
Solving this expression for i reveals that the yield to maturity on this bond
is 11.58 percent, which is considerably higher than both the coupon rate
stated on the bond (6 percent) and the current yield (6.3 percent). The reason
is that the yield to maturity takes into account the money received from both
the coupon payments and the face value payment. Since the face value of
the bond exceeds the price paid for the bond, the yield to maturity is greater
than both the coupon rate and the current yield. In contrast, if the face value
of the bond were less than the bond’s price, the yield to maturity would be
lower than the coupon rate and the current yield.
Valuing Coupon Bonds. How do financial intermediaries value cou¬

pon bonds? The price of a coupon bond reflects the present value of all
future payments the bondholder receives. A coupon bond that pays its face
value (F) in T years also generates a coupon payment (PMT) at the end of
each year until maturity. If the market interest rate is i, the price of a coupon
bond (PCb) will be
_ PMT PMT PMT PMT + F

PcB ~ (1 + O' + (1 + if + (1 + if + ' + (1 + i)T '
Notice that at the end of year T, when the bond matures, the owner receives
a coupon payment plus the face value of the bond.
Suppose a coupon bond pays $100 each year for two years, plus the face
value of $1,000 at the end of the second year. In this case, the owner will
receive $100 at the end of the first year and $1,100 at the end of the second
year ($100 in interest, plus the payment of the face value). The price of the
coupon bond will depend on the market interest rate as determined by the
supply of and demand for loanable funds. For instance, suppose the supply
of and demand for loanable funds are given by Sl and D] in Figure 8.3.
Point A denotes the market equilibrium; the interest rate is 8 percent. What
is the maximum price you would be willing to pay for the bond?
Given the 8 percent interest rate, the price of the coupon bond (PCB) is
the present value of all payments the owner receives:
_ $100 $1,100
PcB ~ 1.08 + 1.082 ’
which is $1,035.66. In other words, given a market interest rate of 8 percent,
you would be willing to pay up to $1,035.66 for this coupon bond. This
shows that an investor is sometimes willing to pay more than face value for
a bond. Again, this occurs because the coupon rate, 10 percent, exceeds the
market interest rate, 8 percent.
If the market interest rate changes, so will the price of the bond. For
example, suppose the government needs to borrow additional loanable funds
to finance the deficit. This causes the demand for loanable funds in Figure
8.3 to increase to D2. The increased demand for loanable funds drives up
the equilibrium interest rate to 10 percent at point B. Given this higher
interest rate, the price of the coupon bond falls to
$100 $1,100
Pcb = —- + ,
c 1.10 1.102
or $1,000. Thus, as the interest rate rises, the price of the coupon bond falls,
as the solid curve in Figure 8.4 shows. Here each point on the horizontal
axis indicates the price (present value) of a coupon bond at various interest
rates. When the interest rate is 8 percent, the price of the two-year bond is
$1,035.66. When the interest rate rises to 10 percent, the price of the bond
falls to $1,000. Thus, an inverse relationship exists between the price of a
coupon bond and the market interest rate, just like that between the interest
rate and the price of a discount bond. An investor who owns this two-year
coupon bond would see it decline in value due to the higher interest rate
caused by increased federal borrowing to finance the deficit.
The solid curve shows Figure 8.4

the relationship be¬ Relationship Between Interest Rate and Prices of
tween the interest rate 10 Percent Coupon Bonds with Different Maturities
and the price of a
two-year coupon
bond, while the
dashed curve shows
the relationship be¬
tween the interest rate
and the price of a
three-year coupon
bond. Both bonds
have a face value of
$1,000 and a coupon
rate of 10 percent. As
the market interest
rate rises, the prices of
both bonds fall, but
the price of the three-
year bond falls more
rapidly than that of
the two-year bond.
In using Figure 8.4, it is important to remember that the solid curve is

drawn holding constant both the future payments and the maturity of the
bond (the date on which the bond issuer pays back the face amount). The
dashed curve in Figure 8.4 shows the inverse relationship between the in¬
terest rate and the price of a coupon bond that matures in three years instead
of two. The $1,000 face value and the 10 percent coupon rate on the three-
year bond are identical to those on the two-year bond; all that differs are the
maturity dates. Because of the longer maturity, the effect of a change in the
market interest rate on the price of the three-year bond is greater than that
for the two-year bond. When the interest rate is 10 percent, the price of both
bonds is $1,000, since the coupon rate on both is also 10 percent. If the
interest rate falls to 8 percent, the price of the two-year bond rises to
$1,035.66 and the price of the three-year bond rises to $1,051.54. Similarly,
when the interest rate rises from 10 to 12 percent, the price of the two-year
bond falls to $966.20 and the price of the three-year bond falls further to
$951.96. A change in the market interest rate has a greater effect on the
prices of bonds with a longer remaining time to maturity.
Box 8.3 describes how to read financial information on corporate bonds.
Inside Money
How to Read Information on Corporate
Bonds in the Wall Street Journal
mmsmm
The Wall Street Journal reports information on had a closing price equal to 1093/s percent of its
corporate bonds daily. The information reported face value, or $1,093.75 for a bond with a
here is from the Wednesday, February 16, 1994, $1,000 face value.
edition and reports prices from Tuesday, Feb¬ The final column reports the change in the
ruary 15, 1994. closing price between the previous business day
The first column, labeled Bonds, gives the (February 15) and the day before that (February
name of the company issuing the bond, fol¬ 14). It too is expressed as a percentage of the
lowed by the coupon rate, then by the year in face value. The ATT bond was up five-eighths of
which the bond matures. If a zr appears where a point, or .00625 x $1,000 = $6.25.
the coupon rate should be, it means this bond is Note how the current yield is calculated
a zero coupon bond. with this information. The purchase price is
The second column gives the current yield $1,093.75, and the annual coupon is $81.25, so
on the bond. Consider the last ATT bond listed. the current yield is $81.25/$ 1,093.75 = 0.074,
This bond pays a coupon rate of 8VS (or 8.125 or 7.4 percent.
percent) and matures in 2024; ATT pays $81.25
CORPORATION BONDS
per year in coupon interest payments, for each Volume, $33,870,000
$1,000 bond. The current yield of 7.4 percent Cur Net
Bonds Yld vol Close Chg.
means that if the bond was purchased on Feb¬ AForP 5s30 7.3 63 68 Va + %
Actava 9'/298 9.5 20 100 - >/4
ruary 15, 1994, the coupon interest accruing Actava 10s99 9.8 24 101 % - '/4
...
AlskAr 6% 14 cv 84 87'/2
AlskAr zr06 218 40% + %
from it would represent an interest rate of 7.4 Albnylnt 5s02 cv 4 96 - 1
AlldC zr98 7 76% - Vs
percent of the price paid. AlldC zr97 100 83% - Vt
AlldC zr07 ... 5 38% ...
The third column gives the volume of bonds AlegCp 6V2I4 cv 10 103 ...
AAirl 5’/d98 5.5 20 95
traded the previous business day. It is expressed ATT 4%96 4.4 30 99% - Vs
ATT 6s00 6.0 375 100 %
in terms of the number of $ 1,000 bonds traded. ATT 8s/s31 7.5 20 114'/4 - Vi
ATT 7Vs02 6.7 90 106'/8 + Va
Thus, 101 of these ATT bonds were traded the ATT 8Va22 7.5 30 109 ...
ATT 8Va24 7.4 101 109% + %
previous day.
The fourth column tells the price of the
bond when trading stopped on the previous
Source.: Reprinted by permission of the Wall Street Jour¬
business day. It is expressed as a percentage of nal, © 1994 Dow Jones Company, Inc. All Rights Re¬
the face value of the bond. Thus, this ATT bond served Worldwide.
Summary Assuming the market interest rate is 9 percent, determine the price and yields
Exercise 8.2 of the following: (a) a $10,000 Treasury bill that matures in one year and
(b) a $1,000 corporate bond that has a coupon rate of 10 percent and matures
in one year.
The Equilibrium Price and Quantity of Bonds 251
(a) The price of the Treasury bill is
$10,000
$9,174.31.
1.09
The discount yield on this bond is
10,000 - 9,174.31 360

Discount yield = X = .081,
10,000 365
which is 8.1 percent. The yield to maturity is the interest rate, i, that satisfies
$10,000
$9,174.31
1 + i
Solving for i, the yield to maturity is 9 percent, the same as the market
interest rate used in our present value calculations. This is to be expected,
of course, since we used this interest rate in determining the price of the
Treasury bill.
(b) The annual coupon payment for this bond is $100. This coupon
payment, along with the face value, will be received in one year. Thus, the
price of the corporate bond is the present value of these payments when the
interest rate is 9 percent:
$100 + $1,000
Pcb — $1,009.17.
1.09
Notice that this amount is greater than the face value of the bond because
the coupon rate of interest ($100/$ 1,000 = 10 percent) is greater than the
market rate of interest (9 percent). The current yield on this bond is
$100
Current yield - -- = .099,
J $1,009.17
or 9.9 percent. Since the price paid for this bond is based on the market rate
of interest, the yield to maturity is the market rate of interest, 9 percent.
Notice that the current yield is greater than the yield to maturity because the
price of the bond exceeds the bond’s face value.
The Equilibrium Price and Quantity of Bonds
Our analysis of discounted corporate bonds and Treasury bills, as well as

bonds with coupons such as Treasury bonds and corporate bonds, has re¬
vealed an important principle: An inverse relationship exists between the
price of a bond and the market interest rate. For this reason, it is important
to understand how changes in the economy affect interest rates and thus
bond prices. We now introduce two different but equivalent methods of
analyzing bond markets. One approach relates the interest rate to the supply
of and demand for loanable funds; the other relates the prices of bonds to
the supply of and demand for bonds. We also show the equivalence of the
two approaches by demonstrating that for a given interest rate, the price of
a bond can be determined from our graph of the inverse relationship between
bond prices and interest rates.
Bond Prices: The Loanable Funds Approach

One way to analyze bond prices is to combine what we have already learned
about interest rate determination and present value analysis into a convenient
framework. Parts a and b of Figure 8.5 illustrate this approach. Part a depicts
the supply of and demand for loanable funds, with the interest rate on the
vertical axis and the dollar value of loanable funds on the horizontal axis.
The supply of loanable funds is provided by lenders, who are willing to offer
more loanable funds at higher interest rates. The supply of loanable funds is
therefore an upward-sloping curve. Borrowers, on the other hand, seek more
funds as the interest rate falls. Thus, the demand curve for loanable funds
slopes downward. Equilibrium is determined by the intersection of the supply
and demand curves for loanable funds and changes with fluctuations in either
Figure 8.5
Changes in the Supply of Loanable Funds and Bond Prices
In part a, the market interest rate is determined in the loanable funds market. With de¬
mand of D1 and supply of S\ the market interest rate is 10 percent. In part b, the price of
a 10 percent coupon bond is $ 1,000 when the market interest rate is 10 percent. How¬
ever, when the supply of loanable funds increases in part a to 52, the market interest rate
falls to 8 percent. In part b, this causes the price of the bond to rise to $1,035.66.
Loanable Funds 1,000 1,035.66 Bond Price ($)

(Millions $)
(a) Loanable Funds (b) Bond Prices

supply or demand. For example, when the supply curve is Sl, equilibrium is
at point 1 and the interest rate is 10 percent. If the supply of loanable funds
increases to S2, as would occur if the tax rate on interest income were
reduced, equilibrium moves to point 2 and the equilibrium interest rate falls
to 8 percent.
Part b of Figure 8.5 illustrates the inverse relationship between the
interest rate and the price of a bond. When equilibrium is at point 1 in part
a, the interest rate is 10 percent. Tracing this interest rate over to part b, we
see that the corresponding price of a 10 percent coupon bond is $1,000.
When equilibrium in part a shifts to point 2, the interest rate falls to 8
percent. Tracing this over to part b, we see that the price of the bond rises
to $1,035.66.
Figure 8.5 shows a graphical technique for determining the impact of
changes in the supply of and demand for loanable funds on bond prices. A
change in any of the determinants of demand or supply listed in Table 8.1
that results in a higher interest rate in part a will lead to lower bond prices
in part b. Conversely, any change in demand or supply that leads to a lower
interest rate in part a will lead to higher bond prices in part b. Thus, bond
prices will systematically fluctuate with changes in such variables as infla¬
tionary expectations, taxes, budget deficits, and wealth.
Box 8.4 uses a real-world example to illustrate the inverse relationship
between bond prices and interest rates.
Bond Prices: The Suppy and Demand of Bonds Approach

An alternative way to determine the equilibrium prices of bonds is to directly
graph the demand and supply curves for bonds as a function of a bond’s
price instead of using the two-part approach used in the previous section.
We will base our analysis in this section on the same scenario that existed
at point 1 in part a of Figure 8.5 to emphasize the equivalence of this
approach to the loanable funds approach. Recall that at point 1 in that figure,
the interest rate is 10 percent, the coupon rate is 10 percent, and the corre¬
sponding price of a bond in part b of Figure 8.5 is $1,000.
Since a bond is a promise to pay a set amount over the life of the bond
and the coupon rate is assumed to be 10 percent, the lower the price of the
bond, the higher the bond’s yield to maturity. In turn, the higher the yield
to maturity relative to the market interest rate, the more attractive the bond
is to investors. For this reason, the demand for a bond is a decreasing function
of the bond’s price, as illustrated by the curve DB in Figure 8.6, on page
255. Notice that this demand curve holds everything constant except the
price and quantity of bonds sold. The demand curve for bonds is composed
of those wishing to give up money today in return for the future stream of
funds promised by the $1,000 bond with a coupon rate of 10 percent. The
lower the price of the bond, the more bonds investors will want to hold.
The Inverse Relationship Between

Bond Prices and Interest Rates
This table reports the asked prices and yields to bond. Second, despite the differences in cou¬
maturity for each day during the first week of pons, the yields to maturity on these two bonds
June 1993, for two $10,000 government bonds were very similar. Thus, the market price ad¬
maturing in June 1994. The first bond carried a justed for the difference in coupon rates. Third,
5 percent coupon rate and the second an 8.5 the asked price and the yield to maturity varied
percent coupon rate. Notice several things. First, over the week. When the asked price increased,
the price of the bond with the 8.5 percent cou¬ the yield to maturity decreased, and vice versa.
pon interest rate was higher than the price of This example illustrates the inverse relationship
the 5 percent coupon rate bond, reflecting the of bond prices and yields for an average week.
higher coupon payments on the 8.5 percent
5% Coupon Rate 8.5% Coupon Rate

Date
Asked Price Yield Asked Price Yield
Fri., 5/28/93 $10,140.63 3.66% $10,509.38 3.64%
Tue., 6/1/93 10,143.75 3.62 10,512.50 3.60
Wed., 6/2/93 10,140.63 3.65 10,512.50 3.59
Thur., 6/3/93 10,143.75 3.61 10,512.50 3.55
Fri., 6/4/93 10,128.13 3.76 10,490.63 3.74
Source: Bond prices calculated on a one-year, $10,000 Wall Street Journal, June 1, 2, 3, 4, and 7, pp. Cl 6, Cl 9,
government bond maturing in June 1994. Data from the Cl9, Cl4, and Cl4, respectively.
The suppliers of bonds, on the other hand, wish to sell the right to a
specified amount over the life of the bond. Since the seller of a bond receives
money today in return for future payments promised over the bond’s life,
the higher the price of a given bond, the more lucrative it is to sell the bond,
holding other things constant. For this reason, the supply of bonds is an
increasing function of the price of a bond, as illustrated by the curve SB in
Figure 8.6. Notice that this supply curve holds everything else constant; all
that varies along the supply curve is the price and quantity of bonds.
Equilibrium in the bond market is determined by the intersection of the
supply and demand for bonds, such as at point A in Figure 8.6. In this
The quantity de¬ Figure 8.6

manded of bonds will Equilibrium Price and Volume of Bond Transactions
increase as the price
of bonds decreases,
holding other things
constant. The quantity
supplied of bonds will
increase as the price
of bonds increases,
holding other things
constant. Equilibrium
in the bond market
occurs at point A,
where demand equals
supply. When the
market interest rate is
10 percent, the equi¬
librium price of 10
percent coupon bonds
is $1,000. In this case,
4,000 bonds trade in
the market.
4,000 Quantity of Bonds

(Number Traded)
example, the equilibrium price of bonds is $1,000 and the equilibrium quan¬
tity of bonds is 4,000. Notice that this equilibrium price of bonds is precisely
the same as that in part b of Figure 8.5 when the market interest rate is 10
percent.
This analysis applies to both the primary and secondary markets. When
bonds trade in the primary market (the market for new bonds), those de¬
manding bonds are providing loanable funds to those issuing the bonds.
When bonds trade in the secondary market (the market for existing bonds),
the demander of a bond is providing funds to the party wishing to sell an
existing bond. Likewise, in the primary market, the suppliers of bonds are
corporations and governments. These parties issue bonds to obtain loanable
funds. In the secondary market, the suppliers of bonds are those who own
existing bonds and wish to sell them to obtain funds immediately.
Summary (a) Determine the impact of an increase in inflationary expectations on

Exercise 8.3 interest rates and bond prices, (b) Suppose the government passed a law that
exempted interest income from federal taxes but the change left the size of
the budget deficit unchanged. What would be the impact on the interest rate
and bond prices? Illustrate your answers graphically.
(a) The initial equilibrium in the loanable funds market is at

point 1 in part a below, where S° and D° intersect. Given the initial interest
rate of i°, the price of a bond is given by PB in part b. The increase in
inflationary expectations decreases the supply of loanable funds from S° to
S1 in part a, since lenders now expect the future interest receipts and face
value received at maturity to have less purchasing power. Similarly, in¬
creased inflationary expectations increase the demand for loanable funds
from D° to D] in part a, since borrowers now expect to be able to pay back
lenders with money that has less purchasing power than before. Thus, equi¬
librium moves from point 1 to point 2 in part a, and the interest rate rises to
il. In part b, the higher interest rate translates into a decrease in bond prices
from PB to PB. Thus, increased inflationary expectations lead to higher
interest rates and lower bond prices.
Interest Interest
Rate Rate
(a) Loanable Funds
(b) The initial equilibrium in the loanable funds market is at point 1 in

part a below, where the interest rate is i\ Part b reveals that initially bond
prices are given by PB. When the government announces that interest income
on bonds is exempt from federal taxes, the supply of loanable funds increases
Interest Interest
Rate Rate
(a) Loanable Funds (b) Bond Prices

Valuing Stocks and Other Assets 257
from Sl to S2 in part a. This is because at any interest rate, bondholders pay

less taxes on bond income than before and are thus willing to lend more
funds. Hence, equilibrium moves from point 1 to point 2 and the interest
rate falls to i2 in part a. In part b, this translates into an increase in bond
prices to P\. Thus, when bond income is exempted from federal taxes, the
result is lower interest rates, higher bond prices, and a greater equilibrium
quantity of loanable funds.
Valuing Stocks and Other Assets

Unlike debt instruments, equity instruments (such as common stock) rep¬
resent the ownership of a share of a firm. Accordingly, the owner of a share
of common stock owns a share of the future earnings (profits) of the company
issuing the stock. Individuals, insurance companies, mutual funds, and pen¬
sion funds hold stocks in their portfolios and continually seek out stocks that
are undervalued. Most banks and bank holding companies are also owned
by shareholders; you can buy a share of Chase Manhattan Bank, Chemical
Bank, or Mellon Bank on the New York Stock Exchange. Box 8.5 describes
how to interpret financial data for corporate stocks.
When is a stock a good buy? To answer this question, we must determine
the value of a share of stock.
When a bank or other company has earnings, it can either (1) return
them to shareholders in the form of dividends or (2) retain them within the
firm. Dividends represent a direct payment to shareholders. Earnings that
are retained by the firm increase the value of the firm in that they can either
be invested in projects within the firm that will enhance future earnings or
be invested elsewhere at the market interest rate and be paid out as dividends
in the future.
The value of a firm at any point in time is the present value of all of the
firm’s future earnings:
EE\ EE2 EEr

Value of a firm (8.3)
(l + 0 + (l + j)2 + ' " + (i + ,y’
where EEt represents the firm’s expected earnings at the end of year t, i is
the interest rate, and T is the number of years over which the firm will
operate.
Two aspects of corporate stock make stock valuation more difficult than
valuing, say, a U.S. Treasury bond. First, if you purchase a U.S. Treasury
bond, you purchase a stream of known future payments. When you purchase
a stock, you purchase a share of the company’s future earnings, which are
far from known. Second, the maturity of a bond is known. The life of a firm,
T, is not known; the firm may live on forever, or it may go bankrupt in one
year. These elements make determining the value of a firm a complicated
Inside Money
How to Read Information on Stocks
in the Wall Street Journal
The Wall Street Journal reports information on est, lowest, and closing price of the stock for
stock prices daily. The information reported here the day. For Du Pont, the high price was
is from the Wednesday, February 16, 1994, edi¬ $55,125, the low price was $54.50, and the
tion and reports prices from the previous day. closing price was $55. The last column indicates
The Hi Low columns give the highest and the difference between the closing price on that
the lowest price, respectively, of the stock over day and the closing price on the previous day.
the last 52 weeks. The Stock column reports the Thus, the closing price of Du Pont stock on Feb¬
name of the issuing company. The Sym column ruary 15 was five-eighths of a dollar (62.5 cents)
indicates the company's ticker symbol. higher than the closing price of the stock on
The Div column tells the current annual divi¬ February 14th.
dends per share. Thus, for Du Pont, each share
pays annual dividends of $1.76. The column la¬ 52 Weeks YW Vel Net
Hi Lo Stock !5ym Div % PE 1«s Hi Lo Close Cits
beled Yld % reports the annual dividend yield, 64% 39V8 Diebold DBD 1.20 2.3 21 1797 51% 51 51% + %
49V4 28% DigitalEqp DEC dd 5136 29% 29% 29% — %
calculated as a percentage of the price of a 48% 33Ve DillardStrs DOS .08 2 16 2549 35 34% 35 + %
9% 6Vb DimeSvgNY DME 9 1684 8% 8% 8%
share. It is calculated as Div divided by the clos¬
31 Vs 20% DiscountAuto DAP 26 80 25% 24% 24% - %
ing price (the Close column). For Du Pont the 48% 36 Disney DIS 30f & 34 9318 46% 46% 46% + %
37% 25% DoleFood DOL .40 1.3 23 1087 30% 30 30% — %
dividend yield was $1,76/$55, or 3.2 percent. 49V2 41 Vs DominRes D 2.54f 6.0 13 1972 42% 42% 42% + %
The column labeled PE is the price-earnings 6% 4V4 Domtar g DTC 227 6% 6% 6% + %
47% 33V4 Donaldson DCI 56f 1.2
22 44 47% 46% 46% —
%
ratio. It is the closing price divided by the earn¬ 31% 26% Donelley DNY .56 1.8
27 2459 31% 30% 31% + %
61 % 44% Dover DOV .92 1.5
22 682 61% 60% 61% +11%
ing per share from the most recent four quar¬ 65 49 DowChem DOW 2.60 4.1
27 5099 64% 63% 64 + %
ters. For Du Pont, the price of a share is 66 40% 26% DowJones DJ 84f 27
2.1 2180 40% 39% 40 — %
27% 14V4 DowneySL DSL ,48f 2.6
10 60 18% 18% 18% - %
times as large as the earnings per share over the 26 141/4 DrPepper DPS 17 1130 25% 24% 25 - %
12V2 8% Dravo DRV 25 344 12% 12 12% + %
most recent four quarters. Since the price of a 25% 17% Dresserlnd Dl .60 Z6 25 8942 22% 21% 22% + %
share is $55, this means earnings per share were 48V2 35% Dreyfus DRY .76 1.6 17 1278 46% 46% 46%
12% 10 DreyfStrGvFd DSI .90 8.4 139 10% 10% 10% + %
$55/66 = $0.83 per share. 11 9% DreyfStrMunBd DSM .66 6.5 348 10% 10 10% + %
IIV2 IOV4 DreyfusMuni LEO .73 6.5 275 11% 11% 11% + %
The eighth column, Vol 100s, is the number 56% 44V2 DuPont DO 1.76 3.2 66 9369 55% 54% 55 + %
of shares traded the previous business day ex¬

pressed in hundreds of shares. Thus, 936,900
Source: Reprinted by permission of the Wall Street Jour¬
shares of Du Pont stock were traded on Febru¬ nal, © 1994 Dow Jones Company, Inc. All Rights Re¬
ary 15. The next three columns report the high¬ served Worldwide.
forecasting exercise. However, once this forecast of future earnings is made,

determining the value of the price of a share of stock is a relatively straight¬
forward process.
In particular, the price of a share of stock is simply the value of the firm
divided by the number of shares of stock issued by the company:
Value of the firm

Stock price = --———. (8 4)
Number of shares
Notice in equation 8.3 that the higher the interest rate, the lower the value
of the firm. Holding the number of shares of the stock constant, this implies
a lower stock price in equation 8.4. Thus, holding everything else constant,
the higher the interest rate, the lower the price of a given firm’s stock.
We now examine in detail the valuation of two particular types of stocks:
income stocks and growth stocks. Income stocks are the stocks of companies
that have a low level of retained earnings and thus pay most of their earnings
to shareholders. Earnings paid to shareholders are called dividends. An
investor who purchases an income stock primarily purchases a future stream
of dividend payments. In contrast, growth stocks are the stocks of companies
that retain most of their earnings to reinvest them within the firm. The
purchaser of a growth stock essentially purchases a company that acquires
more and more assets today to pay higher dividends in the future.
Income Stocks
Income stocks have a low level of retained earnings and thus pay sharehold¬
ers yearly dividends that are a very large fraction of the firm’s earnings. As
a consequence, income stocks tend to have relatively high dividend yields.
The dividend yield on a stock is simply the ratio of the annual dividends
per share to the stock price:
Annual dividends per share

Dividend yield =
where Ps is the price of a share of the stock. For example, suppose a firm’s
stock sells for $10 per share and the firm pays out annual dividends of 60
cents per share. Then the stock’s dividend yield is $0.60/$ 10, or 6 percent.
The stocks of most electric utilities are income stocks; most of their earnings
in any given year are paid out as dividends.
In practice, the valuation of an income stock can be difficult because the
earnings and dividends paid out may vary considerably over time. While we
postpone an analysis of risk until the next chapter, we can gain a basic
understanding of how the price of an income stock is determined by making
a few simplifying assumptions. Suppose a company does not retain any of
its earnings but instead pays them all out as dividends. Furthermore, suppose
the market for the firm’s product is stable and annual earnings are expected
to remain constant at a level of EE from now on. In this case, the firm is
260 Chapter8 The Time V alue of Money and Asset Pricing
expected to pay out dividends per share of d = EE/N each year, where N is
the number of shares of stock issued by the firm. Effectively, the ownership
of such a stock is a perpetuity that pays dividends per share of d = EE/N
in every year from now on. The value of this perpetuity is the price of a
share of this income stock (PIS)'-
Annual dividends per share d
Interest rate i
As an example, suppose an electric utility company is expected to have
earnings of $1 million each year from now on and has issued 500,000
shares of stock. Thus, annual dividends per share are expected to be
d — $1,000,000/500,000 = $2 per share. If the market interest rate is 5
percent, the price of a share of this utility stock will be
$40.
Growth Stocks
Growth stocks differ from income stocks in that most or all of the firm’s
current earnings are reinvested in the firm. Consequently, the dividend yield
on a growth stock will be considerably lower than that on an income stock;
in fact, the stock may not pay current dividends at all (a zero dividend yield).
Nonetheless, since the value of the firm is the present value of all future
earnings and earnings invested back into the firm increase the firm’s value,
considerable growth in earnings will occur over time. As the firm matures,
it ultimately begins to pay dividends to stockholders—much larger divi¬
dends, in fact, than it would have paid during its early years. Usually growth
stocks are shares of relatively young companies.
Like valuing income stocks, valuing a growth stock can be difficult in
practice because the firm’s future earnings are far from certain. Again we
can use a few simplifying assumptions to gain some basic insights into the
pricing of growth stocks without getting bogged down with risk consid¬
erations.
Suppose a firm’s earnings last year were EE, as investors expected. These
earnings are expected to grow at a rate of g each year from now on. Thus,
at the end of this year, expected earnings are EEX = (1 + g)EE, at the end
of next year they are expected to be EE2 = (1 + g)2EE, at the end of three
years they are expected to be EE3 = (1 + g)3EE, and so on. The present
value of the growth firm in this case is given by
Value of growth firm

= EEl + EEl + EE°> + = V=o EEt
(1 + i) (1 +i)2 (1 + ;)3 (1 + if
Using the fact that EEt = (1 + g)rEE, we can simplify this to

t
1 + 8
Value of growth firm EE X xr=i
1 + i
If expected earnings growth exceeded the interest rate (g > /), this expression
would be infinite, reflecting the fact that the firm is infinitely valuable. Of
course, this is unlikely since a firm cannot continue to grow forever at a rate
that exceeds the market rate of interest. The more plausible case is that where
the expected growth rate is less than the market rate of interest.
When g < i, the formula for the value of a growth firm can be simplified
to3
1 + 8
Value of growth firm = EE X
i ~ g
If N shares of stock are outstanding, the price of a share of the firm’s stock
will be the fraction 1 IN of the value of the firm. In other words, the price of
a growth stock will be
EE 1 + 8
X
N i ~ g
where EE is expected earnings last year, g is the expected annual growth in
those earnings, i is the interest rate, and N is the number of shares of stock
outstanding. For example, suppose a firm has expected earnings of $1 million
and these earnings are expected to grow at an annual rate of 3 percent from
now on. If 500,000 shares of stock are outstanding and the interest rate is 7
percent, the price of a share of this firm’s stock will be
_ $1,000,000 1 + .03
= $51.50.
GS ~ 500,000 .07 - .03
3 To see why
"V 00 1 + g 1 + g
A= i
1 + i i ~ g’
let 8 = (1 + g)/( 1 + i). It is a well-known property of infinite series that
xr=i ^ =
whenever |8| < 1. Since (1 + g)/(\ + i) plays the same role as 8 in this formula, it follows that
-it
1 + g 1 + g
Xoo 1 + g 1 + i 1 + i 1 + g
t= l i
1 + + g) 1 + i - 1 - g i - g’
(1+t) 1 + i
as required.
If the interest rate falls to 5 percent, the stock price will increase to
_ $1,000,000 1 + .03
= $103.
cx ~ 500,000 .05 - .03
This example illustrates that the price of a growth stock increases as the
interest rate declines; an inverse relationship exists between the interest rate
and the price of a share of growth stock. Similarly, the greater the growth
rate in earnings, the higher the price of a growth stock.
Summary Suppose a firm’s earnings were $5 million last year. There are 5 million
Exercise 8.4 shares of the company’s stock outstanding, and the market interest rate is 4
percent. Determine the price of a share of this firm’s stock if (a) earnings
are expected to remain constant forever and the firm pays all earnings out
as dividends and (b) the firm’s earnings are expected to grow at an annual
rate of 3 percent.
Answer: (a) Earnings per share are expected to be $5,000,000/5,000,000

= $1 and paid out as dividends each year forever. Thus, by purchasing a
share of this firm, the investor buys a perpetual stream of $1 dividend
payments. The price of this income stock is thus
$25.
(b) Using our formula for the price of a growth stock (here EE = $5,000,000,
N = 5,000,000, g — .03, and i = .04), we find the price of a share of the
growth stock to be
$5,000,000 1 + .03
= $103.
|I 5,000,000 04 - .03
Equilibrium Price and Volume of Stock Transactions
Our analysis of the valuation of stocks showed an inverse relationship be¬

tween the price of a share of stock and the market interest rate, just as was
the case for bond prices. Changes in the economy thus affect stock prices in
much the same way that they affect bond prices.
Part a of Figure 8.7 depicts the supply of and demand for loanable funds,
with the interest rate on the vertical axis and the quantity of loanable funds
on the horizontal axis. When the supply curve is S] and the demand curve
is D], equilibrium is at point 1 and the interest rate is 7 percent. To find the
price of a share of stock, we trace this interest rate over to part b, where we
graph the inverse relationship between the interest rate and stock prices.
Equilibrium Price and Volume of Stock Transactions 263
Thus, at the 7 percent equilibrium interest rate in part a, the corresponding

stock price is $51.50 in part b.
If the demand for loanable funds decreases to D2 in part a of Figure 8.7,
the new equilibrium is at point 2 and the market interest rate falls to 5
percent. Tracing this interest rate over to part b, we see that as a result of
the decline in the interest rate, the price of a typical share of stock rises to
$103. This method of determining the impact of changes in the supply of
and demand for loanable funds on stock prices is virtually identical to the
way we analyzed the determination of bond prices. Any change in demand
or supply in part a of Figure 8.7 that results in a lower interest rate will lead
to higher stock prices in part b, and vice versa.
We can also graph the demand and supply curves for the stock as a
function of its price to determine the stock’s equilibrium price. Since the
ownership of a share of stock represents a share of the future earnings of the
firm, holding constant other things such as the interest rate, the lower the
price, the more desirable it is to purchase the stock. Consequently, graphing
the demand for a stock as function of its price yields a downward-sloping
demand curve like DStock in Figure 8.8. Notice that the interest rate is held
constant at 7 percent along this demand curve.
On the other hand, the higher the price that can be received by selling a
share of stock, the greater the amount of stock that will be offered for sale.
Thus, the supply curve for a stock is an increasing function of the stock’s
price, such as SStock in Figure 8.8. The intersection of the supply and demand
curves determines equilibrium in the stock market. Point A denotes the
equilibrium price, $51.50, and the equilibrium quantity sold, 5,000 shares.
Summary Suppose you read that the growth in Beta Company’s earnings will be lower
Exercise 8.5 than expected. Looking at the stock market—the secondary market in
stocks—and assuming other things remain the same, what would you expect
to happen to the price of the stock? To the number of shares traded? Illustrate
your answer with a graph.
One way to answer this question is to recall that the price of a

share of stock is an increasing function of the growth in the firm’s earnings.
Thus, a decline in the growth rate of earnings will reduce the present value
of the firm and lead to a lower stock price. However, this method does not
answer the second part of the question; to do that we must graph the supply
of and demand for a firm’s stock in the secondary market before and after
the announcement of lower-than-expected growth in earnings. This approach
will allow us to see the impact of the announcement on both the price and
volume of shares traded.
Conclusion 265
The figure illustrates the initial supply (S0) and demand (Z)°) for Beta’s stock.
The initial equilibrium is at point A, which means the equilibrium price of
a share of Beta’s stock is P° and the corresponding quantity of shares traded
on a typical day is Q°. When the lower-than-expected earnings growth is
announced, the demand curve shifts to Dl. This occurs because buyers of
the stock will now pay less for any given quantity of the stock due to the
lower growth forecasts. Simultaneously, the supply of Beta’s stock shifts to
S1 as owners of Beta stock attempt to get rid of their shares on the secondary
market thanks to Beta’s lower projected earnings. Since demand decreases
and supply increases, the equilibrium price of the firm’s stock clearly falls
to a level such as Pl. However, the volume of shares traded may rise, fall,
or remain the same depending on the relative magnitude of the shifts. This
is the same ambiguity that typically arises when both demand and supply
shift. The new equilibrium shown at point B assumes the increase in supply
exactly offsets the decrease in demand, but this need not be the case.
Conclusion
In this chapter, we used present value analysis to determine the prices of
financial assets and found an inverse relationship between asset prices and
the interest rate. This analysis not only provides a means of pricing financial
assets like stocks and bonds but, coupled with the supply and demand
analysis we learned in Chapter 4, also shows how these prices fluctuate with
changes in inflationary expectations, budget deficits, tax rates, wealth, and
other variables that shift the demand for or supply of loanable funds. In the
next chapter, we extend this analysis by examining how risk affects asset
prices.
KEY TERMS
present value current yield
future value equity instrument
debt instrument income stock
discount yield dividend
face value growth stock
yield to maturity dividend yield
coupon rate

1. Suppose you plan to purchase a new 6. Carefully explain what would happen to
house at the end of five years. How much the price and quantity of Treasury bills
would you have to deposit in the bank to¬ traded in the money market if inflationary
day to have enough for the $20,000 down expectations increased.
payment if the interest rate is (a) 1 per¬
cent, (b) 5 percent, (c) 10 percent, (d) 15 7. You just purchased a coupon bond that
percent, and (e) 20 percent? has a face value of $1,000 and a coupon
rate of 8 percent. The bond matures in one
2. Based on your results in problem 1, graph
year.
the relationship between the interest rate
(a) . If the interest rate was 6 percent
and the present value of $20,000 received
when you purchased the bond, how much
in five years.
did you have to pay for this bond? What
3. A recent advertisement offers a two-year was the current yield? The yield to ma¬
subscription to a magazine for $20. If a turity?
one-year subscription costs $11 and you (b) . Suppose that as soon as you pur¬
can invest your money at an interest rate chased the bond at the price determined in
of 9 percent, would you save money by part a, the interest rate soared to 10 per¬
sending in a check today for a two-year cent. If you immediately sold the bond,
subscription? Explain. what would you receive for it? As a
holder of the bond, are you pleased that
4. Suppose you paid $9,000 for a Treasury
the interest rate increased to 10 percent?
bill that matures in one year. Determine
Explain.
(a) the discount yield and (b) the yield to
maturity.
8. Suppose the government enacted a policy
5. If the market interest rate is 8.5 percent, that prevented corporations from deduct¬
how much would you be willing to pay ing interest payments on coupon bonds
for a Treasury bill that matures in one from their income taxes. How would this
year? Given this price, determine the dis¬ affect prices of corporate coupon bonds
count yield and the yield to maturity, and traded in (a) the primary bond market and
explain why they differ. (b) the secondary bond market?
9. Suppose a one-year Treasury bill has a (a) . 1 percent?

discount yield of 5 percent. If your alter¬ (b) . 2 percent?
native to purchasing this bond is to de¬ (c) . 3 percent?
posit the funds in a bank account that (d) . Based on your results in parts a-c,
pays 5 percent interest, what will you do? graph the relationship between the growth
Why? rate and the price of Young-R-Us’s stock
price.
10. A utility company is expected to have
earnings of $5 million each year from 13. Young-R-Us Company (problem 12) pres¬
now to eternity. The company pays out all ently pays no dividends, even though it
of its annual earnings as dividends to had earnings of $2 per share last year. If
owners of its 2 million shares of stock. If the growth rate in earnings is expected to
the interest rate is 6 percent, be 4 percent per year forever, what is the
(a) . What is a reasonable price to pay for price of the firm’s stock if the interest rate
the stock? is
(b) . Based on the price in part a, what is
(a) . 5 percent?
the stock’s dividend yield? (b) . 6 percent?
(c) . Based on the price in part a, what is (c) . 7 percent?

(d) . Based on your results in parts a-c,
the ratio of the stock’s price to earnings
per share (called the P/E ratio on Wall graph the relationship between the interest
Street)? rate and the price of Young-R-Us’s stock
price.
11. Why would anyone purchase a stock that 14. Suppose a firm announces a two-for-one
does not pay dividends? stock split; that is, for every share of stock
owned by investors, the firm gives them
12. Young-R-Us Company presently does
another share for free. What impact does
not pay out dividends, even though it
this practice have on the price of the
had earnings of $2 per share last year.
stock?
If the interest rate is 4 percent, what is
the price of the firm’s stock if earnings 15. Who receives funds when a stock or a
are expected to grow forever at an bond is purchased in the primary market?
annual rate of In the secondary market?
Selections for Further reading

Bodurtha, J. N., Jr., and N. C. Mark. “Testing the Cecchetti, S. G., P. Lam, and N. C. Mark. “Mean Re¬
CAPM with Time-Varying Risks and Returns.” version in Equilibrium Asset Prices.” American
Journal of Finance, 46 (September 1991), 1485- Economic Review, 80 (June 1990), 398-418.
1505. Craig, E. D. “Impact of Federal Policies on Munici¬
Cecchetti, S. G. “Evaluating Empirical Tests of Asset pal Bond Financing.” National Tax Journal, 34
Pricing Models: Alternative Interpretations.” (September 1981), 389-394.
American Economic Review, 80 (May 1990), 48-
51. Continued on p. 268
Fabozzi, F. J., and T. B. Thurston. “State Taxes and Week Effect.” Economic Letters, 26 (1988), 353—
Reserve Requirements as Major Determinants of 356.
Yield Spreads among Money Market Instru¬ McCulloch, J. H. “An Estimate of the Liquidity Pre¬
ments.” Journal of Financial and Quantitative mium.” Journal of Political Economy, 83 (Feb¬
Analysis, 21 (December 1986), 427-436. ruary 1975), 95-119.
Falk, B. “Formally Testing the Present Value Model Schilling, D. “Estimating the Present Value of Future
of Farmland Prices.” American Journal of Agri¬ Income Losses: An Historical Simulation, 1900-
cultural Economics, 73 (February 1991), 1-10. 1982.” Journal of Risk and Insurance, 52 (March
James, C. M., and R. O. Edmister. “The Relation be¬ 1985), 100-116.
tween Common Stock Returns Trading Activity Spatt, C. S., and F. P. Sterbenz. “Warrant Exercise,
and Market Value.” Journal of Finance, 38 (Sep¬ Dividends, and Reinvestment Policy.” Journal of
tember 1983), 1075-1086. Finance, 43 (June 1988), 493-506.
Lee, T. K. “Settlement System and the Day-of-the-
CHAPTER
Risk; Uncertainty,
and Portfolio Choice
II
n the previous chapter, we saw that interest rates affect the prices of
financial assets that banks and individuals hold, such as stocks and bonds.
We assumed there was a single interest rate and borrowers fully kept their
promises to repay loans.
In reality, a host of factors affect the prices of financial assets, chief
among them the various forms of risk. For example, default risk is the risk
that a borrower will not fully repay the principal and interest on a loan.
Market risk, the risk that the price of a security will fall between the time
the security is purchased and the time the owner wants to sell it, can also
affect an asset’s price.
Different types of financial instruments by their nature entail different
levels of risk. Moreover, financial instruments issued by various borrowers
represent different degrees of risk due to variations in borrowers’ ultimate
ability to repay. Thus, there is generally no single interest rate in the market
for loanable funds; instead, interest rates vary for different borrowers and
for different types of financial instruments. A key reason for differential
interest rates is different degrees and types of risk among various financial
instruments.
Lending money always involves some form and some degree of risk. In
this chapter, we see how different types of risk affect the prices banks and
investors are willing to pay for financial assets and how those prices in turn
affect the interest rates borrowers must pay to obtain funds. First, we define
risk and describe the various types of risk that affect financial instruments.
Then we see how risk affects asset prices and interest rates. Finally, we
examine some methods banks and individuals use to reduce financial risk.
Uncertainty and Risk

Both uncertainty and risk are widely used words, and most of us understand
their commonly used definitions. For instance, The American Heritage
Dictionary of the English Language defines uncertainty as “the condition
of being uncertain, doubt” and uncertain as “not known or established; not
269
270 Chapter 9 Risk, Uncertainty, and Portfolio Choice
determined, undecided; not having sure knowledge; subject to change, var¬

iable.” It defines risk as the “possibility of suffering harm or loss; danger,”
but also as “the danger of probability of loss to an insurer; the variability
of returns from an investment; the chance of nonpayment of a debt.”1
In this chapter, we introduce the specialized definitions of these terms
as economists use them. We begin with risk. Like the dictionary definition,
economists think of risk as a possibility of loss. The term risk applies to
outcomes that cannot be exactly predicted. However, economists distinguish
between situations in which we cannot exactly predict outcomes, but know
which outcomes might occur and are able to assign probabilities to those
outcomes and situations in which we lack such information. The former
situation is a situation of risk and the latter a situation of uncertainty.
As a simple example, consider a coin toss in which you bet $50 that
heads will come up instead of tails. Before you toss the coin, you cannot
exactly predict the outcome. However, you can list all possible outcomes—
heads and you win $50 or tails and you lose $50—and the probability of
each outcome occurring. In this example, there is a 50-50 chance of each
outcome, or a probability of 1/2. These are objective probabilities—the true
chance of the event occurring—and this is a situation of risk.
However, more complicated situations are still considered situations of
risk. For instance, your grade in this course is an outcome that you cannot
predict exactly. However, you can list the possible outcomes—A, B, C, D,
or F—and you have an idea of the probabilities of each. In this case, the
probabilities are perhaps a bit more difficult to determine and clearly depend
on your willingness to expend resources (study time, money for books or
tutors, etc.) to learn the material in this course. However, even here you can
estimate the probabilities of the various outcomes. These probabilities are
subjective, since they are formed by you and may not be the same as the
true or objective probabilities of the various outcomes. (You may be more
pessimistic or more optimistic than warranted.) However, as long as you
think these probabilities are correct, you will act as though they are objective.
What distinguishes risk from uncertainty? Economists usually reserve
the word uncertainty to describe a situation in which individuals cannot
form probabilities for outcomes, either because they cannot identify the
possible outcomes or because they cannot estimate the probabilities to assign
to the outcomes. Such a situation occurs when we have no historical evidence
and hence no ability to draw on past experience to help us assign probabilities
to outcomes. Because of this, economists rarely analyze such situations, and
examples seem rather far-fetched. Imagine trying to figure out the probability
of global thermonuclear war. It is difficult to estimate this probability since,
fortunately, no historical precedents exist. You might consider the probabil¬
ity of occurrence of world wars in general, but the very existence of nuclear
1 Excerpted from The American Heritage Dictionary of the English Language, 3rd ed. (Boston:
Houghton Mifflin, 1992).
weapons may have lessened the importance of these historical probabilities

for predicting future wars, especially nuclear wars. In any case, we concern
ourselves in this chapter with situations of risk, in which probabilities—at
least subjective probabilities—can be assigned to the various outcomes. We
will say more about the difference between subjective and objective proba¬
bilities in Chapter 12, when we introduce the concept of rational expec¬
tations.
In the following sections, we deal with situations of risk. In these situ¬
ations, we can list the possible outcomes and assign a probability of occur¬
rence to each. This allows us to formalize two concepts: risk and expected
return. In financial markets, the outcomes might be various possible rates of
return on a financial instrument and the expected return an appropriately
defined average of these possible rates of return, an average that is weighted
by the probability that each rate of return will actually occur. In this appli¬
cation, the concept of risk is the chance that the rate of return will differ
from the expected or average return. The higher the probability that the rate
of return that occurs will be far from the average return, the higher the risk.
We also show that these concepts are simply special cases of two statistical
concepts, expected value and variance, that we discuss next.
Expected Value
Imagine a coin toss in which you will receive $1 if the coin turns up heads
but must pay $1 if it turns up tails. Each of the two possible outcomes is
equally likely, which means the probability of each occurring is 1/2. The
expected value (or mean) of this risky prospect is the sum of the outcomes
weighted by their probabilities. Thus, for our example the expected value is
1 1
Expected value = -($1) H—(— $ 1) = 0.
^ 2
In other words, there is a probability of .5 (or 50 percent) that you will make
$1 and a probability of .5 (or 50 percent) that you will lose $1. On average,
you will neither make nor lose money by undertaking this risky prospect.
Even though the expected value of this prospect is zero, you will never earn
zero by making the bet; you will either make or lose $ 1.
More generally, suppose you own a risky asset. The value of the asset
(x) is not certain. There is a probability px that its value is xx, a probability
p2 that its value is x2, and so on up to a probability pn that the value is xn.
We use pi to denote the probabilities assigned to different values of the asset.
These probabilities must sum up to 1 if we have listed all possible outcomes,
since one of them must occur. Moreover, we must list all possible outcomes
to be in a situation of risk as opposed to one of uncertainty. Using summation
notation, we write this condition as
n
2
i= 1
Pi = P\ + Pl + ••• + Pn = E
Remember, xt denotes all the possible values of the asset and pt denotes the
corresponding probabilities, so the expected value (or mean) of this risky
asset, denoted Ex, is the sum of the possible outcomes weighted by their
probabilities of occurrence:
n
EX =/?!*! + p2X2 + • • * + pnXn = 2 PiXr

1
i=
Notice that the expected value of a risky asset collapses information

about the possible value of the asset into a single number. This number
measures the average value that would occur if a risky situation were played
out many times (technically an infinite number of times).
Variance
When dealing with risky situations, we also want a measure of the risk: the
possibility of values other than the expected value. For this we turn to the
variance. In our coin-tossing example, the expected value of the risky pros¬
pect is zero; on average, you will neither make nor lose money by placing
the bet. However, you will never earn exactly zero by betting $1 on the
outcome of a coin flip; you will either make or lose $1. The only way you
can earn zero for sure is by not placing the bet. Of course, the expected
value of betting zero (i.e., not betting) is zero, since regardless of whether
the coin comes up heads or tails, you earn zero:
1 1
Expected value of not betting = - ($0) H— ($0) = 0.
Thus, two prospects can have identical means but differ with respect to their
underlying risks. Betting zero or betting $1 on a coin toss both have an
expected value of zero, yet betting on the outcome of a coin toss is more
risky than not betting.
A common measure of the risk of a risky prospect is its variance (a2),
which is a weighted sum of the squared deviations of potential outcomes,
xh from the expected value, Ex. The weights used are the probabilities of
each outcome occurring—the pt mentioned earlier. Thus, the variance of a
risky prospect is defined as
a2 = p\{xx — Ex)2 + p2(x 2 — Ex)2 + • • • + pn(xn — Ex)2

n
= 2
/= 1
iPifri ~ £*)2].
For example, when you bet $1 that a coin will turn up heads, px = 1/2,
xx = 1, Ex — 0, p2 = 1/2, and x2 = — 1. The variance in your profit is
a2 = ^(1 - 0)2 + t(-i - 0)2 = 1.

In contrast, if you bet nothing, px = 1/2, xx = 0, Ex = 0, p2 = 1/2, and

jc2 = 0. Now the variance is
a2 = t (0 — 0)2 + t (0 — 0)2 = 0.
The fact that the variance of betting $ 1 is greater than the variance of betting
$0 means that betting $1 is more risky than betting nothing.
Finally, some economists use a closely related measure of risk called
the standard deviation. The standard deviation is simply the square root of
the variance. Increases or decreases in the variance will also be increases or
decreases in the standard deviation. Thus, when risk is changing, both meas¬
ures tell us that risk is moving in the same direction. Hence we need look
at only one of these measures; in this text, we will concentrate on the
variance.
Risk Preference
Even if two bets had the same expected value and variance, not everyone
would be indifferent when choosing between the two bets. Different people
have different attitudes toward risk. Suppose there are two prospects, a risky
one and a sure thing. Each prospect has the same expected value, but the
riskier prospect has a higher variance. On average, you will earn the same
amount from each. To be concrete, suppose you are offered one of two jobs
in a company. Job 1 pays a salary of $50,000 every year regardless of the
company’s profits that year. Job 2 pays a base salary of $25,000 every year,
plus a bonus that depends on the company’s profits that year. Based on past
history, these bonuses are $0 with probability 1/4, $25,000 with probability
1/2, and $50,000 with probability 1/4. What is the expected value of job 1?
Clearly it is $50,000 per year. What about job 2? The expected value is
$25,000 + (1/4) X $0 + (1/2) X $25,000 + (1/4) X $50,000, or $50,000.
These jobs have the same expected value, but job 2 is riskier than job 1
because it has a higher variance.
Some people will choose job 1 over job 2. These people are risk averse:
They prefer a sure thing to a risky alternative with an identical expected
value. Others will prefer job 2 to job 1. These people are risk loving: They
prefer the risky prospect to the sure thing, even though both have identical
expected values. Finally, some people will be indifferent between the two
jobs. They are risk neutral: They differentiate between risky prospects only
on the basis of expected value. A risk-neutral investor sees assets with
identical expected values as equivalent, even if their variances differ.
Risk-averse individuals are willing to pay money to avoid a risky pros¬
pect. For instance, consider a person who owns a $100,000 house. This is a
risky prospect, for if the house burns down, it will be worth nothing. Most
people are willing to pay several hundred dollars in fire insurance each year
to avoid the risk of loss from fire. In effect, by purchasing fire insurance
they are buying a sure thing: a house that is worth $100,000 regardless of
whether or not it burns down. Of course, most mortgage companies require
borrowers to carry fire insurance. But consider earthquake insurance. Mort¬
gage companies often don’t require earthquake insurance, and in the Los
Angeles earthquake of 1994, only about 40 percent of homeowners had this
coverage. Apparently there are quite a few risk-neutral or risk-loving home-
owners in L.A.
Summary You are considering two investments. Investment A (xA) is certain to yield
Exercise 9.1 profits of $100. Investment B (v5) will yield profits of $910 with probability
.1 and $10 with probability .9. (a) What are the expected value and variance
of each investment? (b) Which investment will you prefer if you are risk
averse? (c) Which investment will you prefer if you are risk neutral?
Answer: (a) Since investment A is certain, its expected value is
Exa = $100,
while its variance is (j2a = 0. For investment B,
Exb = . 1 ($910) + .9($ 10) = $100
and
(j2 = 1(910 _ j 00)2 + 9(10 _ 100)2 = 72,900.
Thus, investment A has the same mean but a lower variance, (b) If you are
risk averse, you will prefer a sure thing to an uncertain prospect with an
identical expected value. Thus, you will prefer investment A. (c) If you are
risk neutral, you will be indifferent between investment A and invest¬
ment B.
Valuing Risky Financial Assets
Now that you have a basic understanding of risk and the methods used to
evaluate risky prospects, we examine the implications of risk for financial
markets. In this section, we look at how different types of risk affect the
pricing of financial instruments as well as interest rates, prices, and the yields
of financial instruments such as bank loans, Treasury bills, corporate bonds,
and municipal bonds.
Individuals will insist on paying less for a risky asset than for a risk¬
free asset for two reasons. First, a risk-free asset generally has a higher
expected repayment than a risky asset. For instance, consider two bonds,
one of which is certain to pay $1,000 to the holder at maturity while the
other will pay $1,000 only if the issuing firm does not go bankrupt. If the
probability of bankruptcy is, say, 10 percent, the expected value of the
payment at maturity of the second bond is .9 X $1,000 + .1 x 0 = $900.

The lower expected payment of the risky bond leads to a lower price (and a
higher yield) than that of the risk-free bond.
Second, even if two financial assets have identical expected values, risk-
averse individuals will insist on paying less for the riskier asset (the one
with a higher variance) than for the less risky one. Since a lower price
implies a higher interest rate (as we learned in Chapter 3, a risk-averse
individual will demand a higher interest rate on riskier assets even if the
expected value is identical to those on less risky assets). The difference
between the interest rate on a risky asset and that on a risk-free asset is
known as the risk premium.
Default Risk
Default risk is the risk that a borrower will fail to make the stipulated debt
payments. For instance, suppose you purchase a 20-year corporate bond that
has a face value of $1,000 and a coupon rate of 7 percent. The issuing
corporation has promised to pay $70 each year to the bond owners and an
additional $1,000 at maturity. Of course, many things can happen in 20
years; the corporation might suffer losses and be unable to make one or
more of its interest payments or, even worse, be unable to pay the $1,000
face value at maturity. When the issuer of a debt instrument fails to make
required payments when due, the instrument is in default.
Default risk is related to the creditworthiness of the issuer of a particular
debt instrument. When the U.S. Treasury issues a Treasury bill or Treasury
bond, default risk is very low; to obtain funds to meet its obligation, the
federal government can collect additional tax revenue or print money.2 For
this reason, Treasury bills, Treasury bonds, and other debt instruments that
are backed by the full faith and credit of the U.S. government are often
referred to as default-free debt instruments. However, the growing federal
debt may add some degree of default risk even to U.S. government-backed
debt instruments, although that risk is minimal.
During the 1980s Mexico, Brazil, and other Latin American countries
had difficulty meeting payment obligations on their government-issued debt.
One reason for these defaults was that the debt was issued in a foreign
currency, usually U.S. dollars. Since these countries were obligated to repay
in U.S. dollars, they could not meet their obligation by simply printing more
money. In contrast, U.S. government debt is denominated in U.S. currency,
so the United States has the ability to repay its debt by printing money.
Doing so would cause tremendous inflation, however, and would in effect
be a de facto default, though not a default recognized as such by the courts.
We discuss inflation risk later in this chapter. For now we simply note that
2 Resorting to printing money to pay interest on the debt can lead to inflation, another risk that we
consider later in this chapter.
the federal government’s temptation to print fiat money to repay its debts is
in essence a temptation to default on that debt.
Corporate bonds carry a higher default risk than U.S. government bonds.
Since a firm’s profits can vary significantly from year to year, the firm may
lack enough funds to make required debt payments when they fall due.
Obviously firms cannot print money or raise taxes to cover their deficiencies,
which makes corporate bonds riskier than U.S. government bonds.
Firms in a strong financial position will have lower default risk than
firms in weaker positions. Moody’s and Standard and Poor’s provide inves¬
tors with ratings of the bonds issued by major corporations. Box 9.1 sum¬
marizes Moody’s rating system. As we will see later in this chapter, low-
quality bonds (junk bonds) have lower prices and higher yields than
higher-quality bonds.
Bank loans also carry default risk, which varies depending on the type
of loan. The typical individual who obtains a short-term consumer loan at a
bank is more likely to default on the loan than, say, AT&T. To compensate
the bank for the higher risk of default, this individual pays a higher interest
rate than the bank charges creditworthy corporations (called the prime rate).
Figure 9.1 illustrates this fact using the loanable funds model developed
in Chapter 4. Part a shows the supply of and demand for consumer loans,
and the equilibrium interest rate on consumer loans is 10 percent. Part b
shows the supply of and demand for loans to the bank’s more creditworthy
corporations; the equilibrium interest rate—the prime interest rate—is 7
percent.
Default Risk Premium. The difference between the 10 percent rate

charged on consumer loans and the 7 percent prime rate charged to more
creditworthy customers reflects a default risk premium, a premium a bor¬
rower must pay to compensate a lender for the risk of default. Here the
default risk premium is the amount consumers pay to banks as compensation
for the higher default risk on consumer loans. As we will see, however, even
the most creditworthy customers pay a premium over and above the rate at
which the U.S. government can obtain loanable funds.
The default risk premium applies not only to banks but also to bonds
and other assets that are traded on major exchanges at a dollar price. For
simplicity, we consider bonds that are identical in every respect except their
default risk. Each bond is a one-year bond with a $1,000 face value and a
coupon rate of 5 percent. Thus, each bond promises to pay the holder $1,050
at the end of one year (a $50 interest or coupon payment, plus $1,000 face
value). We also assume individuals are risk neutral; at the end of this section,
we examine the implications of risk aversion for the pricing of risky assets.
We begin by considering a default-free bond issued by the government.
If the default-free interest rate (the interest rate charged to those who are
known to never default on a debt obligation) is / = 5 percent, the price of
Valuing Risky Financial /Assets 277
What Bond Ratings Mean
There are two main sources of bond ratings: panying table shows Moody's ratings and their
Moody's and Standard and Poor's. The accom¬ meanings.
Rating
(Moody's) Interpretation
Aaa These bonds have the lowest risk of default.
Aa These bonds carry a slightly higher long-term investment risk.
A These bonds are still good investments, but they have a higher default risk.
Baa These bonds are reliable at least in the short run, but are questionable over
the long run.
Ba These bonds have a relatively high degree of risk.
B These bonds are considered undesirable investments due to their high degree
of risk.
Caa The issuers of these bonds are already or may soon be in default.
Ca These bonds are often already in default; investment in such bonds would be
only for speculative purposes.
C This is the lowest rating given to bonds. The probability that these bonds will
ever be considered a good investment is very low.
the government bond is simply the present value of $1,050, as we discussed

in Chapter 8. Thus,
$1,050
Pg ~ 1.05 ’
or $1,000. The price of the government bond equals the face value because
the coupon rate is the same as the default-free rate of interest.
Now consider a bond issued by a corporation that has a bond rating of
B. Given this rating, there is only, say, a 90 percent chance that the firm will
pay its obligations in full, in which case the bondholder will receive $1,050
at the end of one year. There is a 10 percent chance that the firm will be
unable to meet its debt obligations. This could mean the firm can only pay,
say, 50 cents on the dollar, but we will suppose it pays nothing to the
bondholder in the event of default. Thus, the expected value of the payment
received at the end of the year is
Expected value of payment = .9 ($1,050) + .1 ($0) = $945.
Even though the bond states that $1,050 will be paid at the end of the year,
the 10 percent chance of default lowers the expected value of the payment
to $945.
Now suppose an investor considering the purchase of this bond wants
to know how much she can pay for it and still earn the 5 percent interest
rate she could earn on a default-free investment. The price this risk-neutral
investor will be willing to pay for this B-rated bond is the present value of
the expected payment:
_ $945
Pb ~ To? ’
or $900. Notice the similarity between default-free bonds and risky bonds.
However, the default risk on the B-rated corporate bond means the bond
sells at a lower price than the default-free government bond does, even
though both bonds promise to pay $1,050 at the end of one year. The higher
price investors are willing to pay for the government bond reflects the fact
that the government will almost certainly keep its promise to repay.
The Inverse Relationship Between Bond Prices and Interest

Rates Revisited. The fact that these two bonds have different prices
also implies that the interest rate for the default-free government bond will
be lower than that for the risky B-rated corporate bond. To see this, consider
Figure 9.2, which graphs the inverse relationship between bond prices and
the interest rate. The interest rate associated with a bond price of PG =
$1,000 is 5 percent; the interest rate associated with a bond price of PB =
$900 is 16.7 percent. The difference between the interest rate on the B-rated
corporate bond and that on the otherwise identical government bond is the
default risk premium. In this case, the default risk premium is 11.7 percent.
If the B-rated corporate bond does indeed meet its obligations, you will earn
a return that is 11.7 percent higher than that on the government bond. Of
course, since there is a 10 percent chance that the company will default, you
require this premium to compensate for this risk.
Another way to look at the default risk premium associated with the B-
rated corporate bond is to calculate the yield to maturity of each bond. These
calculations assume both bonds do in fact meet their obligations. Recall from
Chapter 8 that the yield to maturity on the government bond is the interest
rate, iG, that satisfies the condition
$1,050
$1,000
1 + Ig
Solving this expression for iG reveals that the yield to maturity of the gov¬
ernment bond is iG — 5 percent—the default-free interest rate. In contrast,
the yield to maturity on the B-rated corporate bond (assuming the company
does not default) is
$1,050
$900 =
1 + Ib
Solving this expression for iB reveals that the yield to maturity on the
corporate bond is iB = 16.7 percent.
In equilibrium, the yield to maturity on a bond will equal the market
interest rate for loans issued to borrowers with similar credit risks. Figure
9.3 illustrates the market for default-free loans and for loans to companies
with a credit rating of B (very risky loans). The market equilibrium interest
rate for default-free government bonds is 5 percent. The market interest rate
on loans issued to borrowers with a 10 percent chance of defaulting is 16.7
percent. The difference in the rates for default-free and risky loans reflects
the default risk premium. The main reason for the default risk premium is
that the supply of loanable funds shifts to the left on riskier loans. At every
interest rate, the quantity of loanable funds supplied is lower on riskier loans.
The default risk premium will be lower on loans issued to borrowers
less likely to default. Consider a very creditworthy corporation that has
issued a Aaa-rated bond. This bond, like the earlier B-rated bond and gov¬
ernment bond, has a face value of $1,000, matures in one year, and has a
coupon rate of 5 percent. However, this firm has only a 1 percent chance of
defaulting on the loan. Given this probability of default, the expected pay¬
ment to the bondholder at the end of the year is
Expected paymentAaa = .99 ($1,050) + .01 ($0) = $1,039.50.
The present value of this expected payment given the 5 percent default-free
interest rate is the price a risk-neutral investor would be willing to pay for
the bond:
$1,039.50
PAaa ~ 1.05 ’
or $990. A risk-neutral investor would pay more for this Aaa bond than for
the B-rated bond, but slightly less than for a default-free government bond.
Since the yield to maturity on a bond will equal the market rate of interest on loans issued
to borrowers with similar credit risks, the default risk premium can be illustrated by the
difference between the interest rate for risky loans and the interest rate for default-free
loans. In this figure, the demand for risky and default-free loans is assumed to be the same
for simplicity. The default-free interest rate is 5 percent, while the interest rate on B-rated
bonds is 16.7 percent. The default risk premium—the difference between these two inter¬
est rates—is 11.7 percent and is the additional interest necessary to induce lenders to issue
loans to risky borrowers.
Loanable Funds
Given this price, the yield to maturity if the Aaa bond does not default is
$1,050
$990 =
1 + lAaa
Solving for iAaa gives us the yield to maturity of this Aaa bond: iAaa = .06,
or 6 percent. Thus, the less risky Aaa bond has a default risk premium of
only 1 percent over the default-free government bond—a considerably lower
default risk premium than the B-rated bond has. This 1 percent default risk
premium is required to compensate the investor for the (small) chance that
the company will default. Box 9.2 shows the interest rates on Aaa and Baa
corporate bonds and on one-year Treasury bills.
Until now, our analysis of valuing risky assets assumed investors are
risk neutral and care only about the expected value of a risky asset. Since a
risk-averse investor prefers a sure thing to a risky prospect with an identical
expected value, it follows that risk-averse investors will pay less for a risky
bond than risk-neutral investors will. The lower price risk-averse investors
will pay is equivalent to saying that risk-averse investors demand an even
higher interest rate on risky assets. In other words, the default premiums in
markets dominated by risk-averse investors will be even higher than the risk
premiums required by risk-neutral investors.
One class of particularly risky bonds is the so-called junk bonds. Box
9.3 explains junk bonds and convertible bonds.
Liquidity Risk
As we saw in Chapter 2, debt instruments differ with respect to their under¬
lying liquidity—the ease with which the owner can sell the debt instrument
without loss to obtain money. Some bonds, such as federal government
bonds, are widely held and easily sold in well-established secondary markets;
thus, these bonds are very liquid. Other bonds, such as those issued by small
companies and municipal governments, are not widely held and are less
widely traded in secondary markets. We say the market for these bonds is
“thin.” When the secondary market for a bond is very thin, a person who
wishes to sell a bond for cash could have a very difficult time. He or she
may be able to do so only at a very low price or, in some cases, at no price
at all.
Buying debt instruments that are not very liquid involves considerable
liquidity risk. Liquidity risk is risk concerning the price an investor will
receive if a bond or other debt instrument must be liquidated prior to ma¬
turity. As we will see in the next subsection, liquidity risk is closely related
to interest rate risk, which affects the price at which a bond can be sold if
liquidated prior to maturity.
While virtually all debt instruments carry some degree of liquidity risk,
some are riskier than others. In general, liquidity risk is lower on debt
instruments whose issuers are widely known to be creditworthy. For instance,
Interest Rates on Aaa, Baa, and

U.S. Treasury Bills, 1971-1992
The accompanying figure graphs the interest short-term prospects are adequate but whose
rates on one-year U.S. Treasury bills and two long-term prospects are suspect). We would ex¬
classes of corporate bonds, Aaa bonds and Baa pect the Baa rate to exceed the Aaa rate to
bonds. Moody's assigns a rating of Aaa to compensate investors for the additional risk of
bonds of the highest guality (those with the default on Baa bonds, and indeed this is the
lowest degree of default risk); It assigns a rating case. The government bond rate is generally be¬
of Baa to medium-grade bonds (bonds whose low even the Aaa bond rate.
Source: Citibase electronic database.
U.S. government bonds tend to have relatively little liquidity risk, since
millions of investors are familiar with their terms and the creditworthiness
of the U.S. government. In contrast, bonds issued by a small town or a small
company are not very liquid. Very few investors are familiar with the issuers’
ability to repay the debt, and it is difficult to determine the specific terms on
M
What Are Junk Bonds and
Convertible Bonds?
Junk bonds and convertible bonds are two instrumental in the takeovers and mergers of
very different things. A junk bond is a corporate that decade.
bond that has a very low rating (or perhaps no A convertible bond gives the bondholder
rating at all). This, of course, implies that the the option to have the loan repaid with stock
market associates a very high element of risk instead of cash. The date on which the bond¬
with these bonds. In exchange for the high holder can exercise this option and the number
degree of risk, a junk bond will offer a high of shares of stock for which each bond can be
interest rate. Junk bonds became popular in exchanged are specified when the bond is
the 1980s, when corporations issued these issued. Convertible bonds are not necessarily
bonds and used the proceeds to purchase junk bonds, nor are junk bonds always con¬
other companies. Thus, junk bonds were vertible bonds.
the bonds. Thus, it is more difficult for buyers and sellers to get together
and agree on a price.
The more liquid an asset, the more attractive it is to buyers. Consider
part a of Figure 9.4, which illustrates the supply of and demand for loanable
funds to large U.S. corporations. Since corporate bonds are relatively liquid,
the equilibrium interest rate is 6 percent. Part b of Figure 9.4 illustrates the
supply of and demand for loanable funds to small corporations. Since bonds
issued by small corporations are less liquid than those issued by larger ones,
the equilibrium interest rate must be higher, even ignoring differences in
default risk. In this case, the interest rate on the less liquid bonds is 8 percent.
The difference in these rates reflects the liquidity premium—the additional
interest rate lenders must receive as compensation for the liquidity risk.
Since a decrease in liquidity leads to a reduction in the supply of loanable
funds, corporations whose bonds are not listed (or traded) on the secondary
market for bonds (such as the New York Bond Exchange) can be adversely
affected. To see this, consider part a of Figure 9.5, which illustrates the
demand for and supply of loanable funds to a firm. When the firm’s bonds
are listed on a major exchange, equilibrium is at point A, and the interest
rate the firm must pay for loanable funds is 8 percent. When the firm’s bonds
are not traded in the secondary bond market, their liquidity decreases, and
lenders are less willing to lend to this firm. This decrease in liquidity shifts
the supply of loanable funds to the left. The new equilibrium is at point B
Figure 9.4
Liquidity Premium
The more liquid an asset, the lower the interest rate needed to compensate lenders for
their funds. In this figure, loans to large corporations (part a) are more liquid than loans to
corporations (part b). Consequently, the equilibrium interest rate on loans to large corpora¬
tions is 2 percent lower than that to small businesses because small-business loans are less
liquid than corporate loans. This 2 percent liquidity premium is the additional interest rate
lenders must receive as compensation for holding the less liquid asset.
Interest Interest
Rate (%) Rate (%)
(a) Loans to Large Corporations (b) Loans to Small Corporations
in Figure 9.5, where the interest rate is 12 percent. When the firm seeks
additional funds, it will have to pay a higher interest rate.
Part b of Figure 9.5 illustrates the impact of the decrease in liquidity on
bond prices (and thus on owners of existing bonds). When the firm is listed
on the bond exchange, the interest rate of 8 percent translates into a bond
price of $1,000. When the firm is not listed, the rise in the interest rate to
12 percent leads to a reduction in the price of an existing bond to $900. The
value of an investor’s loan portfolio will fall due to the decrease in liquidity
of the bonds.
Interest Rate Risk and Call Risk

Variability in interest rates leads bond prices to fluctuate, creating risk to
the value of an investor’s bond holdings. This interest rate risk—the risk
of a change in the overall level of interest rates—leads to greater liquidity
risk. Interest rate risk also imposes an additional source of risk on bonds
Figure 9.5
Prices and Yields for Bonds Listed and Bonds
Not Listed on the Secondary Bond Market
A decrease in the liquidity of an asset will lower that asset's price and raise the interest
rate necessary to compensate lenders who hold the asset. Such a situation occurs when a
corporation's bonds are not listed on (not traded in) the secondary bond market. Part a
shows the loanable funds market for such a corporation, and part b indicates the relation¬
ship between the bond's interest rate and the price. The original interest rate for the listed
bond is 8 percent at point A in part a. This corresponds to a price for the iisted bond of
$ 1,000 in part b. When the bond is not listed, it thus becomes less liquid, the supply of
loanable funds decreases, and equilibrium moves to point B in part a. This leads to a higher
interest rate of 12 percent and a lower bond price of $900.
Interest Interest
Rate (%) Rate (%)
with call features. In particular, callable bonds are subject to the risk that
the issuer will call the bond, that is, pay off the bond’s face value prior to
maturity. This happens when a bond is issued with a high coupon rate and
the overall level of interest rates then declines, making it profitable for the
bond issuer to call in the old bonds and reissue them with a lower coupon
rate. In this case, the owner receives the face value earlier than anticipated,
but will not receive the same amount in coupon payments that he or she
would have received had the bond not been called.
If interest rates did not fluctuate, this call risk would not be a problem
for investors. Regardless of when the bonds were paid off, the proceeds
could be used to purchase bonds with identical coupon payments from some
other issuer. Unfortunately, if the interest rate fell after the callable bond
Valuing Risky Financial /Assets 287
was purchased, it would not be possible to obtain a new bond that yielded
as much as the old bonds. Let us see why this is the case.
Suppose you purchase a one-year bond with a coupon rate of 8 percent
and a face value of $1,000. For simplicity, assume there is no default risk
and this bond is not callable. If the market interest rate is 8 percent, the price
of this bond is
$80 + $1,000
$1,000.
1 + .08
Since the market interest rate of 8 percent equals the coupon rate, the yield
to maturity of this bond is also 8 percent.
Now suppose that immediately after you purchase the bond, the interest
rate drops to 4 percent. Since this bond is not callable, the interest rate
change does not affect your 8 percent coupon payment; you will still receive
an 8 percent yield to maturity on your initial investment despite the fall of
the market rate to 4 percent. Notice, however, that investors who want to
purchase the bond from you after the decline in the interest rate will be
willing to pay
$80 + $1,000
$1,038.46. (9.1)
1 + .04
In effect, by purchasing the bond prior to the decline in the interest rate, you
were able to buy the bond for $38.46 less than those who purchase this bond
today. This is why your yield is still 8 percent when the market rate has
fallen to 4 percent. Alternatively put, your bond has increased in value from
$1,000 to $1,038.46 due to the decline in the interest rate.
Notice that the price change is even greater on bonds with longer times
to maturity. Suppose you purchase a two-year bond with a coupon rate of 8
percent and a face value of $1,000. This bond will pay you $80 in one year
and $1,080 in two years (the $80 in interest from year 2, plus the principal).
When the market interest rate is also 8 percent, the price of this bond is
$80 $1,080
$1,000.
1 + .08 + (1 + .08)2
Now what happens if immediately after you purchase this bond, the
market interest rate drops to 4 percent? The price of the bond immediately
changes to
$80 $1,080
$1,075.44.
(1 + .04) + (1 + .04)2
Thus, the price of the two-year bond increased by $75.44, whereas the price
of the one-year bond increased by only $38.46. This difference is due to the
different times to maturity. With the one-year bond, the higher interest rate
is locked in only for one year, while the two-year bond locks in the higher
interest rate for two years.
The story differs if interest rates rise right after you purchase the bond.
To see why, suppose the interest rate rises to 12 percent immediately after
you purchase a one-year bond. In this case, your bond will still yield 8
percent. However, when the interest rate is 12 percent, similar bonds now
sell for
$80 + $1,000
B = $964.29. (9.2)
1 + .12
Your bond has declined in value due to the rise in interest rates.
This example illustrates the interest rate risk associated with purchasing
a bond: If interest rates rise, your bond will fall in value (from $1,000 when
the interest rate is 8 percent to $964.29 when it rises to 12 percent). If you
liquidate your bond right after purchasing it, you will lose a hefty $35.71—
nearly 4 percent of your initial $1,000 investment.
Capital Gains and Interest Income on a Bond. The preceding

examples highlight a general feature of investing in the bond market. The
owner of a bond actually receives income from two sources. One source is
the interest income on the bond at the time of purchase. The other source is
what is called the capital gain. The capital gain is the change in the price
relative to the purchase price. In our example of the decline in the market
interest rate, the initial coupon rate and the market interest rate were both 8
percent. An investor who purchased the one-year bond received an 8 percent
yield to maturity. After the market interest rate fell to 4 percent, the price of
the bond rose by $38.46. This is the capital gain on the bond created by the
fall in the market interest rate, and the capital gain is itself a return over and
above the 8 percent yield. The capital gain is a return of $38.46 on a purchase
price of $1,000, or $38.46/$ 1,000 = 3.8 percent. The total return is the sum
of the coupon interest rate and the capital gain, or
Total return = 8% + 3.8% = 11.8%.
How does this scenario change if the bond has a call feature? If the
interest rate rises from 8 to 12 percent and the bond is called, you will be in
great shape; you will receive $1,000 (but no coupon payment, since the bond
is paid off early). This $1,000 will be more than enough to let you purchase
a bond from another firm. In fact, equation 9.2 reveals that when the interest
rate rises to 12 percent, you can purchase a similar bond at $964.29 and earn
a yield to maturity of 12 percent. You even have $35.71 left over! In short,
if interest rates rise and your bond is called, you will do quite well.
On the other hand, if the market interest rate falls to 4 percent, the
$1,000 you receive when the bond is called will not be enough to let you
purchase another bond. Equation 9.1 reveals that at a 4 percent market
interest rate, similar bonds are selling for $1,038.46—more than the $1,000
you are paid when your bond is called. Even if you scrape up an extra $38.46
to buy a bond, the yield to maturity will be only 4 percent—considerably
lower than what you would have earned had your bond not been called.
This example might tempt you to conclude that you could gain or lose
by purchasing this callable bond depending on whether interest rates are
higher or lower when it is called. This is incorrect; you can only lose by
purchasing this bond. To see why, notice that when the interest rate rises
from 8 to 12 percent, the issuing company has no incentive to call your bond
and pay you $1,000 for it. Instead it can buy back the bond in the secondary
market at a much lower price of $964.29. In effect, since you are providing
the firm with loanable funds at a lower rate (8 percent) than the current
market rate (12 percent), it has no incentive to exercise the call feature!
In contrast, if interest rates fall to 4 percent, the firm will want to call
your bond. The firm can issue a new bond—one identical to the bond you
purchased—and receive $1,038.46 for it. The firm can use the proceeds of
this bond to pay you $1,000 and call the bond. In the process, the firm gains
$38.46 in immediate cash and now pays the lower market rate of 4 percent
instead of the 8 percent it was paying you. Thus, the issuer will call a bond
only when doing so is to your disadvantage; you are a victim of adverse
selection.
Of course, investors in financial markets recognize this problem. As a
result, they pay lower prices for callable bonds than for noncallable ones.
The lower price compensates investors for the additional risk that the bond
will be called. Figure 9.6 illustrates this principle using the inverse relation¬
ship between interest rates and bond prices. Point A represents a bond that
is not callable, and point B represents a bond with a call feature. Since
callable bonds sell at lower prices than noncallable ones, the interest rate on
callable bonds (/c) is higher than that on noncallable bonds (iN). The differ¬
ence between these two interest rates is the call premium—the additional
yield purchasers of callable bonds must receive to be willing to expose
themselves to call risk.
The Inflation Premium and Inflation Risk

When a lender provides current funds to a borrower, the borrower promises
to pay back principal and interest in the future. Since inflation erodes the
purchasing power of money, a dollar paid back in the future does not pur¬
chase the same real quantity of goods as a dollar loaned to the borrower
today.
Because investors recognize the impact of inflation on the real value of
the interest receipts they earn on loans, they require a premium to cover
their expectation of inflation. This inflation premium can be seen in the
Fisher equation we discussed in Chapter 3. To recap that discussion, suppose
r is the interest rate that would be charged in the absence of any inflation;
this is known as the real interest rate. If i is the nominal interest rate (the
interest rate stated in the loan contract) and tt6 is the expected rate of inflation,
then, by the Fisher equation we have
i = r + TTe.
Thus, the nominal (or stated) interest rate on a loan equals the real interest
rate plus the expected rate of inflation. The higher the expected rate of
inflation, the higher the nominal interest rate.
In addition to the inflation premium (or Fisher premium) due to expected
inflation, there is an inflation risk premium. Inflation risk is the risk that
the actual inflation rate will differ from the expected inflation rate. If the
inflation rate is higher than expected, the real interest rate will be lower than
expected. Recall that in Chapter 3 we distinguished between the ex ante and
ex post real interest rates. The ex ante real interest rate is the nominal interest
rate minus the expected inflation rate, while the ex post real interest rate is
the nominal interest rate minus the actual inflation rate. When the actual
inflation rate is higher than expected, the ex post real interest rate is lower
than the ex ante real interest rate.
For example, suppose the expected inflation rate is 2 percent and the
nominal interest rate on your mortgage loan is 7 percent. You expect the
inflation rate to stay at 2 percent, and the nominal interest rate on your
mortgage is fixed. The ex ante real interest rate on your mortgage loan is 5
percent. What happens if the actual inflation falls to 0 percent? In this case,
the ex post or actual real interest rate on your mortgage loan is 7 percent, a
higher rate than you intended to pay. You, the borrower, are worse off.
However, if the actual inflation rate rises to 6 percent, the actual or ex post
real interest rate on your mortgage loan will be 1 percent, a lower rate than
you expected to pay. You are better off, whereas the lender is worse off.
Both borrowers and lenders face inflation risk. Borrowers benefit from
unexpected increases in the inflation rate but are harmed by unexpected
decreases, while lenders benefit from unexpected decreases but are harmed
by unexpected increases. If risk averse, both will require an inflation risk
premium when they borrow or lend funds.
Tax Risk
Finally, market interest rates and bond prices are affected by tax risk—the
risk of changes in the tax treatment of interest income. In the United States,
interest income on corporate and government bonds is subject to personal
income tax. However, many municipal bonds are issued under a federal law
that exempts interest earnings from federal income taxation. This allows
municipalities to charge a lower interest rate and still give investors the same
aftertax interest rate as other bonds whose interest is subject to federal
taxation.
For example, suppose you are in the 38 percent tax bracket; out of each
additional dollar you earn in interest income, you must pay 38 cents to the
federal government. You have $10,000 to invest for one year and must
choose between a corporate bond with a yield to maturity of 9 percent and
a municipal bond with a yield to maturity of 6 percent. Which bond should
you purchase?
Assume for simplicity that both bonds are free of default risk and the
only difference between the bonds is the tax treatment of interest income.
Since the interest received on the municipal bond is exempt from federal
taxes, the aftertax yield on such a bond is simply the 6 percent nominal
yield. For bonds with taxable interest, the aftertax yield is the nominal yield
times (1 — Marginal tax rate). Since you must pay taxes on the income you
receive from the corporate bond, the aftertax yield is (.09)(1 — .38) =
.0558, or 5.58 percent. The municipal bond pays a 6 percent yield before
and after taxes, so you will earn a higher aftertax yield by purchasing the
municipal bond.
Because changes in tax rates on interest income affect the aftertax return
from an investment, the risk of future changes in tax policy indirectly leads
to interest rate risk. As we already saw, this results in reduced prices for
financial assets and higher market interest rates. However, since the interest
income on municipal bonds is exempt from federal taxes, the prices of
municipal bonds often rise during periods of increased uncertainty about
federal tax policy.
One particularly interesting test of the impact of risk on interest rates is
how the danger of war affects interest rates. Box 9.4 compares the interest
rates on bonds of several European countries just prior to and after Ger¬
many’s invasion of Poland at the start of World War II.
Summary (a) National Trash Collection has $10 million in outstanding bonds that will
Exercise 9.2 mature in one year. Each bond has a coupon rate of 15 percent on its face
value of $1,000, and the current default-free interest rate is 5 percent. A
recent report reveals there is a 90 percent chance National Trash Collection
will go bankrupt by year end. If this happens, the firm will be unable to
make interest payments and will be able to pay only 10 cents on the dollar
of the face value of its outstanding debt. What price would you expect its
bonds to trade for after this announcement? (b) NewCorp, a relatively young
company, has just had its bonds listed on the New York Bond Exchange.
Your friend owns some of the firm’s outstanding bonds and asks you what
impact the listing will have on the value of her bonds. How do you respond?
Explain your answer graphically.
Answer: (a) National Trash Collection is scheduled to pay $1,150 in one

year ($150 in coupon payments, plus $1,000 in face value). But there is a
90 percent chance bondholders will receive no coupon payment and only 10
cents on the dollar of face value, or $100. Thus, the expected payment by
National Trash Collection in one year is
.9($ 100) + . 1($ 1,150) = $205.
The present value of this expected payment given the default-free interest
rate of 5 percent is
$205
Present value of expected payment = ^ + = $195.24.
Thus, risk-neutral investors would pay $195.24 for this bond, and risk-averse
investors would be willing to pay even less.
International Banking Box 9=4
The Beginning of Worid War ii

and Interest Rates
History reveals that political events can affect kets react to the outbreak of World War II? To
yields and prices of financial assets. On Septem¬ find out, we gathered data on interest rates in
ber 1, 1939, Germany invaded Poland, leading London for bonds from countries in Europe and
quickly to declarations of war by France and elsewhere for the week ending August 23,
Britain. While not totally unexpected, the exact 1939, and again for the week ending Sep¬
timing of the invasion was a military secret, and tember 27, 1939. These figures are reported in
diplomats hoped until the end to avoid the com¬ the accompanying table.
ing global conflagration. How did financial mar¬
Price (British Pounds) Yield
Type of Bond Aug. 23, 1939 Sept 27, 1939 Aug. 23, 1939 Sept 27, 1939
British Consol 2.5% 63.5 62.0 6.2% 6.5%
British victory bonds 4% 103.0 102.0 3.7 3.8
Argentinean 4.5% 76.0 75.0 7.8 8.0
Belgian 4% 76.5 52.5 7.5 14.5
Danish 3% 96.5 95.0 3.2 3.3
The price of Belgian bonds fell from 76.5 to decline in price and a yield increase of from 3.2
52.5 pounds sterling in one month, and the cor¬ to 3.3 percent. Interestingly, Denmark was not
responding yield increased from 7.5 to 14.5 per¬ directly involved in the initial hostilities on the
cent. This reflected British investors' fears for the continent. British bonds also changed very little:
fate of Belgium in a continental war with Ger¬ The yield on consol bonds (perpetuities) in¬
many. If Germany invaded Belgium, the Belgian creased from 6.2 to 6.5 percent, and the yield
bonds would not be repaid. The higher yields on victory bonds rose from 3.7 to 3.8 percent.
following the invasion of Poland thus reflected a Argentinean bonds were equally stable.
higher default premium on Belgian bonds.
Meanwhile investors appeared more san¬
guine about the fates of other countries. Danish Source: The Economist, August 26, 1939, p. 428, and
bonds barely budged, with a 1.5 pound sterling September 30, 1939, p. 624.
(b) NewCorp’s bonds are now more liquid. This increases the willingness
of lenders to loan funds to NewCorp, which in turn increases the supply of
loanable funds to NewCorp from S° to Sl in part a. This reduces the interest
rate at which NewCorp can borrow from /0 1° h- Part b shows the inverse
relationship between the interest rate and bond prices. The decline in the
interest rate from i0 to q leads to an increase in the price of NewCorp’s
bonds from P0 to Px. Thus, your friend will have a more valuable bond
portfolio.
Interest Interest
Rate Rate
PQ P1 Bond
Prices
(a) Loans to NewCorp (b) NewCorp's Bond Price
Minimizing Financial Risk

Having seen in the previous section the many risks inherent in financial
markets, you may wonder why investors participate in them at all. One
reason, of course, is the potential to earn income. Another is that investors
can take steps to mitigate some of the risk. In this section, we examine some
of those steps. In particular, we describe three methods: (1) diversification,
(2) futures markets, and (3) options markets. The only way to eliminate all
financial risk is to forgo participation in financial markets. This, however, is
not what investors have in mind when they seek to minimize risk; after all,
nothing ventured, nothing gained. Instead they wish to minimize the risk
associated with obtaining any given levels of return on their investments.
For instance, suppose two assets have identical expected returns but
different variances in returns (risk). If an investor can choose only one of
these two investments, choosing the one with the lower variance will mini¬
mize the risk of obtaining a given level of return. As we will see next,
however, it is not always in an investor’s best interest to choose to make a
single investment. An investor may be better able to reduce financial risk by
making several investments simultaneously.
Diversification
Diversification means spreading your investment dollars over two or more
assets; it is the investment analogue of not putting all your eggs in one
basket. To reduce risk, you should make diversification a guiding principle
behind your portfolio selection; that is, you should take into account how
your choice of investment options will influence the riskiness of your port¬
folio.
To see how diversification can reduce financial risk, suppose you have
$100 to invest and are considering two investment opportunities. The first
option is to invest your entire $100 in a budget hotel chain. The risk lies in
your not knowing how much profit this investment will earn. Suppose that
during economic booms, budget hotels lose business to luxury hotels; your
profit from investing in the budget hotel will be only $2. During recessions,
however, budget hotels do much better; the $100 investment in a budget
hotel will yield a profit of $10. If a recession and an economic boom are
equally likely, your $100 investment will yield an expected profit of
Expected profit from budget hotel = .5($2) + .5($10) = $6.
The variance in profits if you invest in the budget hotel is
^budget hotel = -5(2 - 6)2 + .5(10 - 6)2 = 16.
The first row of Table 9.1 summarizes this information.

The second option is to invest your entire $100 in a luxury hotel chain.
This investment does better during economic booms, when your $100 in¬
vestment will net you $10 in profits, whereas you can earn only $2 during
a recession. Given the 50-50 chance of a boom or a recession, your $100
investment in a luxury hotel will yield an expected profit of
Expected profit from luxury hotel = .5($10) + .5($2) = $6.
The variance in profits if you invest in the luxury hotel is
^luxury hotel = -5(10 ~ 6)2 + .5(2 - 6)2 = 16.
The second row of Table 9.1 summarizes this information.

Notice in Table 9.1 that by investing your entire $100 in either a budget
hotel or a luxury hotel, you will earn the same expected profits of $6, which
represents a 6 percent expected return ($6/$ 100 = .06). Notice too that the
$100 investment in a budget hotel has exactly the same variance in profits
as the investment in a luxury hotel. From the standpoint of risk and return,
these two investments are identical. There is one important difference, how¬
ever. The investment in a budget hotel will yield low profits during an
economic boom but higher profits during a recession. In contrast, the luxury
hotel will do better during booms than during recessions. As we will see,
this difference in the investments gives rise to the benefits of diversifying.
Table 9.1
Diversifying Risk
This table illustrates the benefits of diversifying an investment portfolio. By sinking your entire $100
investment in either a budget hotel or a luxury hotel, you will make either $2 or $10 in profits. By
investing $50 in both projects, you guarantee profits of $6.
Profit During Profit During

Economic Boom Recession Expected Variance
Investment (Probability = 50%) (Probability = 50%) Profit in Profit
$100 investment in $2 $10 $6 $16

budget hotel
$100 investment in 10 2 6 16
luxury hotel
$50 investment in 6 6 6 0
budget hotel and
$50 investment
in luxury hotel
Suppose you invested half of your $100 in the budget hotel and half in
the luxury hotel. Your total investment is now diversified. During a boom
you will earn $1 from the budget hotel, but you will also receive $5 from
the luxury hotel. This amounts to total profits of $6 during an economic
boom. During a recession, your investment in the budget hotel will yield $5
in profits, but your investment in the luxury hotel will net you only $1.
The third row of Table 9.1 summarizes the result of your diversified
investment. Regardless of whether a boom or a recession occurs, your profits
if you diversify will be $6 and the variance in those profits will be zero. By
diversifying you have completely eliminated the risk associated with booms
and recessions. Moreover, you have done so without lowering your expected
profits. This example demonstrates the benefits to investors of owning a
portfolio of different financial assets.
In reality, of course, it is generally not possible to completely eliminate
risk by diversifying. As we saw in the previous section, real-world invest¬
ments involve many types of risk, not simply the risk of a boom or a
recession. Some risks are not diversifiable. To diversify, you must be able
to purchase two or more securities whose individual returns move differently
in response to a risky event. Some risks make the returns on all securities
move in the same direction. As an extreme example, there is no way to
diversify your portfolio against the effect of the sun going supernova and
destroying all life on earth. Less extreme examples might be the risk of
nuclear war or the risk of political insurrection (although you could keep
some assets overseas).
Diversifying is not the only way to reduce financial risk. Investors may
also use futures markets or options markets.
Futures Markets
A futures contract is a deal made now to take place at a specified future
date at a price and quantity specified today. Futures contracts are traded in
the futures market. We present a simple example that illustrates how futures
markets reduce risk for buyers and sellers of assets.
Suppose Exxon wishes to expand its refinery this year so that it can
produce more gasoline next year. The expansion project will be profitable
only if Exxon can obtain a supply of oil for $20 per barrel next year. Given
that the firm does not know what the price of oil will be next year, it decides
not to make the investment. Furthermore, suppose the owner of an oil lease
is considering installing additional wells today to increase next year’s oil
production. The new wells will be profitable only if the price of oil is at
least $20 per barrel next year. Due to uncertainty about what price oil will
be in one year, the lease owner decides not to invest in additional wells
today.
This example reveals that uncertainty about future oil prices can keep
both suppliers and demanders of oil from making investments today. Futures
contracts, however, can eliminate the uncertainty about future oil prices, thus
allowing both parties to profitably engage in investment activities today. If
the two firms wrote a contract that specified Exxon’s refinery would pur¬
chase, say, 5 billion barrels of oil from the owner of the oil lease in one year
at a price of $20 per barrel, the two firms would eliminate some of the
uncertainty about the price at which they can buy (and sell) oil in one year.
The contracted price ensures that both parties will find it profitable to make
their investments, and the contract enhances the investment by reducing
uncertainty about future prices.
Futures contracts are traded in major markets, including the Chicago
Board of Trade (CBOT), the Chicago Mercantile Exchange (CME), the
International Monetary Market (IMM) in Chicago, the Commodities
Exchange (COMEX) in New York, and the New York Futures Exchange
(NYFE). Well-established futures markets exist not only for oil futures but
also for many commodities and assets ranging from hog bellies to financial
instruments (including certificates of deposit, Treasury bills, and Treasury
bonds).
In addition to stock price listings, major newspapers print listings of
future contracts. Box 9.5 gives an example of a futures contract listed in the
Wall Street Journal. Financial futures contracts are similar to commodity
How to Read Information on

Futures Markets in the Wall Street Journal
Information on futures markets is regularly re¬ item listed, corn. Written in parentheses next
ported in the Wall Street Journal. Consider the to Corn is CBT, which stands for the Chicago
information presented in the February 8, 1994, Board of Trade, the exchange on which this
edition, which reported prices for Feb. 7, 1994. commodity trades. Next comes 5,000 bus, this
In the table called Futures Prices, look at the first Continued on p. 299
FUTURES PRICES
Lifetime Open
Monday, February 7,1994. Open High Low Settle Change High Low Interest
Open Interest Reflects Previous Trading Day. Mar 77.60 77.80 75.30 75.60 — 1.63 78.36 55.62 18,557
May 77.65 77.75 75.80 76.00 — 1.55 77.75 57.47 15,464
Lifetime Open July 77.84 77.94 76.00 76.36 — 1.29 77.94 58.30 9,555
Oct 72.15 72.25 71.50 71.55 — .70 72.25 59.81 1,999
Open High Low Settle Change High Low Interest
Dec 69.70 69.75 68.85 69.08 — .71 69.99 59.48 6,811
GRAINS AND OILSEEDS Mr 95 70.45 70.45 69.95 69.77 — .68 70.45 64.00 202
CORN (CBT) 5,000 bu.; cents per bu. Est vol 13, 500; vol Fri 9,027; open iint 57,,663, - 454.
Mar 287W 292 287Ci 291'/2 + 3 311% 232% 91,682 ORANGE JUICE■ (CTN)1-15,000 lbs.; cents per lb.
May 293 297'/4 292% 296% + 2 %
316'/. 238’/2 94,345 Mar 104.00 104.50 103.20 103.30 134.25 84.50 10,588
3,709
July 293% 298V2 293% 298 + 2 %
316'/2 241 80,874 May 107.00 107.10 106.15 106.15 134.50 89.00
Sept 278% 282 278% 28T/2 + 2 292V4 240'/2 17,692 July 109.50 109.80 109.45 109.15 + .15 135.00 106.00 1,526
Dec 263V2 266 263’/2 265’/2 + V/l 273% 236'/2 45,018 Sept 111.15 + .15 133.65 107.50 830
Mr95 270 27214 270 271% + l'/2 279'/2 253'/2 2,934 Nov 113.15 + .15 134.00 108.00 439
May 275 275 275 275 + V/l 282 273 263 Ja95 115.15 + .15 132.00 103.50 529
July 275 276'/2 275 276’/2 + l'/2 282'/2 274'/. 585 Est vol 1,400; vol Fri 1,500; open int 17,743, +435.
Est vol 38,000; vol Fri 40,434; open int 333,464, +584.
OATS (CBT) 5,000 bu.; cents per bu. METALS AND PETROLEUM
Mar 129'/4 130'/2 129V4 129% 163'/2 129 11,759 COPPER- HIGH (CMX) -25,000 lbs.; cents per lb.
May 134'/* 135% 134 134% 164 134 5,769 Feb 88.10 89.40 88.10 89.40 + .25 99.20 73.00 514
July 137 139 137 138'/2 + '/2 161'/. 137% 1,729 Mar 88.40 89.40 87.85 89.30 + .30 107.50 73.00 35,315
Sept 143 143 142V4 142'/4 + Vi 154'/2 141 301 Apr 88.65 + .45 90.10 74.50 782
Dec 145'/2 145'/2 145'/2 145% + >/4 157'/. 143 419 May 87.70 88.55 87.20 88.55 + .45 102.20 73.60 13,902
Est vol 1,000; vol Fri 634; open int 19,978, +25. June 87.70 87.70 87.70 88.65 +. .55 89.50 74.10 847
SOYBEANS (CBT) 5,000 bu.; cents per bu. July 87.90 88.70 87.30 88.70 + .60 102.95 74.20 6,598
Mar 573 678 670'/? 673% - % 754 589% 59,318 Aug 87.90 87.90 87.90 88.80 + .60 88.70 75.30 380
May 677'/2 682'/2 675 677% - 1% 751 592'/2 43,534 Sept 87.70 87.90 87.70 88.85 + .60 103.30 74.90 3,544
July 679 684 676'/2 67914 - 1 Vi 750 594'/2 35,226 Oct 88.10 88.10 88.10 89.00 + .55 88.30 75.20 138
Aug 669V2 675 669'/2 671 - 1 % 735 628 6,792 Nov 88.00 88.30 88.00 89.15 + .55 88.30 77.75 165
Sept 651'/2 655 651 651'/4 - 3>/4 689'/2 617 3,769 Dec 88.20 88.40 88.00 89.25 + .50 101.90 75.75 3,460
Nov 636'/2 640'/4 635 635'/2 - 2’/2 665% 581'/2 19,388 Mr 95 88.50 88.90 88.50 89.55 + .50 89.70 76.30 1,143
Ja95 642 644% 642 641'/2 - l'/2 670 6I8V2 1,526 May 89.40 89.40 89.40 89.65 + .50 89.40 76.85 333
Mar 650'/2 + 2 673 642 297 July 89.90 89.90 89.90 89.75 + .50 89.90 78.00 225
July 651% + 1 % 672 642V2 215 Sept 90.30 90.30 90.30 89.85 + .50 90.30 79.10 125
Nov 613 616 613 617 + 1 636 613 881 Est vol 6,700; vol Fri 14, 213; open int 67 ,558, +89.
Est vol 43,000; vol Fri 54,86; open int 170,946, - 1,145. GOLD (CMX)-100 troy oz.; $per troy OZ.
SOYBEAN MEAL (CBT) 100 tons ; $ per ton. Feb 382.50 383.70 379.00 378.90 — 7.70 415.70 333.80 1,073
Mar 194.00 196.20 193.70 194.90 + .60 237.50 185.20 32,879 Mar 386.30 384.30 384.30 379.40 — 7.70 398.30 376.50 13
May 193.90 196.10 193.80 194.70 + .10 232.00 185.50 20,062 Apr 385.80 386.50 380.50 380.70 — 7.60 418.50 335.20 67,609
July 194.20 196.00 193.90 194.80 230.00 193.00 17,510 June 387.50 388.60 382.50 382.70 — 7.60 417.20 339.40 29,093
Aug 192.90 194.20 192.90 193.00 + .40 225.00 191.50 7,051 Aug 389.30 389.60 385.50 384.80 — 7.60 415.00 341.50 5,016
Sept 190.20 192.00 190.20 190.90 + .10 210.00 189.30 4,027 Oct 392.10 392.30 391.80 387.10 — 7.50 417.00 344.00 4,063
Oct 189.00 190.00 189.00 189.50 + .50 206.00 187.10 2,186 Dec 393.50 394.50 389.50 389.40 — 7.50 426.50 343.00 13,153
Dec 187.80 189.60 187.80 188.30 + .20 209.00 185.90 6,143 Fb95 391.90 — 7.40 411.00 363.50 2,393
Ja95 190.00 190.00 189.00 189.00 + .70 200.00 186.50 581 Apr 394.30 — 7.40 425.00 385.50 3,015
Est vol 14,000; vol Fri 15,101; open int 90,439, --746. June 402.00 402.00 400.00 396.70 — 7.40 430.00 351.00 3,872
SOYBEAN OIL (CBT) 60,000 lbs., cents per lb. Aug 399.20 — 7.40 412.50 380.50 463
Mar 28.10 28.24 27.98 28.09 - .06 30.75 21.13 29,306 Oct 401.90 — 7.30 413.30 410.20 159
May 28.02 28.18 27.95 28.03 - .08 30.45 21.30 25,104 Dec 407.00 410.00 405.00 404.70 — 7.20 439.50 358.00 2,601
July 27.80 27.95 27.75 27.84 - .08 29.70 21.55 18,065 Ju96 413.40 - 7.00 447.00 370.90 830
Continued from p. 298 eighth column the lowest lifetime price, $2.7425
tells the units in which corn trades (5,000 bush¬ per bushel. The last column reports the amount
els). Next comes cents per bu., which indicates of open interest in this contract. Open interest is
that the prices quoted are in cents per bushel. the number of contracts that involve actual de¬
Beneath this initial information is data on livery and have not been canceled by offsetting
prices, quantities, and delivery dates. The first trades. As the expiration date rears, the open in¬
column gives the month in which the contract terest in the contract usually increases.
expires (i.e., the month in which delivery is due). Below the information for all the months
The next four columns provide price informa¬ are data on estimated volume, volume on Friday
tion. For example, for July 1995 the price per (February 4), and open interest. These figures
bushel of corn opened at $2.75 (275 cents), are total figures for all delivery dates. The esti¬
with a daily high of $2,765 and a low of $2.75 mated volume of all contracts traded on Mon¬
before closing at the settle (or closing) price of day, February 7, was 38,000. The number of
$2,765 per bushel. Note that the futures con¬ contracts traded on Friday, February 4, was
tract for 5,000 bushels at the settlement price of 40,434. The open interest in contracts for all
$2,765 per bushel would cost $13,825. months was 333,464. The + 584 is the addition
The sixth column gives the change in the clos¬ to outstanding contracts from the previous trad¬
ing price from the previous day. The closing price ing day.
on February 7 was 1.5 cents per bushel higher
than it was on February 4. The seventh column
reports the highest price over the life of the nal, © 1994 Dow Jones Company, Inc. All Rights Re¬
July 1995 contract, $2,825 per bushel, and the served Worldwide.
futures contracts. In a financial futures contract for U.S. Treasury bills, a

buyer might agree today to purchase $1 million in one-year Treasury bills
to be delivered in three months at a price of 95. The price of 95 reflects that
when the T-bills are delivered in three months, the buyer will pay $950,000.
Notice that if the price of T-bills rises over the next three months (and yields
on T-bills thus fall), the buyer is unaffected; she or he has locked in a price
of $950,000 for the $1 million in T-bills (and thus has locked in a yield to
maturity of 5.3 percent). The seller, however, has contracted to sell the T-
bills for $950,000 and must do so even though the prevailing market price
exceeds $950,000. Of course, the price of T-bills could have fallen, and the
seller would have a contract to sell at $950,000 even though the market price
was less than $950,000.
Options Markets
An option contract is a contract that gives its owner a right—but not an
obligation—to sell or buy an item at a prespecified price, called the strike
price, until the option expires. Options trade in options markets and, like
futures, provide a way for individuals to reduce the financial risk of fluctu¬
ations in asset prices. Unlike with futures contracts, however, the buyer of
an option contract does not have to exercise the option; that is, he or she
need not buy the asset. Options are bought and sold at an options price,
which is distinct from the price of the underlying financial asset. Options
are traded for major stocks, debt instruments, foreign currency, and even
market indexes such as the Standard and Poor’s 500 Stock Index.
A call option gives a buyer the right to buy a financial asset at a
stipulated price up to a specified expiration date. For stocks, options contracts
are almost always for 100 shares. Suppose the current price of IBM stock is
$57 per share. You pay $5 for a call option on IBM, with a strike price (the
price at which you have the option to purchase the stock) of $60 and an
expiration date of three months. Then you have paid $5 per share for the
right to buy 100 shares of IBM stock at a price of $60 per share anytime
between now and the expiration of the contract (three months). Clearly, if
the price of the stock stays below $60, you will not find it profitable to
exercise the option; it will be cheaper to buy the stock at the lower market
price. But if the price of the stock rises above $60, you can buy the stock at
$60 per share by exercising your option. In essence, a call option acts like
insurance against a dramatic rise in the stock price. This feature is particularly
useful when you want to buy IBM stock in the future and want to ensure
you don’t pay more than $60 for it.
A put option lets a buyer sell a financial asset at a specified price up to
the option’s expiration date. Suppose you pay $5 for a put option on IBM,
with a strike price (the price at which you can sell the stock) of $60 and an
expiration date of three months. Then you have paid $5 per share for the
right to sell 100 shares of IBM stock at a price of $60 per share anytime in
the next three months. Clearly, if the price of the stock stays below $60, you
will find it profitable to exercise the option; you will earn more by exercising
the option and selling the stock for $60 than by selling it at the lower market
price. But if the price of the stock rises above $60, you can sell the stock
for more than $60 by choosing not to exercise your option. The buyer of a
put option thus buys insurance against a dramatic fall in the price of the
stock.
Participants in futures and options markets can get information about
their investments by looking at the tables printed in major newspapers. Box
9.6 explains the futures options listings in the Wall Street Journal.
Summary Natalie, a long-time family friend, has accumulated $3 million in her com¬
Exercise 9.3 pany’s stock through an employee stock ownership plan. Under the terms
of the agreement, she cannot liquidate the stock until she retires in three
months. Natalie owns no other financial assets and has asked for your advice
regarding what she should do. She is concerned that the value of her stock
may fall in value during this three month period. Can the futures or options
market help her? What do you tell her?
How to Read Information on Futures

Options Markets in the Wall Street Journal
In addition to stock options, the Wall Street in the futures market itself, we have the com¬
Journal publishes information about options modity listed, the exchange on which it trades,
markets for agricultural products, precious met¬ the units in which a contract is expressed, and
als, and foreign currency. Consider the informa¬ the price per unit.
tion on corn presented in the first table under
Futures Options Prices on February 8, 1994. As Continued on p. 302
FUTURES OPTIONS PRICES

Monday, February 7, 1994. LIVESTOCK
AGRICULTURAL CATTLE-FEEDER (CME)
55,000 lbs.; cents per lb.
CORN (CBT) Strike Calls-Settle Puts-Settle
5,000 bu.; cents per bu. Price Mar Apr May Mar Apr May
Strike Calls-Settle Puts-Settle 76 0.10 0.40 0.65
Price Mar May Jly Mar May Jly 78 2.72 0.25 0.80 1.05
270 21'/2 27 30 '/a 1 2% 80 1.22 1.15 1.20 0.75 1.60 2.00
280 11% 10% 22'/2 1/4 2'/2 5'/2 82 0.37 0.50 0.60 1.90 2.90 3.40
290 4 12 17 2V? 5'/2 9Va 84 0.05 0.12 0.22 3.55 ....
300 % 7% 12% 974 IOV2 15 86 0.02 0.02 0.10
310 '/a 4% 9'/2 I8V2 18 21'/2 Est vol 234 Fri 99 calls 155 puts
320 '/a 272 7 2872 26 29 Op int Fri 1,9011 calls 5,787 puts
Est vol 5,000 Fri 3,943 calls 1,615 puts CATTLE-LIVE (CME)
Op int Fri 100,091 calls 88,276 puts 40,000 lbs.; cents per lb.
SOYBEANS (CBT) Strike Calls-Settle Puts-Settle
5,000 bu.; cents per bu. Price Apr Jun Aug Apr Jun Aug
Strike Calls-Settle Puts-Settle 70 5.00 0.20 0.52 0.77
Price Mar ;May Jiv Mar May Jiv 72 3.22 2.52 0.42 1.15 1.45
625 48% 5472 59 74 13/4 43/4
74 1.80 1.40 U5 0.97 2.00 2.45
650 2472 34 4272 Ya 6 12
76 0.85 0.65 0.57 2.02
675 6 20 29 7 17 25 78 0.32 0.27 0.20 3.47 ....
700 1 1174 20 2772 3374 41 80 0.07 0.07 5.22
725 74 63/4 1472 5172 5372 60 Est voli 5,481 Fri 14,9941 calls 6,476 puts
750 74 37/a 103/4 7672 7572 82 Op int Fri 10,815 calls 22,095 puts
Est vol1 11,000 Fri 7,135i calls 3,814 puts HOGS -LIVE (CME)
Op int Fri 100,267 calls 155,658 puts 40,000 lbs.; cents per lb.
SOYBEAN MEAL (CBT) Strike Calls-Settle Puts-Settle
100 tons; $ per ton Price Apr Jun Jiv Apr Jun Jiv
Strike Calls-Settle Puts-Settle 46 4.87 0.17 0.05
Price Mar May Jly Mar May Jly 48 3.10 7.50 0.40 0.15 0.30
185 9.85 29.70 .10 .90 1.60 50 1.77 5.70 1.05 0.32 0.55
190 5.40 6.85 8.00 .60 2.15 3.10 52 0.85 4.07 3.40 2.12 0.70 1.05
195 2.15 4.30 5.80 2.25 4.60 5.85 54 0.30 2.67 2.25 1.25 1.85
200 .90 2.80 4.50 5.80 7.85 9.25 56 0.12 1.60 1.35 2.17
210 .20 1.50 2.65 15.20 16.45 17.50 Est vol1 488 Fri 1,112' calls 1,660 puts
220 .05 .75 1.60 25.10 . Op int Fri 4,2031 calls 2,362 puts
Est vol 600 Fri 601 calls 204 puts
Op int Fri 26.365 calls 15,044 puts METALS
SOYBEAN OIL (CBT) COPPER (CMX)
60,000 lbs.; cents per lb. 25,000 lbs.; cents per lb.
Strike Calls-Settle Puts-Settle
Strike Calls-Settle Puts-Settle
Price Mar May Jiv Mar May Jly Price Mar May Jiv Mar May Jiv
26 2.100:2.170 2.140 .010 .150 .300 86 4.15 4.95 5.65 0.85 2.40 2.95
27 1.100 1.380 1.500 .020 .350 .650 88 2.80 3.85 4.55 1.50 3.55 3.85
28 .400 .840 1.050 .300 .800 1.200 90 1.80 3.00 3.65 2.50 4.45 4.95
29 .120 .550 .770 1.000 1.520 1.900 92 0.85 2.20 2.90 3.55 5.65 6.20
30 .060 .360 .550 1.950 2.350 2.700 94 0.45 1.75 2.25 5.15 7.20 760
Continued from p. 301 ing prices. Any blank would indicate that no call
The first column tells the strike price, the option exists at that price for that expiration
price at which the option to buy or sell the fu¬ date. See, for example, the blank for soybean
tures contract may be exercised. The first option meal calls at a strike price of 185 for July.
listed has a strike price of 270 cents per bushel. The fifth through seventh columns give the
The next three columns give the prices of call same information on put options, options to sell
contracts with various expiration dates. For corn 5,000 bushels of corn at the listed strike prices.
at $2.70 per bushel, the price for the option ex¬ As before, these options expire in March, May,
piring in March is $21.50; that is, for 21.50 and July.
cents per bushel, you can buy an option to pur¬ The information below the price summaries
chase 5,000 bushels of corn at $2.70 per for each commodity is similar to that at the bot¬
bushel. The current futures price is $2.915 per tom of the Corn table in the Futures Prices sec¬
bushel (see the settle price of March corn fu¬ tion (Box 9.5). There is an estimated volume for
tures in Box 9.5), so the option basically is priced all options on corn. Then Friday's volume is bro¬
so that you can purchase corn for the same ken down into call and put options, and Friday's
price in the futures market ($2,915 per bushel) open interest (contracts outstanding) is broken
by purchasing the option for $2.70 per bushel down into call and put options.
for $.215 per bushel.
The third and fourth columns report prices
of call options that do not expire until May and nal, © 1994 Dow Jones Company, Inc. All Rights Re¬
July, respectively. Notice that all prices are clos¬ served Worldwide.
Answer: There are three things for Natalie to consider: (1) the impact of
a dramatic decline in her company’s stock price before she retires, (2) the
implications of having all her money tied up in a single financial asset when
she retires; and (3) the impact of a decline in market interest rates before
her retirement. More specifically, when Natalie retires, she should consider
diversifying her risk by selling her stock and purchasing a portfolio of assets
designed to generate retirement income. Such an income-generating portfolio
would consist of various income stocks, corporate and U.S. government
bonds, and certificates of deposit.
However, even if Natalie plans to acquire such a diversified portfolio of
assets when she retires, she is subject to two types of risk over the next three
months. First, for every 1 percent decline in her company’s stock price,
Natalie will lose $30,000 in retirement funds. Of course, a rise in the stock’s
price will increase her retirement income; the point is that she faces risk
about the value of her retirement funds in three months. If Natalie is risk
averse, she can avoid the risk of a price decline by purchasing a three-month
put option on her company’s stock. Such a put option would lock in a price
at which she could sell her stock when she retires. Whether or not she should
Key Terms 303
buy the put option depends on the market price of the put option and its
strike price, as well as her own risk preferences.
Second, Natalie risks a decline in interest rates over the next three
months. If interest rates decline, when she sells the stock to buy the diver¬
sified portfolio of income-generating assets, she will have to accept a lower
return than she could earn today. To minimize this risk, Natalie may wish
to consider using the futures or options markets to lock in today the price
she will pay in three months for the portfolio of assets.
Conclusion
This chapter showed how six types of risk—default risk, liquidity risk,
interest rate risk, call risk, inflation risk, and tax risk—affect interest rates
and the prices of financial assets. We saw why risky financial assets tend to
sell at lower prices and return higher yields than less risky assets do and
illustrated this principle with graphs of the loanable funds market. We also
looked at strategies for reducing risk, incuding diversification and uses of
the futures and options markets. In the next chapter, we will complete our
picture of the pricing of financial assets by examining how the length of
loans can affect the prices and yields on financial assets.
KEY TERMS
uncertainty call premium

risk inflation risk
expected value (mean) inflation premium
variance tax risk
risk averse diversification
risk loving futures contract
risk neutral futures market
default risk capital gain
default risk premium option contract
liquidity risk strike price
liquidity premium options market
interest rate risk call option
callable bond put option
call risk

1. Briefly explain the difference between risk and (c) the default premium paid by
and uncertainty. XYZ?
2. Suppose you roll a single die and receive 6. Bonds issued by two different companies
$1 for each dot that appears on your first have the same face value, coupon rate,
roll. and maturity date, yet one bond is selling
(a) . What is the expected value of this at a much higher price than the other
risky prospect? bond.
(b) . What is the variance? (a) . Provide three potential reasons for
(c) . How much would a risk-neutral per¬ these price differences.
son be willing to pay to roll the die under (b) . Does the high-priced or the low-
the above terms? priced bond have the greater yield to ma¬
(d) . What can you say about the amount a turity? Explain.
risk-averse person would be willing to pay
7. BUS, Inc., has a $1 million bond out¬
to roll the die under the same terms?
standing that will mature in one year. The
3. Use supply and demand graphs to illus¬ bond pays a coupon rate of 10 percent,
trate the impact of the following changes which equals the default-free market inter¬
on the interest rate a firm pays for loan¬ est rate.
able funds. (a) . If investors believe BUS is certain to
(a) . An increase in the firm’s default risk meet its payment obigations, how much
(b) . A decrease in the liquidity risk of the will they be willing to pay for this bond?
firm’s bonds (b) . What is the default risk premium in
(c) . An increase in inflationary expecta¬ the market for BUS bonds?
tions 8. Consider the same scenario in problem 7,
(d) . Increased uncertainty about the future but suppose a recent report reveals a
tax treatment of interest income 2 percent chance that BUS will not make
.
4 At one point in the late 1970s, the interest any debt payments.
(a) . How much will a risk-neutral inves¬
rate on one-year U.S. Treasury bills was
tor pay for a BUS bond?
12 percent, while the interest rate on one-
(b) . What is the yield to maturity on this
year debt issued by the Swiss government
bond given the new information?
was only 3 percent.
(a) . Use supply and demand curves to il¬ (c) . What is the default risk premium
given the new information?
lustrate the equilibrium in the U.S. and
(d) . Would your answers in parts b and c
Swiss markets for one-year government
differ if investors were risk averse?
loans.
Explain.
(b) . What could account for the differ¬
ences in these interest rates? 9. Explain why an investor who purchases a
callable bond requires a call premium.
5. Moody’s recently downgraded XYZ’s
bonds from Aaa to C. What impact would .
10 One of the stocks in Joe’s portfolio is sell¬
you expect this to have on (a) the price of ing at an all-time high. Joe would like to
XYZ’s bonds, (b) the interest rate XYZ wait six months to sell to postpone paying
must pay for additional loanable funds, taxes on his capital gains.
(a) . What does Joe risk by waiting six the same amount for bonds with identical
months to sell his stock? face values, coupon payments, and matur¬
(b) . What method for avoiding this risk ities even if the issuers have different
can you suggest to Joe? probabilities of defaulting.” Is this state¬
ment true or false? Explain.
11. “If one person gains from a financial
transaction, someone else loses. Thus, fu¬ .
14 What are the similarities between futures
tures markets benefit either buyers or sell¬ contracts and option contracts? What is
ers, but not both.” Is this statement true the major difference?
or false? Explain.
.
15 You work for a major computer firm and
.
12 Provide two reasons risky bonds sell at face the risk that a rival company will
lower prices (and higher yields) than earn huge profits at the expense of your
default-free bonds. firm. Can you suggest a method for diver¬
sifying the risk you face of losing your
13 . 4 ‘Risk-neutral investors do not care about
job? (Hint: Should you buy shares of your
the variance in returns; they care only
stock or of your rival’s stock?) Explain
about the expected value of a risky pros¬
carefully.
pect. Thus, risk-neutral investors will pay
Allen, L., and T. Thurston. “Cash Futures Arbitrage Mortgage Yield.” Journal of Money, Credit, and
and Forward Futures Spreads in the Treasury Bill Banking, 13 (August 1981), 352-364.
Market.” Journal of Futures Markets, 8 (October Kolari, J. W., and V. P. Apilado. “The Cyclical Ef¬
1988), 563-573. fect of Default Risk on Industrial Bond Yields.”
Berck, P., and S. G. Cecchetti. “Portfolio Diversifica¬ Journal of Economics and Business, 37 (December
tion, Futures Markets and Uncertain Consumption 1985), 311-325.
Prices.” American Journal of Agricultural Eco¬ Miller, M. A. “Age-Related Reductions in Workers’
nomics, 67 (August 1985), 497-507. Life Insurance.” Monthly Labor Review, 108 (Sep¬
Boyle, G. W., and L. Young. “Asset Prices, Com¬ tember 1985), 29-34.
modity Prices, and Money: A General Equilibrium, Perry, L. G., D. A. Evans, and P. Liu. “Bond Rating
Rational Expectations Model.” American Eco¬ Discrepancies and the Effect on Municipal Bond
nomic Review, 78 (March 1988), 24-45. Yields.” Quarterly Journal of Business and Eco¬
Dillingham, A. E. “The Influence of Risk Variable nomics, 30 (Winter 1991), 110-127.
Definition on Value-of-Life Estimates.” Economic Roper, D. E. “The Role of Expected Value Analysis
Inquiry, 23 (April 1985), 277-294. for Speculative Decisions in the Forward Currency
Driskill, R., S. McCafferty, and S. M. Sheffrin. Market: Comment.” Quarterly Journal of Eco¬
“Speculative Intensity and Spot and Futures Price nomics, 89 (February 1975), 157-169.
Variability.” Economic Inquiry, 29 (October Tripathy, N., and R. L. Peterson. “Portfolio Risk, Ad¬
1991), 737-751. justment Costs, and Security Dealers’ Bid/Ask
Harris, W. G. “Inflation Risk as a Determinant of the Spreads.” Journal of Financial Services Research,
Discount Rate in Tort Settlements: Reply.” Jour¬ 5 (October 1991), 131-142.
nal of Risk and Insurance, 52 (September 1985), Witte, W. E. “A Short-Run Analysis of the Effects of
533-536. Portfolio Realignments Due to Money Market In¬
Kaufman, H. M., and D. E. Schlagenhauf. “FNMA novations.” Journal of Economics and Business,
Auction Results as a Forecaster of Residential 34 (1982), 193-199.
CHAPTER
The Term Structure

of Interest Rates
nancial assets, such as between government bonds and corporate bonds, are
due to differences in the assets’ risk of default. However, even bonds with
identical default risks can have different yields. In this chapter, we examine
how otherwise identical assets that mature at various times can have different
yields. Economists propose three hypotheses to explain this phenomenon.
Each theory attempts to explain some interesting data about the historical
relationship between yields and terms to maturity. We begin by examining
these data with the help of a tool known as the yield curve.
The Yield Curve
In this chapter, we focus on debt issued by essentially identical lenders; the

only differences we consider are differences in the times at which the debt
instruments mature. For instance, in April of 1994 an investor (such as a
bank or an individual) could purchase a Treasury bill maturing in August
1994 with a yield to maturity of 3.79 percent and a Treasury bill maturing
in April 1995 with a yield to maturity of 4.66 percent. Since both assets are
issued by the U.S. government, the only difference between them is their
term to maturity, and it is the term to maturity that causes the difference in
yields. This relationship is called the term structure of interest rates.
Suppose you are considering placing $100,000 in a certificate of deposit
(CD) at an insured bank. You face virtually no default risk; the only issue
you must address is whether to obtain a one-, two-, or six-month CD. On
February 5, 1994, a typical bank offered the following rates on CDs with
different terms to maturity: 3 percent on a one-month CD, 3.1 percent on a
two-month CD, and 3.2 percent on a six-month CD. Thus, the interest rate
depended on the CD’s term to maturity. Graphing this relationship between
the interest rates on these CDs and their terms to maturity generates the
curve in part a of Figure 10.1. This is known as the yield curve, and it
summarizes the yields one can earn by purchasing otherwise identical debt
instruments of varying maturities.
306
The Yield Curve 307
A yield curve is a snapshot taken at a given point in time. At another

time, the term structure of interest rates will differ from that shown in part
a of Figure 10.1. For instance, on April 19, 1990 (part b), the one-month
rate for certificates of deposit was 7.7 percent, the two-month rate was 7.8
percent, and the six-month rate was 7.9 percent.
Yield curves can also be graphed for other debt instruments, such as
Treasury bills or bonds. All you need remember when graphing a yield curve
is to hold constant the date at which the yields are relevant and the default
risk associated with the particular instrument.
308 Chapter 10 The Term Structure of Interest Rates
To illustrate that yield curves can be graphed for other financial assets
and are a snapshot at a given point in time, consider the picture for U.S.
government securities in June 1983. For purposes of this discussion, we will
refer to Treasury bills, notes, and bonds as simply government bonds. In
June 1983, a government bond maturing in 1 year had a yield of 9.66 percent,
a 3-year bond had a yield of 10.32 percent, a 7-year bond had a yield of
10.83 percent, a 10-year bond had a yield of 10.85 percent, and a 30-year
bond paid 10.93 percent. Thus, in June 1983 you could purchase government
bonds yielding anywhere from 9.66 to 10.93 percent, depending on the time
to maturity. The yield curve for government bonds on June 1983 is the
upward-sloping curve in part a of Figure 10.2.
The yield curves we have seen so far are all upward sloping: The longer
the term to maturity, the higher the yield. While this has historically been
the rule regarding yield curves, exceptions have occurred. In April 1990, for
instance, there was little difference between the yields of assets with short
terms to maturity and those with longer terms. Part b of Figure 10.2 illustrates
the yield curve for government securities on April 1990, which was essen¬
tially flat. On this date, a 3-year Treasury note had the same yield as a 30-
year Treasury note. Interestingly, a few episodes of downward-sloping yield
curves have occurred. For instance, the yield curve for government bonds
on February 1981 in part c of Figure 10.2 was downward sloping. This
Figure 10.2
Upward-Sloping, Downward-Sloping, and Flat Yield Curves
These yield curves are for government securities at three different points in time. Part a shows that in
June 1983, the yield curve for government securities was upward sloping. Part b shows that the yield
curve was flat in April 1990. Part c shows a downward-sloping yield curve for February 1981.
Source: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, various issues, and Citibase elec¬
tronic database.
Explanations of the Term Structure of Interest Rates 309
indicates that yields on longer-term bonds were lower than those on short¬
term bonds. The remainder of this chapter explains the factors that influence
the shapes of yield curves, why they are normally upward sloping, and why
they are sometimes flat or downward sloping. Box 10.1 shows yield curves
for bonds issued by various countries as of June 10, 1993.
Summary Given the following information, graph the relevant yield curves: A three-
Exercise 10.1 year government bond yields 7 percent; a three-year Aaa corporate bond
yields 8 percent; a seven-year government bond yields 9 percent; and a
seven-year Aaa corporate bond yields 10 percent.
| Answer: Since government bonds are less risky than Aaa corporate bonds,
we must graph one yield curve for the corporate bonds and another for the
government bonds. The figure plots these two yield curves. Notice that the
yield curve for Aaa corporate bonds lies above the curve for government
bonds. This reflects the default-risk premium on Aaa bonds and the differ¬
ences in the tax treatment of the interest income these bonds generate. Recall
that interest on Aaa corporate bonds is taxed by both the federal and state
governments, while interest on U.S. government bonds is taxed only by the
federal government.
Explanations of the Term Structure

of Interest Rates
We now know three important facts: (1) Yields on financial assets differ
according to their terms to maturity; (2) yield curves are usually upward
Yield Curves for Different Countries
The accompanying graph plots the yield curve was downward sloping. Also, notice the differ¬
for government bonds of the United States, ence in the level of interest rates between Japan
Japan, Britain, and Germany on June 10, 1993. and the United States on the one hand and Brit¬
Because of the difference in maturity structures ain and Germany on the other. International in¬
across countries, we include only 1-month, terest rates often vary widely, and interest rates
3-month, 6-month, and 10-year (120-month) in Europe have been relatively high since the
government bonds. Notice that while the United early 1990s as Germany strives to deal with its
States, Japan, and Britain all had upward-sloping budget problems and its reunification with the
yield curves on this date, Germany's yield curve former East Germany.
Source: International Herald Tribune, June 11, 1993;

Financial Times, June 1 1, 1993.
sloping, reflecting higher interest rates on longer-term loans than on shorter-

term ones; and (3) a few episodes of flat and even downward-sloping yield
curves have occurred. We now look at why yield curves are shaped as they
are and why their shapes may change over time. But first, Box 10.2 tells us
how to read the data in the Wall Street Journal on yield curves.
We begin with two very different theories that economists have advanced
to explain the shape of yield curves: the expectations hypothesis and the
segmented-market hypothesis. The third theory—the preferred-habitat hy¬
pothesis—combines essential elements of the first two hypotheses. As we
will see, the preferred-habitat hypothesis does the best job of explaining the
shape of yield curves.
The Expectations Hypothesis

The expectations hypothesis asserts that longer-term interest rates are an
average of the shorter-term interest rates expected to prevail during the life
of the longer-term asset. For instance, if the current interest rate on a one-
year bond is 5 percent and investors expect the interest rate on one-year
bonds to be 7 percent next year, the current interest rate on a two-year bond
will be (approximately) an average of the two one-year rates, or (.05 + .07)/
2 = 6 percent. Let us see what lies behind this theory.
Assumptions. Under the expectations hypothesis, we assume investors

view a series of short-term bonds as perfect substitutes for long-term bonds.
An investor who wishes to invest funds for, say, two years has a choice
between (1) purchasing a bond that matures in two years and (2) purchasing
a one-year bond and, when it matures, using the proceeds to purchase another
one-year bond. In either case, the investor will have invested funds for two
years. Since the expectations hypothesis assumes short- and long-term bonds
are perfect substitutes, the only factor that will affect the investor’s decision
is the expected return to be earned from purchasing a two-year bond com¬
pared to that from purchasing the two one-year bonds. Thus, the expectations
hypothesis also implicitly assumes investors are risk neutral and thus are not
willing to pay a premium to lock in a two-year interest rate. In addition, it
assumes transactions costs are zero so that the cost of buying, say, a two-
year bond is the same as that of buying a series of one-year bonds: zero.
Implications for the Term Structure of Interest Rates. Given

the assumptions that short- and long-term financial assets are perfect substi¬
tutes and that investors care only about the expected return, the only way
borrowers can attract long-term funds is to offer a long-term interest rate
that gives investors the same expected return they could earn on a sequence
of short-term investments. More specifically, suppose you have $10,000 in
cash that you will not need for two years. The current interest rate on a one-
year CD is 4 percent, and next year a one-year CD will pay interest of q
How to Read Information on the Treasury

Yield Curve in the Wall Street Journal
The yield curve in the accompanying figure is

taken from the Wall Street Journal on February Treasury Yield Curve
Yields as of 4:30 p.m. Eastern time
8, 1994. It shows the yield curve on U.S. Treas¬
ury securities (bills, notes, and bonds) with ma¬
turities from 3 months to 30 years. On this day
the yield curve was upward sloping, with the
then current one-year rate at 3.8 percent and
the two-year rate at 4.4 percent. According to
the expectations hypothesis (which we look at in
Chapter 12), the expected future one-year rate
(in February 1995) was 5 percent. Since this box
was written in February 1994, we do not know
how accurate this forecast was. You will be able
to check its accuracy by looking at a February
1995 issue of the Wall Street Journal to see if
one-year rates were really 5 percent at that
time.

nal, © 1994 Dow Jones Company, Inc. All Rights Re¬
served Worldwide.
percent. Then, if you purchase a one-year CD today and, when it matures,

purchase another CD that pays i{ percent, in two years you will have
V5 = $10,000(1 + .04)(1 + h).
Obviously your total return if you buy the series of one-year CDs depends
on what the interest rate (/]) is next year, when your first CD matures.
Unfortunately, since interest rates fluctuate randomly due to changes in
market conditions, you do not know with certainty what next year’s rate on
a one-year CD will be; all you (and other investors in the market) can do is
form an expectation of what you think it will be. Suppose you and other
investors expect the interest rate on a one-year CD to be Exi next year. For
example, if you expect the rate next year to be the same as this year’s rate,
Exi = .04. In this case, by purchasing a series of one-year CDs, you will
compound the 4 percent interest rate for two years, giving you
Vs = $10,000(1 + .04)(1 + .04) = $10,816
at the end of two years. In contrast, if you expect the interest rate in one
year to be Exi = .10, you will earn
V5 = $10,000(1 + .04)(1 + .10) = $11,440
by purchasing a series of one-year CDs. The higher you expect the future
one-year rate to be, the greater will be the value (V5) to you of purchasing
the series of one-year CDs.
Suppose the bank does not want all of its CDs to mature on the same
date; it would like some investors to purchase one-year CDs and some to
purchase two-year CDs. If the bank offers a rate of i2 on a two-year CD, the
total return to the purchaser of such a CD will be
VL = $10,000(1 + i2f.
However, since investors view short- and long-term CDs as perfect substi¬
tutes and are risk neutral, they will not purchase short-term CDs if Vs <
in that case, their expected return will be higher from purchasing the two-
year CD. In contrast, if Vs > VL, investors will not purchase long-term CDs;
the expected return will be higher if they purchase the series of short-term
CDs. In fact, under the expectations hypothesis, the only condition in which
a bank can sell both one- and two-year CDs is that where investors are
indifferent between purchasing one type or the other: VL = Vs. In this case,
the expected return from a two-year CD exactly equals that on the series of
one-year CDs. Of course, the actual return on the series of one-year CDs
may differ from what is expected, but risk-neutral investors don’t concern
themselves with this.
To see what this example implies about the term structure of interest
rates, suppose the current rate on a one-year CD is 4 percent and investors
expect the one-year rate to be 6 percent next year (Exi = .06). In this case,
the interest rate on a two-year CD that equates the expected return of holding
a two-year CD with that of purchasing a series of one-year CDs satisfies the
condition
VL = Cs
or, substituting in for Vs and VL,
$10,000(1 + i2)2 = $10,000(1 + .04)(1 + .06).
The $10,000 sum cancels out because it is on both sides of the equation.
Thus, we see that investors will be indifferent between a one-year and a
two-year CD if
(1 + i2f = (1 + .04)(1 + .06).

Taking the square root of both sides of this equation gives us
1 T i2 = V(1 + -04)(1 + -06) = 1.05.

Since i2 = .05, the only way the bank can induce investors to purchase a
two-year CD, given the current one-year rate of 4 percent and investors’
expectations of a 6 percent rate next year, is to pay 5 percent today on the
two-year CD. At this rate, investors earn exactly the same return from both
investments: Vs = Vl = $11,025.
This analysis reveals that according to the expectations hypothesis, ex¬
pectations about short-term rates determine long-term interest rates. If
expectations about future short-term rates change, the current interest rate
on long-term rates will change. For instance, if the current one-year rate is
4 percent and investors expect that rate to remain at 4 percent next year, the
interest rate on a two-year CD will satisfy the condition
(1 + .04)(1 + .04) = (1 + i2f.

Solving this equation for i2 reveals that i2 = 4 percent. Thus, if investors
expect the interest rate to be constant, the rate on one-year CDs will equal
that on two-year CDs.
Similarly, if investors expect the one-year rate to drop to 2 percent next
year, the interest rate on a two-year CD must satisfy the condition
(1 + .04)0 + -02) = (1 + i2f.

Solving for i2 reveals that the two-year interest rate is 3 percent.
Figure 10.3 illustrates the yield curves associated with the three cases
we examined. Part a shows an upward-sloping yield curve, indicating that
investors expect a higher one-year rate next year. Part b depicts a flat yield
curve, indicating that investors expect the one-year rate to remain at 4
percent. Finally, part c illustrates a downward-sloping yield curve, indicating
that investors expect a lower one-year rate next year.
A More General Formulation. Now that you understand the basic

implications of the expectations hypothesis, it is easy to extend the results
to compare yields on financial instruments with various terms to maturity.
More specifically, let i be the current interest rate on a one-year CD and Eti
be the expected interest rate on a one-year CD purchased t years in the
future. Then, if in is the current interest rate on a CD with n years to maturity,
investors will be indifferent between a series of one-year CDs and the n-
year CD if
(1 + ij = (1 + i)(l + FiDO + E2i) • • • (1 + En_xi). (10.1)

The term on the left-hand side of this equation represents the total return
(per dollar invested) if the investor purchased a CD that matures in n years
and pays an interest rate of in percent per year. The right-hand side represents
Figure 10,3
Yield Curves and the Expectations Hypothesis
In part a, short-term interest rates are expected to increase, giving rise to an upward-sloping yield curve.
In part b, short-term interest rates are expected to remain constant, generating a flat yield curve. In part
c, the yield curve is downward sloping because short-term interest rates are expected to fall.
Interest Interest Interest

Rate (%) Rate (%) Rate (%)
Term to Maturity (Years) Term to Maturity (Years) Term to Maturity (Years)

(a) Short-Term Rate (b) Short-Term Rate (c) Short-Term Rate
Expected to Rise Expected to Remain Constant Expected to Fall
the expected return if the investor purchased a one-year CD that pays i

percent, reinvested the proceeds at the end of the first year in another one-
year CD that pays Exi, reinvested those proceeds at the end of the second
year in another one-year CD that pays E2i, and so on. We may obtain an
exact expression for the relationship between the interest rate on an rc-year
CD and the expected future rates on one-year CDs by taking the nth root of
both sides of the equation 10.1 and solving for in:
l = y/a + o(i + £,o(i + e2i) • • • 0 + £„_,o - i. (io.2)

For example, if the current one-year rate is 4 percent (/ = .04) and the
expected rate on a one-year CD next year is 6 percent {Exi = .06), the
interest rate on a two-year CD (i2) will be
i2 = VO + -04)(1 + -06) “ 1 = -04995,
or 4.995 percent. Notice that this amount is very close to the simple average
of the current one-year rate (/ = .04) and the expected one-year rate next
year (Exi — .06):
i + E\i .04 4- .06
in = .04995 = .05.
In fact, a very useful approximation that we have already used for the exact
formula in equation 10.2 is:* 1
i + Eii + E2i 4- E3i + • • • 4- En_\i
in ~
That is, the interest rate on an n-year CD (in) will, under the expectations
hypothesis, approximately equal the simple average of current and expected
one-year rates over the life of the n-year bond. We will use this approxi¬
mation formula throughout the remainder of this chapter.
Using the Expectations Hypothesis to Forecast Future In¬

terest Rates. One of the more interesting aspects of the expectations
hypothesis is that we can use it to obtain a market forecast of future interest
rates. Let us see how we can accomplish this.
Suppose you have no idea whether short-term interest rates will rise or
fall; all you know is that the yield on a bond with one year to maturity is
i = 4 percent and the yield on a bond with two years to maturity is i2 = 8
percent. Let us suppose other investors in the market expect the interest rate
on a one-year bond to be Exi. Unfortunately, you do not know what Exi is
based on their forecasts; you know only the yields on one- and two-year
bonds. However, given only this information, you can use the expectations
hypothesis to determine what these other investors believe the future one-
year rate will be, even if they do not directly report their estimate to you.
Specifically, under the expectations hypothesis, the rate on a two-year
bond (i2) must equal the average of the current one-year rate (/) and the
market’s expectation of the future one-year rate (Exi):
i + Eii
1 In particular, if we take the logarithm of equation 10.1, we get
n ln(l + in) = ln(l + i) + ln(l 4 E\i) 4 ln(l 4 E2i) + • • • + ln(l 4 En-\i).
Since the logarithm of (1 4 x) approximately equals x when x is small, this equation can be
approximated by
n X in ^ i + E\i + E^i, + • • • + En — \i.
Dividing through by n gives us
i + E\i + E2i + • • • + En _ i i
n
Since you know i = .04 and i2 = .08, we may plug these numbers into the
equation to obtain
.04 + Exi
or, solving for Ex /,
Exi = (.08)(2) - .04 - 0.12.

Thus, based on what you know about current yields on one- and two-year
bonds, you can infer that the market thinks one-year bond yields will rise to
12 percent next year. Box 10.3 applies the expectations model to check
whether the market predictions of April 1990 came to pass.
The Expectations Hypothesis and Yield Curves. Our analysis

of the expectations hypothesis provides a potential explanation for the shape
of yield curves at different points in time: The varying shapes are due to
different expectations of future one-year interest rates. Unfortunately, this
explanation is far from complete. As noted earlier, with a few exceptions,
yield curves have historically sloped upward. According to the expectations
hypothesis, this would imply that investors almost always expect future rates
to be higher than current rates—an outcome the actual data on interest rates
have not borne out. In particular, short-term interest rates tend to fluctuate
randomly over time; they are as likely to rise as they are to fall. Thus, the
expectations hypothesis alone cannot fully account for the historical shapes
of yield curves in the United States. Other hypotheses, presented in the
following sections, were developed to better explain the historical yield
curves.
Summary | Suppose the interest rate on a one-year CD is 5 percent, on a two-year CD

Exercise 70.2 is 4 percent, and on a three-year CD is 8 percent, (a) Describe the shape of
the yield curve, (b) Use the expectations hypothesis to determine the market’s
forecast of the one-year rate next year, (c) What is the market’s forecast of
the one-year rate in two years?
| Answer: (a) The yield curve is V shaped; two-year rates are lower than
both one-year and three-year rates, (b) Under the expectations hypothesis,
the current interest rate on a two-year CD is the average of the expected
one-year rates during the two-year term:
i + E]i
l2 —
2
Calculating Expected Future Interest Rates
The expectations hypothesis asserts that long¬ What did investors in 1990 think the inter¬
term interest rates approximately equal the aver¬ est rate would be in 1991? The expectations hy¬
age of the expected one-year interest rates over pothesis asserts that the two-year interest rate
the term to maturity of the long-term security: in 1990, i2l equaled the average of the current
one-year rate, i, and the expected one-year rate
/ + E*\i + E2i £3/ ~F ■ ■ ■ -t- £n_1/
for April 1991. Based on data for 1990, this
/n = n '
means investors expected the one-year interest
We use the data on the government securities rate in April 1991 to be 9.04 percent. The actual
yield curve summarized in the accompanying ta¬ one-year interest rate in April 1991 was 6.24
ble from April 1990 to calculate various ex¬ percent, so investors' expectations were too
pected future one-year interest rates. We can high. Similarly, investors using the expectations
get an idea about the accuracy of the expecta¬ hypothesis in 1990 predicted that the one-year
tions hypothesis by checking whether the short¬ interest rate in April 1992 would be 8.9 percent.
term interest rates predicted in April 1990 by In reality, it was 4.3 percent. Thus, the expected
the expectations hypothesis actually came to one-year interest rates implied by the expecta¬
pass. tions hypothesis failed to detect the general
downward movement in interest rates that oc¬
curred during the early 1990s.
Interest Rate Term to
in April 1990 Maturity
8.40% 1 year tem, Federal Reserve Bulletin, TI.35, various issues, and
8.72 2 years
8.78 3 years
8.77 5 years
8.81 7 years
8.79 10 years
8.76 30 years
wmmammmmmmm
Since i = .05 and i2 = .04, we have

.05 + E\i
.04 = -L,
2
so Eii — .03. Thus, the market expects a one-year CD to yield 3 percent

next year, (c) Similarly, the current interest rate on a three-year CD is the
average of the expected one-year CD rates during the three-year term:
i + E\i + E2i
We are given i = .05 and i3 = .08, and from part b we know Exi = .03.
Substituting these values in the preceding formula gives us
.05 + .03 + E2i

.08 = --.
3
Solving this for E2i reveals that the market forecast of the interest rate on a
one-year CD in two years is a whopping 16 percent!
The Segmented-Markets Hypothesis

The segmented-markets hypothesis is at the opposite end of the spectrum
from the expectations hypothesis. It states that bonds with different maturi¬
ties are not substitutable for one another; their yields are determined inde¬
pendently of one another. In effect, the segmented-markets hypothesis views
the markets for bonds with varying maturities as separate. In each segmented
market, the yield is determined by the intersection of the supply of and
demand for each type of bond.
Part a of Figure 10.4 illustrates the market for loanable funds with a
term to maturity of one year. Equilibrium in this market occurs where
demand (D°) intersects supply (S°) at point A and the market rate of interest
on one-year loans is 4 percent. In contrast, part b presents a separate market
for loanable funds with a two-year term to maturity. The interest rate on
these bonds is determined by the demand for and supply of funds with a
two-year term to maturity and is 5 percent. Yet a third market exists for
loanable funds, with a term to maturity of three years, in part c. The interest
rate in this market is 6 percent, determined by the demand for and supply
of three-year loans.
If the interest rate on one-year loans increased to 5 percent due to an
increase in the demand in part a of Figure 10.4 from D° to D1, then, according
to the segmented-markets hypothesis, there would be no impact on the
market for two- or three-year loans. The demand and supply curves in parts
b and c would be unchanged, since participants in the financial markets do
not base two- or three-year loanable funds decisions on other interest rates;
the markets are viewed as segmented, or independent of one another. Thus,
Figure 10.4
Determination of Short- and Long-Term Rates under
the Segmented-Markets Hypothesis
According to the segmented-markets hypothesis, short-term bonds are not substitutes for long-term
bonds. Thus, separate markets exist for one-year bonds, two-year bonds, and so on. Parts a through c
show the markets for one-, two-, and three-year loans, respectively. The interest rate on a bond with a
given maturity is determined in its specific market, independently of the interest rates on bonds with
different terms to maturity. An increase in demand for one-year loans in part a, which raises the interest
rate on one-year loans, affects neither the market for two-year bonds nor the market for three-year
bonds.

(a) Market for 1-Year Loans (b) Market for 2-Year Loans (c) Market for 3-Year Loans
a change in short-term interest rates has no impact on long-term rates, and

vice versa.
The Rationale for the Segmented-Markets Hypothesis. The

basic idea behind the segmented-markets hypothesis is that investors have
preferences for financial instruments with particular terms to maturity. Sup¬
pose you have idle cash you will not need for two years, at which point you
will need the money for retirement. According to the segmented-markets
hypothesis, you will have a preference for instruments with a two-year
maturity. Purchasing and holding a two-year bond to maturity is more at¬
tractive than buying a one-year bond in each of the next two years or
purchasing a longer-term bond and selling it prior to maturity, because it
minimizes transactions costs, the time and effort needed to make the invest¬
ment decision, and interest rate risk. If interest rates rose after the initial
purchase of the bond, the price of the bond would decline, potentially leading
to a net loss when you sold it prior to maturity. You can eliminate this
interest rate risk by purchasing a two-year bond and holding it to maturity.
Effectively, by purchasing a bond with the desired term to maturity, you can
lock in the yield for the term of the bond.
The Segmented-Markets Hypothesis and Yield Curves. The

segmented-markets hypothesis thus implies that the relative supply of and
demand for bonds and other financial instruments of varying maturities
determines the shape of the yield curve. For a given demand for loanable
funds, when the supply of short-term loanable funds is greater than the supply
of long-term loanable funds, short-term rates will be lower than long-term
rates. For example, in part a of Figure 10.5 the equilibrium interest rate on
one-year loans is 5 percent, and in part b the equilibrium interest rate on
two-year loans is 6 percent. These interest rates give rise to the yield curve
in part c. Notice that the vertical axis in part c is the same as that in parts a
and b, but the horizontal axis is the term to maturity. The yield curve in part
c slopes upward because the supply of one-year loans is relatively greater
than that of two-year loans.
In contrast, if for a given demand the supply of short-term loanable
funds is less than that for longer-term loans, short-term rates will exceed
long-term interest rates as illustrated in parts a and b of Figure 10.6. This
gives rise to the downward-sloping yield curve in part c.
While the segmented-markets hypothesis provides a possible explanation
for both upward- and downward-sloping yield curves, it is subject to two
criticisms. First, it is not very useful in predicting changes in the pattern of
yields; it states only that changes in preferences for loans of different ma¬
turities will change the shape of the yield curve. Second, the theory implies
that short-term and long-term interest rates are competely unrelated. Yet
empirical evidence reveals that long- and short-term interest rates tend to
move together; when short-term rates rise, long-term interest rates also tend
to rise. For this reason, the segmented-markets hypothesis provides only an
incomplete explanation of the term structure of interest rates.
The Preferred-Habitat Hypothesis

The preferred-habitat hypothesis combines key features of both the ex¬
pectations and segmented-markets hypotheses. It says that while investors
have a preference for loanable funds of a given term, they are willing to
substitute away from their preferred terms if they are compensated for doing
so. The compensation required to induce investors to purchase an asset with
a different term to maturity than their preferred terms is known as the term
premium.
The Rationale for the Preferred-Habitat Hypothesis. The

rationale for the preferred-habitat hypothesis is a combination of the ration¬
ales for the expectations and segmented-markets hypotheses. Since we al¬
ready examined the other two theories, let us look at the preferred-habitat
hypothesis in the context of an analogy.
Suppose you must choose between two jobs: a job in the Bahamas or a
job in Antarctica. The job descriptions are identical; the only difference is
the habitat in which you will live. If we applied the expectations hypothesis
to job choice, you would care only about the expected income from each
job. Thus, in equilibrium an employer in the Bahamas must pay the same
wage as an employer in Antarctica to get you to be indifferent between the
two jobs. In contrast, if we applied the segmented-market hypothesis, you
would have a preference for one job over the other independent of the pay
differentials. For example, you would choose to work in the Bahamas re¬
gardless of how much more you could earn working in Antarctica. In other
words, even if the job in the Bahamas paid $100 per year and the job in
Antarctica paid $2 million annually, application of the segmented-markets

hypothesis would lead you to choose the job in the Bahamas. You would
have a preference for a particular job location and would be unwilling to
substitute alternatives.
In contrast, the preferred-habitat hypothesis, when applied to job choice,
combines these two features in the following way. You have a preference
for the habitat of the Bahamas, but you also care about how much you expect
to earn in your job. All else equal, you would prefer to work in the Bahamas.
But if a company in Antarctica offered you a premium—say, $ 100,000 above
what you expected to earn in the Bahamas—you would choose the job in
Antarctica.2 The premium the company offered would compensate you for
the fact that Antarctica was not your preferred habitat.
2 In the labor market, this premium is often called a compensating wage differential.
In the context of financial markets, the preferred-habitat hypothesis as¬

serts that an investor will choose bonds on the basis of both the expected
return of the bond and the investor’s preference for bonds with a particular
maturity. An investor will purchase a bond with an “undesirable” term to
maturity only if he or she receives a term premium. The term premium is
simply the additional yield required to induce the investor to purchase a
bond with a term that is not exactly in line with the investor’s preferences.
If we let an be the term premium required on a bond with a term to
maturity of n years, the preferred-habitat hypothesis states that the yield on
the bond, in, is determined by the term premium plus the average of the
expected one-year interest rates over the life of the n-year bond:
i + E\i + E2i + E-xi + • • • + En^\i

in = + -L.~~---—, (10.3)
n
Here / is the interest rate on one-year bonds and Eti is the expected interest
rate on one-year bonds in year t. Thus, the interest rate on an n-period bond
equals the average of the one-year rates that are expected to prevail during
the life of the long-term bond, plus the term premium required to induce
investors to hold an n-period bond. The term premium is required to com¬
pensate the investor for the additional risk and/or transactions costs invoved
in purchasing a bond that does not match his or her immediate preferences.
For instance, suppose you have a preference for one-year bonds because you
need funds at the end of the year. Buying a two-year bond and selling it at
the end of one year (prior to maturity) involves risk; if interest rates rise
between the time you purchase or sell the bond, you will be forced to sell
the bond at a lower price than you would obtain otherwise. Moreover, you
will have to pay additional sales commissions to a broker to liquidate the
bond prior to maturity. (At maturity, the firm charges no commission to pay
you the face value of the bond.) For these reasons, you must receive a term
premium to buy longer-term bonds when your preference is for shorter term
bonds.
The expectations and segmented-market hypotheses are actually extreme
versions of the preferred-habitat hypothesis. Under the segmented-markets
hypothesis, the term premium is effectively infinite; there is no finite interest
rate that one can offer to induce an investor to purchase a bond whose
maturity does not exactly match her or his preferences. Under the expecta¬
tions hypothesis, the term premium is zero; in is simply the average of
expected one-year rates during the term of the long-term bond. In this case,
investors do not require any premium to be willing to hold bonds of different
maturities.
The Preferred-Habitat Hypothesis and Yield Curves. The

beauty of the preferred-habitat hypothesis lies not only in its intuitive appeal
in combining the best aspects of the other two hypotheses but also in its
power in explaining the shapes of yield curves. The preferred-habitat hy-
pothesis explains not only upward-sloping yield curves but flat and down-
ward-sloping ones as well. Unlike the other hypotheses, it also explains why
yield curves are more frequently upward than downward sloping. Moreover,
since an increase in the short-term interest rate (/) in equation 10.3 leads to
an increase in the long-term interest rate (/„), the preferred-habitat hypothesis
implies that short-term and long-term interest rates will tend to move to¬
gether. As noted earlier, this prediction is highly in accord with empirical
data on interest rates and is also predicted by the expectations hypothesis.
A simple example will illustrate these facts.
Suppose the term premium required on two-year bonds is a2 = .01 (1
percent). Further, assume the interest rate on a one-year bond is 6 percent.
According to the preferred-habitat hypothesis, the interest rate on a bond
that matures in two years will be
.06 + Eii
h .01 + (10.4)
where Exi is the expected interest rate next year on a one-year bond. Let us
calculate the interest rate on the two-year bond (/2) under three different
scenarios regarding expectations about future interest rates {E{i).
Case A: E7i = .20. In this case, investors expect a substantial increase

in the future interest rate on one-year bonds. If we insert this expectation
into equation 10.4, we get the corresponding interest rate on a two-year
bond:
.06 + .20
h .01 + - = .14.
2
Since the current interest rate on a one-year bond is i = .06 and the interest
rate on a two-year bond is i2 = .14, the yield curve in this case is upward
sloping, as illustrated in part a of Figure 10.7. The upward-sloping yield
curve reflects the fact that short-term rates are expected to rise substantially
in the future.
Case B: E/i = .04. In this case, investors expect the interest rate on
one-year bonds to be slightly lower next year. Setting Exi — .04 in equation
10.4 gives us the interest rate on a two-year bond:
.06 + .04
h — .01 + .06.
Since the current interest rate on a one-year bond is / = .06 and the interest
rate on a two-year bond is i2 = .06, the yield curve in this case is flat, as
illustrated in part b of Figure 10.7. The flat yield curve reflects that short¬
term rates are expected to fall moderately. In fact, the expectations hypothesis
would suggest that the interest rate on two-year bonds is 5 percent. But when
Figure 10.7
The Preferred-Habitat Hypothesis and Yield Curves
Part a shows that when the term premium is positive and short-term interest rates are expected to rise,
the yield curve will be upward sloping. Part b shows that when the term premium is positive and short¬
term interest rates are expected to fall by only a small amount, the yield curve will be relatively flat. In
part c, the term premium is positive and short-term interest rates are expected to fall by a substantial
amount; thus, the yield curve is downward sloping.

Flat
Yield
Curve
1 2
Term to Maturity (Years) Term to Maturity (Years) Term to Maturity (Years)
(a) Short-Term Rates (b) Short-Term Rates (c) Short-Term Rates

Expected to Rise Expected to Fall Expected to Fall
Moderately Dramatically
the term premium of 1 percent is added in, the yield on the two-year bond
is 6 percent—the same as that on the current one-year bond.
Case C; E7i = .01. In this case, investors expect the interest rate on
one-year bonds to fall dramatically next year. Setting Exi = .01 in equation
10.4 gives us the interest rate on a two-year bond:
.06 + .01
h .01 + .045.
~2 ~
Since the current interest rate on a one-year bond is i = .06 and the interest
rate on a two-year bond is i2 = .045, the yield curve is downward sloping
in part c of Figure 10.7. The downward-sloping yield curve reflects the fact
that short-term rates are expected to fall so sharply that even when a term
premium is applied to the two-year bond, the current two-year rate is below
the current one-year rate.
Conclusion 327
These three cases reveal that the preferred-habitat hypothesis is capable

of explaining the various shapes yield curves take. Moreover, even if short¬
term interest rates are expected to remain constant or fall slightly, the yield
curve will slope upward due to the term premium on longer-term bonds. A
downward-sloping yield curve results only when investors expect future
short-term interest rates to fall dramatically. Thus, in most instances yield
curves will slope upward. Taken together, all the implications of the
preferred-habitat hypothesis accord very well with the empirical data on
interest rates presented at the beginning of this chapter.
Summary Suppose investors prefer one-year bonds to three-year bonds and will pur¬
Exercise 10.3 chase a three-year bond only if they expect to receive an additional 2 percent
over the return from holding one-year bonds. Currently one-year bonds yield
3 percent, but investors expect this yield to rise to 4 percent next year and
to 6 percent the year after, (a) Which of the three models of term structure
is relevant in this case? (b) Is the yield curve upward sloping, flat, or
downward sloping? (c) What is the yield on a three-year bond?
(a) Since investors have a preference for one-year bonds but are
willing to hold three-year bonds if they receive a term premium of 2 percent,
this scenario is consistent with the preferred-habitat hypothesis. Under the
expectations hypothesis, investors would require no term premium; under
the segmented-markets hypothesis, investors would not purchase three-year
bonds for any premium, (b) Since short-term rates are expected to rise, the
yield curve must be upward sloping. (Note: You should know this answer
even before working part c of this exercise.) (c) Under the preferred-habitat
hypothesis, the interest rate on a three-year bond is
i + E] i + E2i
h = «3 + ---^
Here a3 = .02, i = .03, Exi = .04, and E2i — .06. Thus, the yield on a
three-year bond is
.03 + .04 -T .06

i3 = .02 H-= .063,
3
or 6.3 percent. Notice that the interest rate on a three-year bond is higher
than that on a one-year bond, which means the yield curve is upward sloping.
Conclusion
In this chapter, we examined the relationship between the yield on a financial
asset and the asset’s term to maturity. Unlike in Chapter 9, where we saw
that differences in risk can explain differences in asset prices, in this chapter
we noted that the term structure of interest rates gives rise to different yields
on financial instruments with identical risk characteristics. Together this and
the last two chapters have provided insight on how the time value of money,
risk, and the term structure of interest rates affect the prices and yields of
financial assets held by banks and other investors. In the next chapter, we
examine how this picture is affected by activities in the foreign exchange
market—the market in which currencies like Japanese yen and British
pounds are bought and sold.
KEY TERMS
term structure of interest rates segmented-markets hypothesis
yield curve preferred-habitat hypothesis
expectations hypothesis term premium

1. What is held constant along a yield curve? Treasury bills. Is the term structure of in¬
What is allowed to vary? terest rates increasing, decreasing, or flat?
Does this imply anything about expected
2. Explain the rationale for the expectations future interest rates? Explain.
hypothesis. What are the theory’s major
weaknesses? 6. Explain the rationale for the segmented-
markets hypothesis. What are this theory’s
3. Suppose that on January 1, 1995, the in¬ major weaknesses?
terest rate on a one-year government bond
7. According to the preferred-habitat hypoth¬
is 5 percent, the interest rate on a two-
esis, if the market expects future one-year
year government bond is 6 percent, and
interest rates to be the same as today’s
the interest rate on a three-year govern¬
one-year interest rate, will the term struc¬
ment bond is 7 percent. According to the
ture of interest rates be increasing, de¬
expectations theory of the term structure
creasing, or flat? Why?
of interest rates, what is the expected in¬
terest rate on one- and two-year bonds is¬ 8. Explain how the expectations and
sued on January 1, 1996? What is the ex¬ segmented-markets hypotheses are ex¬
pected interest rate on a one-year bond treme versions of the preferred-habitat
issued on January 1, 1997? hypothesis. Give an example to illustrate
your explanation.
4. According to the expectations hypothesis,
when would the yield curve slope up¬ 9. The interest rate on a one-year govern¬
ward? When would it slope downward? ment bond is 4 percent, the rate on a two-
year CD is 7 percent, and the rate on a
5. Get a current issue of the Wall Street three-year corporate bond is 9 percent.
Journal and look up the interest rate on What can you infer about the term struc¬
three-month, six-month, and one-year ture of interest rates? Explain.
10. The interest rate on a one-year govern¬ (c) . M’Lissa wishes to earn a high ex¬
ment bond is 4 percent, the rate on a two- pected return on her investments but has a
year government bond is 7 percent, and preference for five-year bonds. She will,
the rate on a three-year government bond however, consider purchasing shorter- or
is 9 percent. longer-term bonds if they offer a substan¬
(a) . Describe the shape of the yield curve. tially higher yield.
(b) . Use the expectations hypothesis to (d) . Jay prefers one-year bonds even if
determine the market’s forecast of the they have a lower yield than any other
one-year rate next year. bond.
(c) . What is the market’s forecast of the
one-year rate in two years? 13. Suppose investors expect inflation to re¬
main stable this year but increase dramati¬
11. Suppose investors prefer one-year bonds cally next year.
to two-year bonds and will purchase a (a) . What impact will this have on expec¬
two-year bond only if they expect to re¬ tations of next year’s short-term interest
ceive an additional 4 percent over the re¬ rate? Explain.
turn from holding one-year bonds. (b) . How will this expectation affect cur¬
Currently one-year bonds yield 5 percent, rent one-year interest rates? Explain.
but investors expect the yield to fall to (c) . How will this expectation affect cur¬
4 percent next year. rent two-year interest rates? Explain.
(a) . What is the yield on a two-year
bond? 14. Suppose a company’s bonds have recently
(b) . Is the yield curve upward sloping, been downgraded from Aaa to C. Assum¬
flat, or downward sloping? ing the firm has outstanding bonds matur¬
ing in 1, 5, and 10 years, how will this
12. Determine whether the following scenar¬ announcement affect the yield curve for
ios are most consistent with the expecta¬ the company’s bonds?
tions, segmented-markets, or preferred-
habitat hypothesis. 15. A mortgage applicant was perplexed when
(a) . Natalie’s sole criteria in choosing he learned that the mortgage company
among bonds with varying maturities is was charging a lower rate on a 15-year
the expected yield she will earn on the mortgage than on a 30-year mortgage. Ex¬
bonds. plain to the applicant why a mortgage
(b) . Since Mitchell plans to retire in 20 company might have such a term structure
years, he will purchase 20-year bonds. of mortgage rates.
Lee, T. K., and Y. K. Tse. “Term Structure of Inter¬ Mankiw, N. G., and J. A. Miron. “The Changing Be¬
est Rates in the Singapore Asian Dollar Market." havior of the Term Structure of Interest Rates.”
Journal of Applied Econometrics, 6 (April-June Quarterly Journal of Economics, 101 (May 1986),
1991), 143-152. 211-228.
MacDonald, S. S., and S. E. Hein. “Futures Rates and McCulloch, J. H. “The Tax-Adjusted Yield Curve.”
Forward Rates as Predictors of Near-Term Treas¬ Journal of Finance, 30 (June 1975), 811-830.
ury Bill Rates.” Journal of Futures Markets, 9
(June 1989), 249-262.
CHAPTER
Foreign Exchange Markets
ach morning, national TV and radio news programs begin their busi¬
ness reports with a quick rundown of the previous day’s activity in the New
York, Tokyo, and London stock markets. These reports usually include
closing quotes for the Dow and Nikkei averages and a summary of exchange
rates (“the yen closed at 107,” “the pound dropped sharply in response to
the jump in U.S. interest rates,” “the Fed intervened to boost the dollar”).
Perhaps you have wondered what these reports mean and why they change
so frequently.
This chapter provides answers. First, we look at what exchange rates are
and how changes in them affect prices of the foreign goods, such as cars,
TVs, and CD players, that you purchase. Next, we examine why exchange
rates change in response to changes in price levels or interest rates on foreign
and domestic deposits and how currency traders’ short-term incentives to
hold U.S. or foreign bank deposits depend on their expectations. We also
look at how interest rates, exchange rates, and price levels are linked by
international arbitrage conditions and how central bank interventions affect
exchange rates. We conclude by using our knowledge of exchange rates to
examine the history of world exchange rate systems, including the European
monetary system.
Fundamentals of Exchange Rates

Thus far, we have been concerned mainly with domestic financial markets
in which U.S. residents hold dollar deposits at banks and/or purchase U.S.
stocks or bonds. An important link between these markets and the global
economy is the foreign exchange market, the market in which different
currencies are bought and sold. This over-the-counter market is composed
of hundreds of banks and other dealers in foreign currency that buy and sell
deposits of foreign currency. A very small fraction of foreign exchange
transactions—primarily those of tourists—is exchanges of actual dollar bills
and notes issued by other governments. Most of the transactions occur over
the phone or by computer and are merely bookkeeping entries that shift bank
deposits denominated in one currency into another.
Foreign exchange markets arise because various countries have different
monetary systems and require different currencies to buy goods, services,
330
Fundamentals of Exchange Rates 331
and financial assets. If your mutual fund wishes to purchase stocks or bonds
in the British financial market, it must pay for them in British pounds (£).
Similarly, a U.S. import dealer who wishes to buy Japanese cars to sell in
the United States must pay for the cars in yen (¥). Transactions such as these
have become increasingly important to the U.S. economy.
To put this importance in perspective, in 1993 U.S. gross domestic
product, a measure of the total output of finished goods and services, was
more than $6 trillion. Of that amount, more than 12 percent was exported
to other countries, and our imports of foreign goods were even larger. The
foreign exchange market makes all of these transactions possible by facili¬
tating the exchange of other currencies into dollars, and vice versa. The
number of units of one currency that must be given up to buy a unit of
another currency is called the exchange rate. As Box 11.1 shows, these
rates are published daily in some business newspapers.
Before we explain how market forces determine exchange rates, you
need to understand how to interpret and use those rates. When you buy a
loaf of bread, you probably see the $2 price tag without thinking about what
it means. The price of $2 means each loaf of bread can be exchanged for
$2. Stated differently, $1 buys one-half loaves of bread. In terms of value,
1
- loaves of bread = $ 1.
2
Just as the price of bread reveals how much bread you can buy for $ 1,
an exchange rate reveals how much one unit of a particular currency costs
in terms of another. For instance, suppose you plan to visit Germany and
need German marks to make purchases there. If it costs 50 cents to buy one
mark (DM), then, in terms of value we would say,
$.50 = DM1,
or there are .50 dollars per mark. The exchange rate of dollars in terms of
marks is 50 cents per mark. Put another way, you can sell each dollar for
two marks:
$1 = DM2.
When expressed this way, we say the exchange rate of marks in terms of
U.S. dollars is two marks per dollar. In the United States, it is more common
to state the exchange rate as the number of marks (or other currency) one
will receive per dollar. Thus, if the exchange rate for German marks is
DM1.7/$1 (as it was in early 1994), there are 1.7 marks per dollar.
The exchange rate lets you determine how much it would cost in dollars
to purchase a good whose price is quoted in another currency. For instance,
suppose you were on a business trip in Japan in early 1993 and noticed that
a camera you could buy for $90 in the United States was priced at 10,000
yen (¥10,000) in Tokyo. Should you have bought the camera in Tokyo or
waited until you returned to the United States?
332 Cha pterII Foreign Exchange Markets
Exchange Rate Information in

the Wall Street Journal
Each edition of the Wall Street Journal contains EXCHANGE RATES

a table of exchange rates such as that shown Tuesday, March 1, 1994
The New York foreign exchange selling rates below apply to
here. This table iists various countries, with their trading among banks in amounts ot $1 million and more, as
quoted at 3 p.m. Eastern time by Bankers Trust Co., Dow Jones
currencies in parentheses, followed by a series Telerate Inc. and other sources. Retail transactions provide
fewer units of foreign currency per dollar.
of exchange rates. The table reproduced here is Currency
U.S. $ equiv. per u.s.s
from the Wednesday, March 2, 1994 edition, Country Tues. Mon. Tues. Mon.
Argentina (Peso). 1.01 1.01 .99 .99
and the rates quoted are for both Tuesday and Australia (Dollar) . .7130 .7133 1.4025 1.4019
Austria (Schilling). .08325 .08350 12.01 11.98
Monday. Bahrain (Dinar) . 2.6529 2.6529 .3770 .3770
Belgium (Franc) . .02845 .02853 35.15 35.05
Rates are given in both U.S. dollar equiva¬ Brazil (Cruzeiro real) . .0015518 .0015710 644.40 636.55
Britain (Pound) . 1.4896 1.4860 .6713 .6729
lents (how many dollars you can purchase with 30-Day Forward . 1.4875 1.4838 .6723 .6739
90-Day Forward . 1.4845 1.4808 .6736 .6753
a unit of the foreign currency) and currency per 180-Day Forward . 1.4812 1.4775 .6751 .6768
I
Canada (Dollar) . .7403 .7398 1.3508 1.3518
U.S. dollar (how many units of the foreign cur- 30-Day Forward. .7402 .7396 1.3510 1.3520
I
90-Day Forward . .7400 .7394 1.3514 1.3525
rency you can purchase with one U.S. dollar). 180-Day Forward. .7394 .7388 1.3524 1.3535
Czech. Rep. (Koruna)
The current market rate (called the spot rate) is Commercial rate . .0334795 .0334795 29.8690 29.8690
Chile (Peso) . .002400 .002400 416.59 416.59
given for all the countries listed. Forward rates China (Renminbi) . .114882 .114882 8.7046 8.7046
Colombia (Peso) . .001220 .001220 819.71 819.71
are also given for the major currencies. The for¬ Denmark (Krone) . .1494 .1499 6.6928 6.6701
Ecuador (Sucre)
ward rate indicates the rate at which you can Floating rate. .000493 .000493 2030.00 2030.00
Finland (Markka) . .18056 .18076 5.5382 5.5321
buy foreign currency to be delivered at a speci- France (Franc) . .17191 .17273 5.8170 5.7895
30-Day Forward . .17147 .17227 5.8320 5.8049
fied future date (usually 30, 90, and 180 days). 90-Day Forward . .17082 .17195 5.8540 5.8155
180-Day Forward . .17008 .17090 5.8795 5.8514
Consider the Canadian dollar. The exchange Germany (Mark) . .5855 .5870 1.7080 1.7035
30-Day Forward. .5841 .5856 1.7121 1.7077
rate on Tuesday was .7403 U.S. dollars per Cana¬ 90-Day Forward . .5822 .5837 1.7175 1.7132
180-Day Forward . .5802 .5817 1.7234 1.7190
dian dollar, or 1.3508 Canadian dollars for every Greece (Drachma) . .004034 .004052 247.90 246.80
Hong Kong (Dollar) .... .12939 .12943 7.7287 7.7263
U.S. dollar. The forward rates (in currency per Hungary (Forint) . .0096740 .0096404 103.3700 103.7300
India (Rupee) . .03216 .03216 31.09 31.09
U.S. dollar) were 1.3510 Canadian dollars per Indonesia (Rupiah) .... .0004664 .0004664 2144.04 2144.04
Ireland (Punt). 1.4289 1.4291 .6998 .6997
U.S. dollar 30 days forward, 1.3514 Canadian Israel (Shekel) . .3339 .3339 2.9945 2.9945
Italy (Lira) . .0005910 .0005925 1691.96 1687.76
dollars per U.S. dollar 90 days forward, and
1.3524 Canadian dollars per U.S. dollar 180
days forward. The exchange market is predicting
a depreciation of the Canadian dollar relative to
the U.S. dollar, since the forward market is indi¬
cating that a U.S. dollar can buy even more Cana¬
dian dollars per U.S. dollar as we move into the nal, © 1994 Dow Jones & Company, Inc. All Rights Re¬
future. served Worldwide.
Fundamentals of Exchange Rates 333
Obviously the answer to this question depends on the exchange rate.

The exchange rate (on February 5, 1993) between yen and U.S. dollars was
¥134/US$. Taking the reciprocal of this tells us the number of dollars needed
to buy one yen: 1/(¥134/US$) = US$.00746/¥. In other words, it costs a
little less than a penny to buy one yen. If we multiply this figure by the
number of yen needed to purchase the camera (¥10,000), we get
$.00746/¥ X ¥10,000 = $74.60.
(Notice the ¥ signs cancel out on the left-hand side of this equation, which
means we have converted yen to dollars). Thus, given the exchange rate on
February 5, it would have been cheaper for you to purchase the camera in
Japan than to purchase it when you returned to the United States. Yet if you
had been in Japan in February of 1994, when the exchange rate was about
¥100/US$, you would have been better off buying the camera stateside.
When the exchange rate falls, even if the price of the camera remains
the same (¥10,000), it will now cost you more in dollars to buy the camera
in Japan. Why? Because the value of the dollar has depreciated against the
yen. Depreciation of the U.S. dollar means the dollar sells for fewer units
of foreign currency than before. For example, when the exchange rate be¬
tween yen and U.S. dollars falls to ¥100/$, a dollar can be sold for only 100
yen. The reciprocal tells us the number of dollars it would take to buy one
yen at this lower exchange rate: l/(¥100/$) = $.01/¥. Due to the depreciation
of the dollar, each yen now costs 1 cent. Multiplying the dollar cost per yen
($.01/¥) by the number of yen needed to purchase the camera (¥10,000)
reveals that the cost of buying the camera in Japan rises to
$.01/¥ X ¥10,000 = $100.
When the dollar depreciates relative to the yen, Japanese goods become
relatively more expensive to Americans. In Chapter 17, we will see how the
depreciation of the dollar affects the balance of trade.
The U.S. dollar might also appreciate relative to the Japanese yen. When
the dollar appreciates, holders of foreign currency have to give up more of
their currency to buy a dollar than before. As the dollar appreciates relative
to the yen, not only will the dollar buy more yen than previously, but
Japanese goods will also become less expensive in terms of U.S. dollars.
Depreciation and appreciation are always relative to another currency. Thus,
when the U.S. dollar appreciates relative to the yen, the yen must be depre¬
ciating relative to the dollar: If a U.S. dollar now buys more yen, a Japanese
yen now must buy less of a U.S. dollar.
Finally, as Box 11.2 shows, the U.S. dollar can appreciate relative to
one currency while simultaneously depreciating relative to another. In the
next section, we look at the market for foreign exchange and see what causes
the dollar to depreciate or appreciate relative to foreign currencies.
334 Chapter 1 1 Foreign Exchange Markets
Exchange Rate Movements

Between the US. Dollar and
Four Major Foreign Currencies
The accompanying figure illustrates exchange generally appreciated relative to the Canadian
rate movements between the U.S. dollar and dollar and the British pound over the last three
four major currencies: the German mark, the decades.
Japanese yen, the British pound, and the Cana¬ In contrast, part b shows the exchange
dian dollar. There has been considerable volatil¬ rates with the German mark and the Japanese
ity in all these exchange rates, and the long-run yen. Here the general trend for each rate is
trends tend to vary across currencies. For in¬ downward, so the dollar has generally depreci¬
stance, part a of the figure graphs the exchange ated relative to the mark and the yen. For in¬
rates with the Canadian dollar and the British stance, it took more than four German marks to
pound. Notice that the general trend in purchase a dollar in the early 1960s, but by
exchange rates between U.S. dollars and these
two currencies is upward, so the dollar has Continued on p. 335
(a) Exchange Rates, Canadian Dollar per U.S. Dollar (b) Exchange Rates, German D-Mark per U.S. Dollar
(Left Scale) and U.K. Pound per U.S. Dollar (Left Scale) and Japanese Yen per U.S. Dollar
(Right Scale) (Right Scale)
Continued from p. 334 relative to some currencies and depreciate rela¬

1993 it took only 1.6 marks. tive to others.
The graphs in the figure indicate not only
that exchange rates are volatile over short per¬
iods of time but also that over long periods of tem, Foreign Exchange Rates, various issues; and Citibase
time, the dollar may simultaneously appreciate electronic database.
Summary (a) Suppose the exchange rate between British pounds (£) and U.S. dollars
Exercise 11.1 is .75 pounds per U.S. dollar. How much would it cost in dollars to purchase
a watch that sells for 28 pounds? (b) If the exchange rate changes to one
pound per U.S. dollar, how would your answer change?
Answer: (a) When the exchange rate is £.75/$, it costs $l/£.75 =

$1.33/1£ to buy a pound. Consequently, it costs
$1.33/£ X £28 = $37.24
to buy the watch, (b) When the exchange rate is l£/$, it costs
$l/£ X £28 = $28
to buy the watch. Since the dollar appreciated relative to the pound, the
dollar cost of the British watch is now lower.
Th£ Market for Foreign Exchange
With this understanding of how to use exchange rates to convert prices from
one currency to another, we now look at how market forces determine
exchange rates. Throughout this section, we will focus exclusively on the
determination of the exchange rate between yen and dollars. The basic
principles developed here are also valid for determining other exchange rates,
such as that between British pounds and U.S. dollars or even that between
Mexican pesos and Italian lire.
Before we begin, however, look at Figure 11.1. This figure shows that
the exchange rate between yen and dollars has fluctuated widely during the
past two decades. Between 1976 and 1979, the dollar depreciated from about
300 to 180 yen per dollar. Between 1979 and 1985, the dollar appreciated
from 180 to 260 yen per dollar, only to depreciate once again to 120 yen
per dollar by the end of 1988. Each of these periods reflects a long-run
depreciation or appreciation of the dollar. However, notice that within each
period, the exchange rate movements were very volatile. Between 1979 and
1985, for instance, the jagged curve in Figure 11.1 represents considerable
Source: Board of Governors of the Federal Reserve System, Foreign Exchange Rates, various
issues; and Citibase electronic database.
exchange rate volatility, although the general trend over the period reflects
the appreciation of the dollar.
In the first part of this section, we examine factors that can explain these
longer-term trends in exchange rates. Then we look at why exchange rates
can deviate from these trends over short periods of time, during which
numerous rises and falls occur as shown in Figure 11.1.
Long-Run Exchange Rate Determination

In the long run, exchange rates are determined by economic fundamentals
like price levels and real incomes in different countries. Figure 11.2 shows
how the long-run exchange rate between yen and dollars is determined by
the forces of supply and demand. As is customary, the vertical axis measures
the “price” of dollars, while the horizontal axis measures the quantity of
U.S. dollars. Since we are focusing on the foreign exchange market between
yen and dollars, the price of dollars is the number of yen one must give up
to purchase each dollar.
The vertical axis mea¬ Figure 11.2

sures the price of dol¬ Determination of the Long-Run Yen/Dollar Exchange Rate
lars in the exchange
market. This price is
the yen/dollar
exchange rate, the
number of yen re¬
quired to purchase a
dollar. The demand
for dollars consists of
holders of yen depos¬
its who want to use
them to buy dollars.
The supply of dollars is
composed of those
who hold dollars.
Equilibrium in this fig¬
ure occurs when the
exchange rate is 100
yen per dollar.
The downward-sloping curve, D, is the demand for U.S. dollars by

Japanese residents and other holders of yen (including banks) who need
dollars as a medium of exchange to buy U.S. products and financial assets
such as stocks and bonds. When the exchange rate is 100 yen per dollar,
these holders of yen have a quantity demanded of U.S. dollars of $500
million. When the exchange rate falls to 50 yen per dollar (the dollar depre¬
ciates), it becomes cheaper to convert yen into dollars to buy U.S. goods
and assets, so the quantity demanded of dollars increases to $1 billion. This
is as we would expect given the law of demand: Other things equal, the
fewer yen required to purchase a dollar, the greater will be the quantity of
dollars demanded by Japanese and other holders of yen.1
The upward-sloping curve, S, represents the supply of U.S. dollars by
U.S. residents, banks, and other holders of dollars who wish to purchase
Japanese goods or assets. Since these items must be paid for with foreign
1 Advanced courses in international trade and finance typically introduce other factors (like import
and export demand elasticities) that can alter the slopes of the supply and demand curves. We
ignore these issues to focus on essential features of exchange markets.
338 Chapter 11 Foreign Exchange Markets
currency, banks trade dollars in the foreign exchange market for foreign
currency. When the exchange rate is 50 yen per dollar, the quantity supplied
of U.S. dollars is $300 million. When the exchange rate rises to 100 yen per
dollar (the dollar appreciates), the quantity supplied of dollars increases to
$500 million, reflecting the law of supply. We can think of Americans and
other holders of dollars as either demanding foreign currency or supplying
dollars, but the latter interpretation allows us to more easily see how the
exchange rate (in yen per dollar) is determined in the foreign exchange
market.
In the long run, the equilibrium exchange rate is determined by the
intersection of these demand and supply curves. In Figure 11.2, the equilib¬
rium exchange rate is 100 yen per dollar and the equilibrium quantity of
dollars bought and sold is 500 million. At the exchange rate of ¥100/$, the
quantity demanded of dollars exactly equals the quantity supplied, and every¬
one willing and able to buy or sell dollars at this exchange rate can do so.
If the exchange rate were below the equilibrium level, a shortage of U.S.
dollars would arise in the exchange market, whereas an exchange rate above
that level would create a surplus of dollars in the exchange market. Either
situation would put pressure on the exchange rate, moving it back to equi¬
librium.
Factors That Shift Long-Run Demand or Supply. Figure 11.1

shows that exchange rates often exhibit long-run trends. We now look at
how these long-run trends can be explained by changes in the supply of and
demand for currency brought on by changes in such factors as real income
or price levels.
Real Income. The level of real income in an economy ultimately reflects

the productivity of the country’s resources. Changes in real income at home
relative to that abroad will shift the demand or supply curves in the foreign
exchange market. For instance, the higher Japanese incomes are, the more
U.S. goods and assets Japanese residents will desire, and thus the greater the
amount of dollar deposits they will need to make these transactions. Hence
the demand for U.S. dollar deposits tends to increase when real income in
Japan rises and fall when it declines. Similarly, the supply of U.S. dollars
increases when real income in the United States rises. This is because more
Japanese goods and assets are desired due to higher real incomes in the
United States, and this leads to a greater number of dollars sold in exchange
for yen. The supply of U.S. dollars will decrease when U.S. incomes decline.
Since real incomes in Japan and the United States affect the demand for
and supply of dollars in the exchange market, a change in Japanese or U.S.
real incomes will affect the long-run exchange rate. Consider Figure 11.3,
where the initial equilibrium in the exchange market is at point A, where D°
and S'0 intersect. At the equilibrium exchange rate, e°, suppliers of dollars
receive e° yen for each dollar they sell in the exchange market. Let us see
what will happen if real incomes in the United States increase, but everything
else (including real incomes in Japan) remains unchanged. With higher real
incomes, people, businesses, and banks in the United States will want to
purchase more Japanese goods and assets than before, which will require
selling more dollars in exchange for yen. This increases the supply of dollars
from S° to S] in Figure 11.3. Thus, as a result of an increase in real income
in the United States, equilibrium moves from point A to point B, and the
exchange rate falls from e° to el. The dollar depreciates against the yen
because at point B, U.S. residents receive fewer yen for each dollar sold than
before the increase in real U.S. income. Notice too that the equilibrium
quantity of dollars sold in the foreign exchange market rises from F° to Fx.
Similarly, holding other things constant, an increase in Japanese real
incomes will lead to an appreciation of the dollar relative to the yen due to
the increased demand for dollars. Thus, changes in real income provide one
explanation for discernible long-run trends in exchange rates. In fact, the
depreciation of the dollar against the yen during the early 1990s that we saw
in Figure 11.1 was due in part to the U.S. recession, which resulted in a
decline in Americans’ real incomes.
Price Levels. Another important long-run determinant of exchange rates

is the price level at home compared to that abroad. When the U.S. price
level rises relative to that in Japan, U.S. goods and assets become more
expensive relative to Japanese goods, potentially reducing Japan’s need for
dollars to buy U.S. goods and assets. Furthermore, during periods of contin¬
ually rising U.S. prices, dollars become a poorer store of value compared to
yen because the purchasing power of dollars is eroded more rapidly, allowing
fewer goods to be purchased in the future for a given quantity of dollars.
For these reasons, a rise in the U.S. price level relative to that in Japan tends
to decrease the demand for dollars. Similarly, holders of dollars are more
eager to sell them in exchange for yen. Thus, an increase in the U.S. price
level relative to that in Japan will increase the supply of dollars.
Figure 11.4 shows how changes in relative price levels affect the long-
run exchange rate between yen and dollars. The initial equilibrium is at point
Ay where the exchange rate (e°) is determined by the intersection of the
demand and supply curves D° and 5°. If the U.S. price level increases relative
to that in Japan, both the demand for and supply of dollars will shift. In
particular, a rise in the U.S. price level increases the supply of dollars from
S° to Sl in Figure 11.4 because holders of dollars are more eager to convert
them into yen. The rise in the U.S. price level also makes holders of yen
less willing to buy dollars, shifting the demand for dollars to the left. If the
decrease in the demand for dollars is small relative to the increase in supply
(the shift from D° to Dl is small relative to the shift from S° to S1), the new
equilibrium will be at point B. In this case, the exchange rate falls from e°
to eB, and the equilibrium quantity of dollars traded rises from F° to FB.
However, if the decrease in demand is large relative to the increase in supply
(the shift from D° to D2 is large relative to the shift from 5° to S1), the new
equilibrium will be at point C. In this case, the exchange rate falls to an
even lower level, ec, and the quantity of dollars traded falls to Fc.
This analysis reveals a very important principle in foreign exchange
markets: All else equal, an increase in a country’s price level will lead to
long-run depreciation of the country’s currency. In Figure 11.4, for instance,
the rise in the U.S. price level leads to a decline in the exchange rate to
either eB or ec. This depreciation of the dollar is due to two consequences
of the rise in the U.S. price level: Although holders of dollars become more
willing to sell their dollars in exchange for yen, holders of yen are less
willing to use their yen to buy dollars. These two forces work together to
reduce the exchange rate when the U.S. price level increases. In fact, the
depreciation of the dollar against the yen during the last half of the 1970s
shown in Figure 11.1 was partly due to the unusually high growth in the
U.S. price level at the time.
Tariffs, Trade Barriers, and Preferences. Tariffs and other trade

barriers that restrict the amount of Japanese goods brought into the United
States reduce the ability of Americans to purchase Japanese goods and
The Marketfor Foreign Exchange 341
therefore reduce Americans’ desire to convert dollars into yen. Thus, restric¬
tions on imports from Japan decrease the supply of dollars, resulting in an
increase in the exchange rate. Similarly, limitations on the amount of Amer¬
ican goods that can be sold in Japan decrease the demand for dollars, re¬
sulting in a lower exchange rate. More generally, any change that reduces
Americans’ ability or desire to purchase Japanese goods leads to. a reduction
in the supply of dollars, while changes that reduce Japanese residents’ ability
or desire to buy U.S. goods decrease the demand for dollars.
The opposite is true of changes that enhance the willingness or ability
of domestic consumers to buy foreign products. For example, during the
1980s many Americans perceived Japanese goods as being of better quality
than goods made in the United States. These preferences led to increased
demand for Japanese goods, most notably cars and electronic products. In
turn, this increased demand for Japanese goods meant U.S. importers had to
convert dollar deposits into yen to pay for those goods, which led to an
increase in the supply of dollars in the foreign exchange market. This ulti¬
mately contributed to the depreciation of the dollar against the yen during
the last half of the 1980s.
Interest Rates. Foreign and domestic interest rates can also influence
long-run exchange rates through their impact on the demand for and supply
of currency. However, we will see shortly that if interest rate parity holds,
the impact of interest rates will primarily be on short-run changes in
exchange rates.
Why might interest rates affect the long-run exchange rate? Other things
equal, the higher the interest rate in the United States relative to that in
Japan, the more attractive holders of yen deposits find it to convert those
deposits into U.S. dollar deposits to earn higher interest. For this reason, a
rise in U.S. interest rates increases the demand for U.S. dollars. Similarly,
those who hold dollar deposits become less willing to convert them into yen
deposits as U.S. interest rates rise. Consequently, an increase in U.S. interest
rates leads to a decrease in the supply of dollars in the exchange market.
Since changes in interest rates affect the demand for and supply of a
currency, they affect the equilibrium exchange rate. To see this, suppose the
exchange market for dollars is in equilibrium at point A in Figure 11.5. The
exchange rate is e°, which corresponds to the intersection of the demand and
supply curves D° and S°. Two things happen if interest rates in the United
States rise relative to those in Japan while other things remain constant:
First, people holding dollar deposits are now less willing to sell them in
exchange for yen; second, people holding yen deposits are more willing to
use them to buy dollars.
Thus, the effect of an increase in interest rates in the United States is to
decrease the supply of dollars from S° to S1 and increase Japanese consum¬
ers’ demand for dollars from D° to D{ in Figure 11.5. As a result of these
two changes, the equilibrium exchange rate moves to point B, and the
exchange rate rises from e° to eB. (Since demand increased and supply
decreased, the new equilibrium quantity of dollars traded may rise, fall, or
remain the same.) At point A a dollar is worth only e° yen, but at point B it
is worth eB yen. The higher U.S. interest rates induce U.S. suppliers to sell
fewer dollars at any given exchange rate, which exerts upward pressure on
exchange rates. In addition, the higher U.S. interest rates induce Japanese
residents to attempt to buy more dollars, which exerts even more upward
pressure on the exchange rate. These two forces lead to a higher exchange
rate and a more valuable dollar.
Purchasing Power Parity. Our theory of exchange rates suggests

that many factors contribute to the determination of exchange rates in the
long run. A simpler theory, called purchasing power parity, asserts that in
the long run the exchange rate is determined solely by price levels in different
countries. We conclude our analysis of long-run exchange rate determination
by examining this theory and its limitations.
The idea of purchasing power parity (PPP) is derived from a consider¬
ation of the law of one price, which states simply that two identical goods
within a relevant market must sell for the same price. The law of one price
is based on the idea that violations of that law will be quickly corrected by
buyers purchasing only the good with the lower price.
In international trade, the law of one price can be applied to two identical
goods—say, a sweater made in the United States and an identical sweater
made in France. For these two sweaters to be sold in the United States, they
must sell for the same price, or consumers will buy only the cheaper one.
Suppose the sweater costs $30 in the United States; then the sweater from
France must also sell for $30 worth of francs. If the exchange rate is 5 French
francs per dollar, the price of the sweater in French francs must be 150
francs.
The law of one price by itself applies to a set of identical goods, such
as our two sweaters. But the idea has led to the notion of purchasing power
parity, which is a relationship not between the prices of an identical good in
the United States and in another country but between the price level in the
United States and the price level in another country. Purchasing power
parity states that the price level in the United States (Pus) times the exchange
rate, e (expressed as foreign currency per dollar), equals the price level in a
foreign country (PF):
Pus X e = PF.
When purchasing power parity holds, the dollar has the same purchasing
power in the United States and in any other country. By converting dollars
into foreign currency at the market exchange rate, you would be able to buy
exactly the same amount of foreign goods with your dollars that you would
in the United States.
Purchasing power parity also implies that a foreign currency will depre¬
ciate if the country’s price level rises relative to the foreign price level and
appreciate if the foreign price level rises relative to the country’s own price
level. For instance, suppose the price levels in the United States and abroad
are both 100. If purchasing power parity holds, the exchange rate must be
unity in order for the purchasing power parity equation to hold. If the U.S.
price level rises to 200 while the foreign price level remains unchanged, the
dollar exchange rate must depreciate to .5 to balance the equation. More
generally, purchasing power parity implies that, other things equal, a 10
percent increase in the U.S. price level will lead to a 10 percent depreciation
of the dollar.
Does purchasing power parity hold? The answer is not exactly, especially
in the short run (see, for instance, Box 11.3). One reason it fails to hold
exactly is the presence of substantial transactions costs in the international
trade of goods and services and the existence of nontraded goods—goods
like real estate that are not traded internationally and for which there is
therefore no tendency for market forces to eliminate international price
differences. Another reason stems from the fact that many foreign and do¬
mestic goods are not identical; there is no reason to expect a Mercedes to
sell for the same price as a Chevy. Finally, we will see shortly that short-
run exchange rates are determined in great part by the expectations of traders
in the foreign exchange market. Largely for these reasons, the historical
evidence does not support the notion that a 10 percent rise in the U.S. price
level relative to foreign price levels will result in a 10 percent depreciation
of the dollar. However, the theory of purchasing power parity is nonetheless
useful because it often accurately predicts the long-run direction of changes
in exchange rates.
Exchange Rates in the Short Run

As we noted earlier, in the short run (seconds, minutes, hours, or days)
exchange rates tend to fluctuate wildly. Since the long-run determinants of
exchange rates discussed earlier (like price levels) do not change minute by
Purchasing Power Parity; the Big Mac

Standard, and The Economist Standard
Economists frequently calculate purchasing The accompanying table presents what we

power equivalent exchange rates—the exchange call "The Economist Standard." Similar to the
rates that would equalize purchasing power Big Mac Standard, it contains the exchange
across countries. International firms often use rates that would allow you to convert the U.S.
these rates to determine the supplemental sala¬ price of an issue of The Economist into a foreign
ries employees living in high-cost countries re¬ currency and purchase the magazine at the for¬
quire to be able to attain the U.S. standard of eign currency price published on the front cover.
living while living abroad. For example, in January 1993, the price of The
One of the more interesting measures is Economist in Canadian dollars was C$4.25, and
what the newsmagazine The Economist pub¬ the price in German marks was DM7.00. The
lishes under the name "Big Mac Standard." This U.S. dollar price was US$3.50. These prices are
measures what the exchange rates would have summarized in the second column of the table.
to be for the local currency price of a The third column presents the implied pur¬
McDonald's Big Mac hamburger to be the same chasing power parity exchange rate on this date
across countries. If the actual exchange rate among the different currencies. For example, in
equals the Big Mac Standard rate, purchasing 1993, C$4.25 would buy an issue of The Econo¬
power parity holds; you can convert the dollar mist, as would US$3.50, so The Economist con¬
price of a Big Mac in the United States into, say, sidered C$4.25 to be equivalent to US$3.50.
yen and have enough yen to buy a Big Mac in Thus, the implied exchange rate between U.S.
Japan. Historically, however, actual exchange dollars and Canadian dollars that makes the lo¬
rates do not allow purchasing power parity to cal currency prices of this magazine the same is
hold for Big Macs. Continued on p. 346
Price of
The Economist
in Local Implied PPP Market Change in Market
Currency, Exchange Rate Exchange Rate, Exchange Rate Exchange Rate,
January 23, for One February 9, Required for March 1,
Country 1993 U.S. Dollar 1993 PPP to Hold 1994
Canada C$4.25 1.2143 1.2676 — 1.3508

France FFr23 6.5714 5.5975 + 5.8170
Germany DM7.00 2 1.6527 + 1.7080
Japan ¥850 242.86 121.25 + 104.55
U.K. £1.80 .5143 .6985 —
.6713
U.S. $3.50 —
Continued from p. 345 power parity rates. The fifth column indicates
C$4.25/US$3.50 = 1.2143 Canadian dollars per whether market exchange rates must increase or
U.S. dollar. Similarly, the implied exchange rate decrease relative to those in February 1993 for
with the Japanese yen is ¥850/$3.50 = 242.86 purchasing power parity to hold.
yen per U.S. dollar. The final column shows the market
The fourth column lists the spot exchange exchange rates on March 1, 1994. Between
rates on February 9, 1993. These exchange rates 1993 and 1994, market exchange rates for
do not match up well with the implied rates French francs, German marks, and British
based on purchasing power parity in column 3. pounds moved in the direction required by PPP,
This is consistent with the view that purchasing but this was not the case for Canadian dollars or
power parity tends not to hold in the short run. Japanese yen. This illustrates that PPP is not
But in the long run, the idea of purchasing aways a good predictor of exchange rate
power parity is that the market exchange rates movements over short time periods like one
will move in the direction of the purchasing year.
minute, they do not explain short-term fluctuations in exchange rates. What

causes these short-run movements?
Since today’s foreign exchange markets are linked by computers, banks
and other dealers of foreign currency, as well as professional traders, can
very quickly convert dollar deposits into deposits of foreign currency. In
fact, these traders make their living by buying dollars at a low exchange rate
one minute and selling them at a slightly higher rate the next. As we will
see, this process often results in minute-by-minute changes in exchange rates
due to changes in traders’ expectations about future exchange rates.
Interest Rate Parity. To currency traders and banks, foreign bank

deposits are very close substitutes for dollar deposits because they can easily
be converted from one currency into another via the foreign exchange mar¬
ket. For this reason, currency traders and banks continually monitor move¬
ments in interest rates and exchange rates across countries to determine the
most attractive form of deposits to hold. For instance, suppose a bank has
$1 million in U.S. deposits that pay interest of ius. If the current exchange
rate is 100 yen per dollar, the $1 million can alternatively be converted into
a 100-million-yen deposit that earns the foreign interest rate of iF. Should
the bank keep the funds as dollar deposits or convert them into yen deposits?
As we discussed in the section on the determination of long-term
exchange rates, the higher the rate in Japan, the more attractive it is to hold
Japanese deposits. Flowever, another factor enters the picture in the short
run because the rate of return of holding foreign deposits includes both the
interest earned on those deposits and the expected appreciation of U.S.
currency. If the dollar appreciates against the yen over the next few days,
money will be lost when the yen are converted back into dollars. We can
see this by letting £er+1 denote the trader’s expectation of the future
exchange rate and letting et represent the exchange rate right now. The
expected rate of return of holding foreign deposits is thus
„ . (Eet+1 - et)
l\pr — If
U
Notice that (Eet+X — et)let is the expected appreciation of the dollar (in
percentage terms) and must be subtracted from the foreign interest rate to
account for the fact that an appreciating dollar means part of the interest
earned on foreign deposits will be lost when the deposits are converted back
into dollars. For instance, if the foreign interest rate is 5 percent and traders
expect the dollar to appreciate 3 percent against the yen (that is, expect the
exchange rate to increase from 100 yen to 103 yen per dollar), the expected
rate of return of holding Japanese deposits is only 2 percent. While 5 percent
interest is earned on the foreign account, 3 percent will be lost when the yen
used to open the foreign account are converted back into dollars.
In deciding whether to hold dollar or yen deposits, traders continually
compare the U.S. interest rate to the expected return they can earn on foreign
deposits. If ius exceeds, RF, they expect to earn more by holding dollar
deposits. Consequently these traders begin converting yen into dollars, which
puts pressure on the exchange rate to rise. Conversely, if ius is less than the
expected return on foreign deposits, RF, traders convert dollars into yen,
which tends to drive down the exchange rate. The point is that traders
constantly evaluate these returns and buy and sell currencies by the minute
to capitalize on short-term profit opportunities. These short-term buy-and-
sell decisions lead to short-term changes in the exchange rate.
Expressed differently, traders in currency markets continually look for
profit opportunities that arise from short-term imbalances between the ex¬
pected returns to holding U.S. versus foreign deposits. They switch deposits
across countries to maximize their expected returns, and this practice ulti¬
mately leads to the returns being equal:
(Eet+[ - et)
lUS ~ lF
et
This relation is called interest rate parity, because it means the interest rate
on dollar deposits equals the expected return on foreign deposits. When
interest rate parity holds, traders cannot profit by switching currency hold¬
ings, and this effectively determines the short-term exchange rate for the
moment, et.
Figure 11.6 graphically illustrates interest rate parity. The horizontal axis
measures the expected rate of return (in dollar terms), and the vertical axis
measures the current exchange rate, et, expressed as yen per dollar. The line
labeled RF is the expected rate of return to holding foreign deposits, and the
vertical line labeled ius is the U.S. interest rate (the rate of return to holding
Figure 11.6
Interest Rate Parity and Determination of the Short-Term Exchange Rate
When the spot exchange rate is ejr, the return to holding dollar deposits (ius) exceeds the
return to holding foreign deposits (RF) at point A. Currency traders quickly convert yen
deposits into dollar deposits, which tends to drive up the spot exchange rate to e*. Con¬
versely, when the spot rate is ef, the return to holding foreign deposits at point B exceeds
that to holding dollar deposits. This induces currency traders to shift out of dollars and into
yen, which drives the spot rate down to e*. At point C interest rate parity holds, meaning
the market is in short-run equilibrium because traders cannot make profits by converting
one currency into another.
U.S. dollar deposits). At point A, the expected rate of return on dollar deposits
exceeds that on foreign deposits (ius > RF), which means profits can be
made by converting foreign deposits into dollar deposits. As traders do so,
the exchange rate rises. Conversely, at point B the expected return is higher
on foreign deposits. Consequently traders convert dollars into foreign de¬
posits, which puts pressure on the exchange rate to fall. Only at point C,
where the current (spot) exchange rate is ef, are the expected returns on
foreign and dollar deposits equal. At this point, traders cannot buy or sell
dollar deposits to earn additional profits because interest rate parity holds;
the market is in short run equilibrium.
Expectations of Future Exchange Rates. We have seen that

banks and other currency traders continually seek out profit opportunities.
Notice that these profit opportunities depend critically on the return to hold¬
ing foreign deposits, which is
n . (Eet+1 - et)
Rp — ip
et
This return depends not only on the current exchange rate (et) and the interest
rate on foreign deposits (ip) but also on traders’ expectations of the future
exchange rate (Eet+X). For this reason, traders’ decisions about whether to
buy or sell foreign deposits today depends in part on what they expect
exchange rates to be in the future. Traders continually update these expec¬
tations based on new information, including minute-by-minute wire reports
and rumors on the street.
What happens if new information surfaces that leads currency traders to
increase their expectations of the future exchange rate? In this case, the
expected return to holding foreign currency falls, inducing them to quickly
sell yen and buy dollars. This puts upward pressure on the current exchange
rate, and short-term appreciation of the dollar occurs. Graphically, an in¬
crease in the expected future exchange rate shifts the RF curve in Figure
11.7 to the left, resulting in a higher exchange rate at which interest rate
ft** <*8 **8

The exchange market Figure 11.7
is initially in equilib¬ Impact of an Increase in the Expected Future Exchange
rium at point A, Rate on the Short-Term Exchange Rate
where the expected
return on foreign de¬
posits equals that on
U.S. deposits. An in¬
crease in the expected
future exchange rate
decreases the ex¬
pected return on for¬
eign deposits, shifting
the Rf curve to the
left. As you can see,
this causes the spot
exchange rate to rise
to e] and creates a
new short-run equilib¬
rium at point B.
parity holds. Conversely, if expectations of the future exchange rate fall, the
Rf curve shifts rightward, and short-term depreciation of the dollar occurs.
You might think traders could profit by using our long-run supply and
demand framework to aid in their forecasts of future exchange rates. For
instance, suppose a report comes over the news wire that leads traders to
expect a higher U.S. price level relative to that in Japan. As we saw earlier
in this chapter, a higher price level, if it does materialize, will in the long
run lead to the depreciation of the dollar. Does this mean currency traders
could profit by immediately converting their dollar deposits into yen? The
answer is no. The problem is that in the short run, uncertainty exists about
not only whether the higher price level will materialize but, if so, just how
high it will be and when its effect on exchange rates will be felt. Add to this
the likelihood that other factors might change as well, which could offset
the higher U.S. price level. Different traders are likely to form different
expectations about all of these things. In addition, most currency traders
have very short-term horizons and cannot afford to hold positions in foreign
currency for two or three years to capitalize on a long-run devaluation of
the dollar.
Consequently, in real life currency traders continually buy and sell cur¬
rency based on their best guesses about how the latest news report or event
will affect exchange rates over the next few days or even minutes. During
the war in the Persian Gulf, for instance, exchange rates fluctuated wildly as
traders bought and sold currency over very short spans of time based on
their own expectations of how long the war would last and how it would
affect exchange rate movements. Of course, each day the Department of
Defense released new information, and each day these traders changed their
buy-and-sell decisions based on their latest expectations.
Thus, short-term movements in exchange rates are largely unpredictable
and arise primarily due to minute-by-minute changes in trader expectations
that occur as new information becomes available. In the next chapter, we
will take a deeper look at why the short-term returns on financial assets are
so difficult to predict and discuss different views of how market participants
form their expectations. We will use all of these tools extensively in Chapter
17, where we examine the macroeconomic ramifications of changes in
exchange rates on such factors as the balance of payments.
Summary (1) Assuming other things remain constant, determine the likely long-term
Exercise 11.2 impact of the following on the exchange rate between the U.S. dollar and
the British pound, (a) Due to a long recession in both the United States and
Britain, real incomes in both countries fall, (b) Interest rates in the United
States fall by 2 percent. (2) Suppose the interest rates in the United States
and Japan are both 6 percent, but the dollar is expected to depreciate by 2
percent relative to the yen. Does interest rate parity hold? In the short run,
what would you expect to happen to the exchange rate?
Answer: (1) (a) Suppose the initial equilibrium in the exchange market
between pounds and dollars is at point A in part a. Due to a decline in real
income in the United States, the supply of dollars decreases from S° to S'1.
The decline in real income in Britain reduces the demand for dollars from
D° to either D] or D2. Depending on the relative magnitude of the shifts, the
exchange rate will either rise, fall, or remain the same. If the demand for
dollars declines by less than the supply, the new equilibrium is at point B.
In this case, the exchange rate rises from e° to eB (the dollar appreciates
against the pound). But if demand decreases by more than supply, the new
equilibrium is at point C. In this case, the exchange rate falls from e° to ec
(the dollar depreciates against the pound).
(b) Suppose the initial equilibrium is at point A in part b. If interest rates
in the United States decline, holders of dollar deposits find pound deposits
relatively more attractive. This increases the supply of dollars from S° to S\
The decrease in U.S. interest rates, however, makes those holding pound
deposits less willing to convert them into dollar deposits. This leads to a
decrease in the demand for dollars from D° to D1. These two forces exert
downward pressure on the exchange rate, and as a result the new equilibrium
is at a point such as B. The exchange rate falls from e° to e] (the dollar
depreciates against the pound).
(2) The expected rate of return (in dollars) of yen deposits is RF = 6%
— ( — 2%) = 8%, which exceeds that on U.S. deposits (6 percent). Since
these returns are not equal, interest rate parity does not hold. In the short
run, currency traders will attempt to profit by converting dollar deposits into
yen deposits, leading to a short-term depreciation of the dollar against the
yen.
£/$ £/$
Spot and Forward Exchange Rates
Our analysis of the short-term determination of exchange rates reveals that

they are often driven by changes in expectations and, furthermore, that they
can fluctuate considerably during a given day. This volatility in short-term

exchange rates exposes users of foreign exchange markets to risk. In this
section, we show how participants in the exchange market can reduce their
exposure to risk by using the forward market and how doing so opens up
arbitrage possibilities for traders seeking short-term profits in foreign
exchange markets.
The spot rate is the current exchange rate, the rate at which you can
exchange a foreign currency for U.S. dollars at this instant. The spot rate for
foreign exchange is akin to getting a price quote on a share of Houghton
Mifflin stock from your stockbroker; it is the price at which you can buy the
asset in question right now. The forward rate is different. You can sign a
forward contract today at an exchange rate and quantity of currency con¬
tracted for today. But the actual exchange will take place at a specific future
date at a rate called the forward exchange rate, because the currency will
actually be exchanged at some forward (future) date.
The forward exchange rates presented in Table 11.1 reveal they do not
always move in concert with the spot rate. For example, the spot exchange
rate for Canadian dollars is C$1.3508 for US$1, while the 180-day forward
rate is C$1.3524 for US$1. It takes more Canadian dollars to purchase a
given amount of U.S. dollars in a forward contract than on the spot market.
Similar to our analysis of yield curves in the previous chapter, this means
traders in the forward market expect the U.S. dollar to appreciate relative to
the Canadian dollar over the next 180 days. In contrast, Table 11.1 reveals
the spot rate for yen is greater than the 180-day forward rate. Since the spot
rate of 104.55 yen per dollar exceeds the 180-day forward rate of 103.67
yen per dollar, the U.S. dollar is expected to depreciate relative to the yen
over the next 180 days.
Participants in the Forward Market

Who would ever want to buy or sell currency in the forward market? The
answer is hedgers, speculators, and arbitrageurs, each of whom we discuss
next.
Hedgers go to the forward market to buy or sell currency now

to protect themselves against (hedge) fluctuations in exchange rates. Con¬
sider a grain dealer with an offer from Japan for a large quantity of soybeans.
The Japanese insist on contracting in yen, and the contract offers the grain
dealer 1 billion yen for a specified amount of soybeans. The grain dealer
looks at the spot exchange rate of 104.55 yen per dollar and calculates that
the Japanese are offering the equivalent of $9,564,801.50. The problem is
that the Japanese will not pay until they receive the soybeans, half a year
from now. The grain dealer wants to accept the deal, on which her costs are
$8 million. If the exchange rate doesn’t change, she will earn a profit of
about 20 percent. However, she worries that the exchange rate could change
Table 11.1
Spot and Forward Exchange Rates
This table presents spot and forward exchange rates for six countries on March 1, 1994, as
well as the market prediction of the future change in the exchange rate. For instance, on
March 1 you could sell 1 U.S. dollar for 1.3508 Canadian dollars in the spot market. Alter¬
natively, on this date you could sell 1 U.S. dollar on March 1 and in return receive 1.3524
Canadian dollars in 180 days. Since the forward rate is greater than the spot rate, the U.S.
dollar is expected to appreciate relative to the Canadian dollar over the next 180 days. The
+ in the final column represents the market's prediction of this appreciation. Notice the -
in the corresponding column for Japan, which indicates the market expects the dollar to
depreciate relative to the yen over the next 180 days.
Forward Market
Market Spot Market 180-Day Prediction for
Exchange Rate, Forward Exchange the Change in
Country March 1, 1994 Rate, March 1, 1994 the Exchange Rate
Canada 1.3508 1.3524 +

France 5.8170 5.8795 +
Germany 1.7080 1.7234 +
Japan 104.55 103.67 —
Switzerland 1.4364 1.4374 +

U.K. .6713 .6751 +
Source: The Wall Street Journal, March 2, 1994, p. Cl6.
drastically in half a year and destroy her profit margin. For example, if the
exchange rate rises to 150 yen per dollar, her 1 billion yen will be worth
only $6,666,666.67, and her tidy profit will turn into a loss.
What is the solution to this grain dealer’s worries? She can hedge her
position by contracting in the forward market for foreign exchange. She signs
a contract now to exchange 1 billion yen for dollars in 180 days at the current
180-day forward exchange rate of 103.67 yen per dollar. In this way she locks
in the sale and the price, and in 180 days she shows up with 1 billion yen
and trades it for $9,645,992.10. This trade will be honored regardless of what
the spot exchange rate happens to be in 180 days. Of course, our trader may
end up kicking herself if the exchange rate falls or thanking her lucky stars
if it rises, but the key point is that she can hedge her position and essentially
eliminate the exchange rate risk in her transaction.
Speculators. Speculators also buy and sell in the forward market, not
to hedge a position but to intentionally take a risky (or open) position.
Speculators bet on the direction of the market. To be concrete, suppose the
current spot exchange rate is C$1.3508 per U.S. dollar, and the speculator
thinks the spot rate in 180 days will be C$1 for one U.S. dollar. Since the
180-day forward rate of C$1.3524 per U.S. dollar exceeds the market spot
rate, the market thinks the U.S. dollar will appreciate relative to the Canadian
dollar. The speculator can bet against the market by buying Canadian dollars
on the forward market. If the market is wrong and the speculator right (i.e.
the dollar depreciates), the speculator will be able to sell his Canadian dollars
on the spot market in 180 days and make a handsome profit. For instance,
if the speculator uses the forward market to buy C$1,000,000 in 180 days
for C$1.3524 per U.S. dollar (at a total cost of US$739,426.20), if the spot
rate falls to C$1 per U.S. dollar in 180 days, he can turn right around and
sell his C$1,000,000 for US$1,000,000 in the spot market—a profit of about
35 percent. Of course, if the speculator is wrong and the market is right, he
may lose money. Speculators make money by taking risky positions in the
foreign exchange market.
There are other ways to speculate in the foreign exchange market, in¬
cluding going to the futures market instead of the forward market. We won’t
go into the details of such transactions here. We merely point out that
speculators play an important role in these markets too and that these markets
work similar to the futures markets we looked at in Chapter 9.
Arbitrageurs. A third group active in the forward market is arbitra¬

geurs. Arbitrageurs are buyers and sellers who try to take advantage of
temporary profit opportunities created by discrepancies among prices in
various markets. They do not speculate; arbitrage possibilities are based not
on outguessing the market but on taking advantage of pricing discrepancies.
For example, if you notice that this textbook costs $65 at your bookstore
but only $45 at a bookstore at another university, you have something like
an arbitrage possibility. You could buy this book for $45 and sell it for
almost $65 to a classmate. This might be risky, however, since you have
only a small number of classmates and it might take considerable time to
sell such a book. In contrast, the foreign exchange market involves millions
of dollars in transactions virtually every minute. This allows arbitrage to
occur very quickly in the exchange market, and at very little risk.
Triangular Arbitrage. One type of arbitrage in exchange markets is

triangular arbitrage. Triangular arbitrage takes advantage of discrepancies
between the spot exchange rate between two currencies and the exchange
rate obtainable by first trading one of the two currencies for a third currency
and then trading the third currency for the other of the two currencies. For
example, suppose the spot rate between U.S. dollars and Canadian dollars
is 1.30 Canadian dollars per U.S. dollar and the spot rate between U.S.
dollars and British pounds is 1.50 U.S. dollars per 1 pound. One way to
obtain British pounds from Canadian dollars is indirectly through purchases
of U.S. dollars; that is, 1.95 Canadian dollars will purchase 1.50 U.S. dollars,
which in turn will purchase 1 pound. If the exchange rate between Canadian
dollars and British pounds differs from this rate, there is an arbitrage pos¬
sibility—a riskless way to make profits.
For example, if the Canadian dollar were trading for 1.90 Canadian
dollars per pound, an arbitrageur could use Canadian dollars to buy pounds
and then trade the pounds for U.S. dollars and the U.S. dollars for Canadian
dollars. For every 1.90 Canadian dollars, she would buy 1 pound, then buy
1.50 U.S. dollars, and then buy 1.95 Canadian dollars, for a profit of .05
Canadian dollars on every 1.90 Canadian dollars that she had initially. Since
these transactions can be executed very quickly, this return can be com¬
pounded rapidly. If exchange rates were ever this far out of line, arbitrageurs
would instantly trade millions of dollars using the above method, earning
huge returns until the exchange rates adjusted to close off this arbitrage
possibility.
Covered Interest Arbitrage. Covered interest arbitrage is an ar¬

bitrage possibility that occurs when two countries have interest rates, a spot
exchange rate, and a forward exchange rate that allow an arbitrageur to
profit. Consider an investor with US$10,000. This investor is considering
buying U.S. Treasury bills with these funds, and six-month Treasury bills
are paying 5.9 percent interest. However, he is also looking into Canadian
Treasury bills, which he reasons are just about as safe as U.S. T-bills and
are paying 6 percent interest. His friends, of course, counsel him to take the
6 percent interest in the Canadian bills, but he is not one to quickly jump at
things. Instead, he reasons that $10,000 in the U.S. Treasury bills, at 5.9
percent interest for half a year, will earn him $10,000 X 1.0295 = $10,295
after six months.
What about the Canadian investment? The investor can use the
US$10,000 to buy Canadian dollars at the going exchange rate of C$1.3508
per U.S. dollar. His US$10,000 will give him US$10,000 X C$1.3508/US$
= C$13,508. He invests this at 6 percent interest for six months and earns
C$13,508 X (1.03) = C$13,913.24. He is also worried about exchange rate
risk, since six months is a long time away, so he decides to eliminate this
risk by using the forward market. He knows that he will have these Canadian
dollars in 180 days, so he also contracts forward to exchange this amount
of Canadian dollars for U.S. dollars at the forward rate of C$1.3524 for one
U.S. dollar. This process—covered interest arbitrage—gives him
C$ 13,913.24/(C$ 1.3524/US$ 1) = US$10,287.81 in six months.
What does all this mean? The investment in Canada, even at a higher
interest rate, pays a lower return after taking into account the forward
exchange rate that would convert interest paid in Canadian dollars back into
U.S. dollars in 180 days. Since these are equally safe investments, our
investor earns more by investing in U.S. Treasury bills. In fact, so will all
other investors in both the United States and Canada. They will flock to U.S.
356 ChapterII Foreign Exchange Markets
Treasury bills, bidding up the price and lowering the yield, while at the same
time the paucity of investors in Canadian Treasury bills will lower their
price and raise their yield, until the return from these two alternative invest¬
ments of equal risk are equalized. Covered interest arbitrage is the act of
choosing between these two equally risky investments according to which
pays the higher return, and the continual presence of covered interest arbi¬
trageurs means departures from equality in these two alternative investments
are very brief indeed, lasting only minutes at a time.
If we let ius and ic denote the risk-free interest rates in the United States
and Canada and espot and forward denote the spot and forward exchange rates
(in foreign currency units per dollar), respectively, we can state the covered
interest arbitrage condition as
1 , • _ v ^ + ic)
^ l US ^spot ^
^forward
The left-hand side is the total return from investing in the U.S. Treasury bill.
The right-hand side is the total return from investing in the Canadian Treas¬
ury bill, which includes the interest rate in Canada times the spot rate divided
by the forward rate. The spot rate divided by the forward rate is called the
foreign exchange term premium and indicates how the forward market per¬
ceives the exchange rates will move. When this equation holds with equality,
investors cannot profit from covered interest arbitrage. When the equality
does not hold, covered interest arbitrage will lead to profits. If you plug the
interest rates and exchange rates used in our earlier example, you will find
this equality does not hold. In this example, Canadians can earn profits by
investing in the United States via covered interest arbitrage, while Americans
simply invest in the United States.
Covered interest arbitrage links the spot and forward exchange rates and
the relative interest rates on equally risky investments between two countries.
However, covered interest arbitrage does not determine the exchange rate,
nor does it tell us how the spot rate will react to a change in the U.S. or
Canadian interest rate. We know that if the U.S. interest rate rises, the
covered interest arbitrage condition will hold after a very brief adjustment,
but we don’t know if it will hold because the spot rate will rise, the forward
rate will fall, the Canadian interest rate will rise, or some or all of these
events will occur.
Summary The yield to maturity on a one-year U.S. Treasury bill is 4.5 percent, and
Exercise 11.3 that on an equally safe one-year Canadian Treasury bill is 7 percent. The
spot exchange rate is 1.25 Canadian dollars per U.S. dollar, and the one year
forward rate is 1.28 Canadian dollars per U.S. dollar. Which Treasury bill
should you buy to maximize your return?
Answer: The U.S. Treasury bill yields 4.5 percent, or a return of 1.045.
In Canada, you can invest with equal safety by first converting your funds
into Canadian dollars at the spot exchange rate of C$1.25/US$1. Then you
invest in the Canadian Treasury bill and earn 7 percent, and you contract in
the forward market today to convert your funds back into U.S. dollars at the
forward exchange rate of C$1.28/US$1. Your return on the Canadian bill is
calculated as (C$1.25/US$1) X (1.07)/(C$1.28/US$1) = 1.0449. Thus, you
are almost indifferent between the two investment options, although the U.S.
T-bill pays an extra 1/100 of 1 percent.
Fixed and Flexible Exchange Rate Systems
Exchange rates can be flexible—that is, set by forces of supply and demand
in the exchange market as discussed earlier in this chapter—or they can be
controlled by central bank intervention. We now compare these systems and
look at how the Fed can actually intervene in a flexible exchange rate system
to influence the exchange rate.
Flexible Exchange Rates

When the exchange rate is determined strictly by the intersection of supply
and demand, with no central bank intervention, we say we have flexible (or
floating) exchange rates. This indicates the exchange rate is free to rise or
fall as dictated by supply and demand in the market for foreign exchange,
as we have thus far assumed in this chapter. Of course, exchange rates don’t
have to be flexible, and throughout much of history the world has had fixed
exchange rates. We discuss this history a bit later in this chapter, but to put
it into context, we first explain methods authorities like the Fed use to
intervene in exchange markets.
Central Bank Intervention

Central bank intervention can occur in any foreign exchange market, even
in a situation of flexible exchange rates. This situation, called a dirty float,
occurs when the monetary authority (e.g., the Fed) buys or sells currency to
influence the market-determined exchange rate. For instance, today the U.S.
dollar exchange rate against the British pound is basically determined by the
laws of supply and demand, but the Fed occasionally intervenes to influence
the rate.
To see how central bank intervention works, consider a situation where
the pound/dollar exchange rate is high by historical standards. The high
exchange rate makes it difficult for U.S. exporters specializing in shipments
to England to sell products there. This might lead to political pressure that
ultimately persuades the Fed to intervene in the foreign exchange market.
What can the Fed do to help U.S. exporters? It can increase the supply
of dollars by purchasing pounds. This adds to the total market supply of
dollars and tends to lower the exchange rate. Figure 11.8 illustrates this
situation. The initial equilibrium is at the intersection of the supply of and
demand for dollars, S° and D°, and the equilibrium exchange rate is e°. The
Fed seeks to lower this exchange rate by supplying additional dollars, which
shifts the supply curve to Sl. This lowers the equilibrium exchange rate to
e\ the desired result (at least from the point of view of U.S. exporters). U.S.
consumers who have a taste for British products, on the other hand, will be
harmed by the Fed’s actions, since they will now have to pay higher prices
for British goods due to the depreciation of the dollar.
Notice that intervention to lower the exchange rate requires the Federal
Reserve System to create dollar deposits, which it supplies to the foreign
exchange market in return for pounds. Thus, the intervention means the Fed
accumulates foreign reserves in the form of British pounds, while also in¬
creasing the supply of dollars in circulation. You might be tempted to con¬
clude that these dollars, since they end up in the hands of non-Americans,
The initial supply of Figure 11.8

dollars is 5°, and de¬ Fed Intervention to Lower the Exchange Rate
mand is D°. The equi¬
librium exchange rate
is e° in the absence of
government interven¬
tion. If the Fed wants
to intervene to lower
the exchange rate, it
can do so by supplying
additional U.S. dollars
in the foreign
exchange market, in¬
creasing the market
supply from 5° to S\
By doing so, it reduces
the equilibrium
exchange rate to e1.
Notice that with
exchange rates quoted
in units of foreign cur¬
rency per dollar, this
reduction in the
exchange rate reflects
a depreciation of the
dollar.
will not change the U.S. money supply. That is incorrect, however, because
non-U.S. consumers generally purchase dollars not to hoard but to spend on
U.S. goods or assets, and these dollars, once spent, are in circulation in the
United States just like any other dollars. Thus, intervention in the foreign
exchange market will tend to increase the U.S. money supply. As we learned
in Chapter 3, this can lead to inflation in the United States. In Chapter 17,
we look at this situation in more detail.
Would the Fed ever intervene if the exchange rate were too low? A low
exchange rate also makes U.S. assets inexpensive relative to British assets,
and hence there may be a large increase in purchases of U.S. assets by U.K.
citizens. This can have severe political repercussions. Remember the concern
a few years ago that the Japanese were “buying up America”? To remedy
this fear, the Fed might intervene to raise the exchange rate, making U.S.
assets more expensive for non-Americans. To do so, the Fed would need to
decrease the supply of dollars or increase the demand for dollars. But the
Fed cannot directly reduce the number of dollars brought to the foreign
exchange market, even though it can reduce the U.S. money supply. Thus,
the Fed cannot easily shift the supply curve for dollars leftward. But it can
add to the demand for dollars by also offering to sell British pounds in
exchange for dollars. By doing so, the Fed in essence increases the demand
for dollars to Dl in Figure 11.9. This additional demand will raise the
exchange rate from e° to el.
Of course, the Fed can act in this way only if it actually has a way to
pay for the dollars with British pounds; that is, the Fed must have a stash of
pounds, or so-called reserves of foreign exchange, before it can intervene in
this manner. Also, unlike the situation in which the Fed was supplying dollars
(something it can do in unlimited quantities if it so desires), the Fed cannot
keep supplying pounds forever; it has only finite supplies of foreign curren¬
cies that it can use to intervene in the exchange markets. Thus, trying to
raise the pound-per-dollar exchange rate is a different problem than trying
to lower it.
Fixed Exchange Rates

How, then, do fixed exchange rates work? The answer depends partly on the
exact system of fixed exchange rates adopted. For example, a nation can
unilaterally decide to fix the exchange rate between its currency and another
currency. Some nations fix the exchange rate between their currencies and
the U.S. dollar, even though the dollar itself floats against other currencies.
In this situation, the unilateral action requires that the country fixing the
exchange rate buy or sell its currency or its foreign reserves to keep the
exchange rate fixed. Alternatively, several nations may band together to set
up a fixed exchange rate regime, in which the countries take on a mutual
responsibility to maintain fixed rates. In this situation, there will be specific
rules or understandings concerning each country’s responsibilities in terms

of interventions.
In all cases, fixed exchange rates result when a country or group of
countries pledge to keep the exchange rate within a narrow band of a target
level. To honor this pledge, these nations set up an upper and a lower limit
on the exchange rate, then offer to buy or sell all the currency demanded or
supplied at those limits. Figure 11.10 illustrates this system. There is an
upper bound, ev, a lower bound, e0 and somewhere in between a target
level, eT These bounds are usually fairly tight, such as plus or minus 1 to 3
percent of the target. When the exchange rate gets to the upper bound, ev,
the Fed and/or the foreign central bank agrees to supply more dollars to keep
the exchange rate within the band. When the exchange rate gets to the lower
bound, eL, the foreign central bank agrees to demand dollars to keep the
exchange rate within the band. A reasonable division of responsibilities
might be for the Fed to supply dollars when the exchange rate gets near the
upper band, because this is relatively easy for it to do, and the foreign central
bank to supply its currency when the exchange rate gets near the lower band,
because it is relatively easy for it to do so. The alternatives, such as the
Figure 11.10
Currency Bands and a Fixed Exchange Rate System
With a fixed exchange rate system, the central banks involved commit to maintaining the
exchange rate within some band. In the figure, the upper and lower bounds of this band
are labeled ea and eL/ respectively, and the target level for the exchange rate is labeled eT.
As long as the demand for and supply of U.S. dollars results in an exchange rate within the
band (as at point A), no action by the Fed is required. If the demand for and supply of
dollars results in an equilibrium exchange rate above the upper bound, the Fed will have to
supply dollars to reduce the exchange rate. If the equilibrium exchange rate is below the
lower bound, the Fed will have to use its foreign currency reserves to demand dollars to
increase the exchange rate.
foreign central bank supplying dollars, will tend to drain that bank’s reserves
of U.S. dollars.
Fixed exchange rates require that monetary policy be devoted to main¬
taining the exchange rates within the fixed bands. Thus, the U.S. Federal
Reserve System would find that its main task is not to maintain a stable
money supply or a stable price level but to maintain a stable exchange rate.
If this means a stable money supply or a stable price level, all the better.
But under a fixed exchange rate system, the Fed must sacrifice its ability to
pursue other goals to almost singlemindedly pursue the goal of exchange
rate stability. For reasons we will see in Chapter 17, many nations do not
want to give up their ability to use monetary policy for domestic purposes,
and hence will not commit to a fixed exchange rate system.
A Brief History of the Foreign Exchange Rate System

You now have a basic understanding of how modem exchange markets work
and how they are affected by central bank intervention. We conclude by
taking a look at the history of exchange rate systems. As we will see, there
have been a number of such systems, ranging from the gold standard to the
present system in which many currencies float against one another.
The Classical Gold Standard

A gold standard pegs currency to a specified amount of gold—say, one
ounce. During most of the 1700s and 1800s, England was on a gold standard,
departing from it during the Napoleonic Wars and during World War I in
1914. England returned briefly to the gold standard in 1926, only to depart
again in 1931 during the Great Depression. The United States was on a gold
standard from 1879 until the midst of the Great Depression in 1933. The
period 1890-1914 is considered the high point of the gold standard. During
this period, countries announced they would buy or sell gold at stated prices.
This would set, or peg, the value of the currency in terms of gold. For
example, the United States pegged the dollar so that an ounce of gold was
worth about 20 U.S. dollars.2 Similarly, the British pegged the pound so that
an ounce of gold was worth about four English pounds. In essence these
actions not only set the value of the currencies in terms of gold but also set
the exchange rates, since a U.S. citizen could exchange $20 for one ounce
of gold and then use the gold to purchase four pounds. Thus, $20 was equal
in value to 4 pounds; that is, the exchange rate was $5 for 1 pound, or .2
pounds for $ 1.
How did the governments maintain these gold equivalents? Basically,
by standing willing to supply enough currency—dollars or pounds—to keep
the supply of and demand for the currency relative to gold in equilibrium at
the desired levels. What would happen, for example, if the U.S. dollar started
to appreciate so that the dollar price of gold was falling? As the equilibrium
price of gold was falling to $19 per ounce, the U.S. government would still
buy or sell gold for $20 an ounce. In this case, domestic and foreign residents
would trade the government gold for dollars, since the government would
give them $20 per ounce while the market price was only $19 per ounce.
Of course, the U.S. government would be accumulating gold. This increase
in gold would lead to an increase in the money supply and an eventual rise
in the price level. When the price level rose, the price of gold would rise,
2 Actually, $20,672 per ounce of gold.

and this would tend to restore equilibrium. In practice the price of gold never
departed much from the pegged levels, because the governments adjusted
their money supplies to fluctuations in the stock of gold to keep its price at
the pegged levels.
Like any fixed exchange rate system, the gold standard had its pluses
and minuses. The biggest minus was perceived to be the constraint the gold
standard imposed on monetary policy. Under a gold standard, a nation was
not free to increase the money supply at will, since doing so would tend to
destroy the pegged price of gold. Instead, any increases or decreases in the
money supply ended up being governed by gold flows into and out of the
government’s treasury. Some viewed this as a plus, however, because it
limited the government’s ability to inflate the money supply and prices.
Bretton Woods
The Great Depression and World War II spelled the end of the classic gold
standard. After World War II, the western nations operated under an
exchange rate system called the Bretton Woods system, named after the
town in New Hampshire in which the system was negotiated. Under this
system, the United States agreed to peg the U.S. dollar so that an ounce of
gold would sell for $35. Then each country fixed the exchange rate between
its currency and the U.S. dollar. This was a modified gold standard in which
each currency was indirectly pegged to gold via the dollar.
Unlike the classic gold standard, under the Bretton Woods system only
the United States fixed the exchange rate between its currency and gold. All
other nations bought or sold their currencies in terms of dollars, maintaining
fixed exchange rates with the dollar. The Bretton Woods system operated
until August 15, 1971, when President Nixon announced the United States
would no longer honor requests to exchange dollars for gold at $35 per
ounce.
What caused the demise of Bretton Woods? The basic problem was an
overvalued dollar, caused in part by the unwillingness of the United States
to pursue the policy required to maintain gold at $35 per ounce. This would
have required a contractionary monetary policy, which, as we will see in
Chapter 16, can cause temporary increases in unemployment and decreases
in output. Not willing to pursue such a policy, the United States instead
opted for more expansionary monetary policy that eventually led to the
downfall of the Bretton Woods system. Other contributing factors were the
unwillingness of other nations to adjust their exchange rates to correct the
problem of an overvalued dollar and an institutional setup that provided no
way for the United States to devalue its currency directly. Exchange rates
between the United States and the war-torn industrialized countries of Japan
and Europe became difficult to maintain as those nations recovered from
their war wounds and became increasingly competitive with the United
States on the economic front.
364 ChapterII Foreign Exchange Markets
For a brief period, nations attempted to create alternatives to the defunct

Bretton Woods system. But by February 12, 1973, the Japanese yen floated
against the dollar, and by March 16, 1973, the European Community fol¬
lowed suit. The world embarked on a floating or flexible exchange rate
system.
Post-Bretton Woods
From 1973 to the present, the United States has had a flexible exchange rate
to one degree or another. Market forces set the exchange rate of the dollar
against other currencies, with only occasional intervention by the Fed. How¬
ever, some nations have unilaterally pegged their currencies to the dollar or
to the currency of some other nation, and the European Community at¬
tempted a partial system of pegged exchange rates.
Some nations of the European Community started the European Mone¬
tary System (EMS) in 1979, in which exchange rates between member
countries were pegged to narrow bands. This system pegged these currencies
relative not to other currencies such as the dollar or the yen but only to one
another. This system seemed to be working relatively well until 1992, when
high German interest rates drove several currencies out of the EMS, most
notably the English pound. Box 11.4 discusses the EMS in more detail.
Summary | Late in the summer of 1992, the EMS experienced difficulty because Ger¬
Exercise 11.41I many raised its interest rates to new heights. The British government was
under great pressure to devalue its currency, because Britain did not want to
match Germany’s increase in interest rates. Explain this situation using the
foreign exchange market.
Answer: Graph the demand for and supply of British pounds, with the
exchange rate being German marks per pound. The supply curve is the supply
of pounds by Britons, while the demand curve is the demand for pounds by
Germans. When Germany raised interest rates, German assets became more
attractive than British assets, everything else constant. Thus, German demand
for pounds declined, since fewer Germans wanted to buy assets in Britain.
The British supply of pounds increased, however, because more Britons
wanted to buy assets in Germany. The net effect was to reduce the exchange
rate in marks per pound—a depreciation of the pound. To counter this, the
British central bank would have had to raise British interest rates, which it
refused to do in the midst of a recession. Instead, it chose to intervene
heavily in the foreign exchange market, purchasing pounds with foreign
reserves to increase demand and prop up the exchange rate. However, so
great were the interest rate differential and the pressure on the pound that in
the end Britain gave up and instead chose to let the pound float, effectively
abandoning the EMS for the time being.
The European Monetary System
The European Community, which has been cord of 1991 confirmed the goals of the DeLors
moving closer toward political and monetary report and established a timetable for achieving
union for some time, established the European those goals by January 1, 1999, at the latest.
Monetary System (EMS) in 1979. The EMS in¬ The EMS is widely credited with reducing
volves a system of fixed but adjustable exchange inflation rates in Europe, mostly because curren¬
rates. The eventual goal was to establish full cies are pegged to the German mark and Ger¬
European Monetary Union (EMU), with a com¬ many has maintained a low inflation rate. Be¬
mon European central bank and a common cur¬ cause of this, other nations also must maintain a
rency. These goals were framed by the DeLors low inflation rate. If not, the ratio of the price
report in 1989, which set goals of permanently level in Germany to the price level in another
fixed exchange rates (in effect, a common cur¬ nation will fall, and the exchange rate with the
rency) and integrated capital markets (with German mark will move outside of the fixed
monetary policy conducted not by the central rate. Thus, the EMS has imposed German mone¬
banks of member nations but by a European tary discipline on other nations.
central bank). More recently, the Maastricht Ac¬ Continued on p. 366
Continued from p. 365 ginning in early 1983 and lasting until early
The accompanying figure graphs the 1986. At this point the franc was devalued, and
exchange rates between the French franc and by early 1987 the franc had depreciated to
the German mark and between the Dutch guil¬ about 3.33 francs per mark. This rate has been
der and the German mark. Notice that the fairly closely maintained since then, so that in
Dutch guilder has maintained a fairly constant early 1993 the rate was 3.40 francs per mark.
exchange rate of about 1.12 guilders per mark The stabilization of the franc after an initial false
since the early 1980s. In contrast, the French start was an EMS success story until the latter
franc has a long history of depreciating relative half of 1993, when the EMS once again en¬
to the mark; the exchange rate rose from fewer countered serious problems.
than 1.5 francs per mark in the 1960s to more
than 3 francs per mark in the 1980s. As part of Source: Board of Governors of the Federal Reserve Sys¬
the EMS, the French franc was held at a fairly tem, Foreign Exchange Rates, various issues; Citibase
constant rate of about three francs per mark be¬ electronic database, and authors' calculation.
Conclusion
In this chapter, we examined the role of foreign exchange markets and
established a theoretical framework for predicting long-run exchange rate
movements by using the model of supply and demand used throughout this
text. We also looked at how purchasing power parity can be used to obtain
a gauge of general long-run movements in exchange rates. We saw that
short-term exchange rates are determined largely by arbitrage and interest
rate parity, and that short-term movements in exchange rates are often dif¬
ficult to predict because of the role of short-run expectations. We concluded
by looking at how central banks intervene in currency markets and briefly
overviewed the foreign exchange rate systems used during this century. In
the next chapter, we take an in-depth look at how participants in financial
markets form expectations and forecasts of economic variables like inflation
and exchange rates.
KEY TERMS
foreign exchange market spot rate

exchange rate forward exchange rate
depreciation hedger
appreciation speculator
purchasing power parity arbitrageur
interest rate parity triangular arbitrage
Key Terms continued
covered interest arbitrage gold standard

flexible (floating) exchange rates Bretton Woods
dirty float European Monetary System (EMS)
fixed exchange rates
1. Based on what you learned in this chapter, 6. Use a diagram to illustrate the impact of
why do you think exchange rates tend to the following on the supply of dollars to
fluctuate over long periods of time? Over Italians: (a) a decrease in Italians’ real in¬
short periods of time? comes, (b) a decrease in the Italian price
level, and (c) a decrease in the interest
.
2 Suppose the exchange rate between Dutch
rate in Italy.
guilders (DF) and U.S. dollars is 1.7 guil¬
ders per dollar. 7. Determine the impact of the following on
(a) . How much would it cost in dollars to the exchange rate between U.S. dollars
purchase a box of rare tulips that sell for and Italian lire:
980 guilders? (a) . Italy experiences a minor recession,
(b) . How much would it cost in guilders while the U.S. enjoys an economic boom.
to purchase a car that sells for $12,500? (b) . Interest rates in Italy fall by 3 per¬
cent.
3. How would your answer in problem 2
(c) . Interest rates in the United States rise
change if the exchange rate changed to
by 4 percent.
three guilders per dollar? Has the dollar
(d) . The price level in the United States
appreciated or depreciated against the
increases by 10 percent.
guilder? Has the guilder appreciated or
depreciated against the dollar? 8. Suppose interest rates in the United States
increase from 4 to 8 percent, while those
.
4 How would your answer in problem 2
in Japan fall from 4 to 2 percent. What
change if the exchange rate changed to
affect would you expect this to have on
1.5 guilders per dollar? Has the dollar
the long-run exchange rate between dol¬
appreciated or depreciated against the
lars and yen?
guilder? Has the guilder appreciated or
depreciated against the dollar? .
9 Suppose the government imposed a tax on
income earned by non-residents on U.S.
.
5 Use a diagram to illustrate the impact of
debt obligations. Would you expect the
the following on the demand for dollars
dollar to appreciate or depreciate relative
by Italians: (a) a decrease in Italians’ real
to the yen? Explain.
incomes, (b) an increase in the Italian
price level, and (c) an increase in the in- 10. Between 1985 and 1990, the dollar depre¬
terest rate in Italy. ciated considerably against the British
pound. Do you think this contributed to a have depreciated against the currency of
decline in the number of Americans who virtually every western industrialized
visited Britain? Explain. country. Why do you think this has hap¬
pened? (Hint: Go to your library and ob¬
.
11 Based on what you learned in this chapter, tain information about changes in the
provide three possible explanations for price level in Mexico and in various South
each of the following events: American countries during the past 10
(a) . Over the last four years, the dollar years.)
depreciated against the pound.
(b) . Today the dollar appreciated against .
13 In light of problem 12, what actions could
the mark. you recommend to the Mexican central
bank to remedy the depreciation of the
.
12 During the past decade, the Mexican peso peso? Could your policy be sustained for
and various South American currencies very long? Explain.

Ahking, F. W. ‘ The Dollar/Pound Exchange Rate in can Data.” Journal of Public Economics, 42
the 1920s: An Empirical Investigation.” Southern (August 1990), 329-356.
Economic Journal, 55 (April 1989), 924-934. Harrison, M. J., P. D. McNelis, and S. N. Neftci. “A
Ahking, F. W. “Further Results on Long-run Pur¬ Diagnostic Check for Model Specification: An Ap¬
chasing Power Parity in the 1920s.” European Eco¬ plication to the Yen-Dollar Exchange Rate.” Eco¬
nomic Review, 34 (July 1990), 913-919. nomic Betters, 33 (May 1990), 69-73.
Akgiray, V., et al. “A Causal Analysis of Black and Honohan, P., and P. McNelis. “Is the EMS a DM
Official Exchange Rates: The Turkish Case.” Welt- Zone?—Evidence from the Realignments.” Eco¬
wirtschaftliches Archiv, 125 (1989), 337-345. nomic and Social Review, 20 (January 1989), 97-
Antonovita, F., and R. D. Nelson. “Forward and Fu¬ 110.
tures Markets and the Competitive Firm under Kulkarni, K. G., and D. Chakraborty. “An Empirical
Price Uncertainty.” Southern Economic Journal, Evidence of Purchasing Power Parity Theory: A
55 (July 1988), 182-195. Case of Indian Rupee and U.S. Dollar.” Margin,
Blenman, L. P. “A Model of Covered Interest Arbi¬ 22 (October 1989-March 1990), 52-56.
trage under Market Segmentation.” Journal of Mark, N. C. “Real and Nominal Exchange Rates in
Money, Credit, and Banking, 23 (November 1991), the Long Run: An Empirical Investigation.” Jour¬
706-717. nal of International Economics, 28 (February
Breece, J. H. “Devaluation, Money and Currency 1990), 115-136.
Substitution.” Economia Internazionale, 40 (May/ McMillin, W. D., and F. Koray. “Does Government
August 1987), 172-191. Debt Affect the Exchange Rate? An Empirical
Driskill, R., and S. McCafferty. “Exchange-Rate De¬ Analysis of the U.S.-Canadian Exchange Rate.”
termination: An Equilibrium Approach with Imper¬ Journal of Economics and Business, 42 (November
fect Capital Substitutability.” Journal of Interna¬ 1990), 279-288.
tional Economics, 23 (November 1987), 241-261. Rogers, J. H. “Foreign Inflation Transmission under
Driskill, R., and S. McCafferty. “Speculation, Ra¬ Flexible Exchange Rates and Currency Substitu¬
tional Expectations, and Stability of the Foreign tion.” Journal of Money, Credit, and Banking, 22
Exchange Market.” Journal of International Eco¬ (May 1990), 195-208. ”
nomics, 10 (February 1980), 91-102. Sayer, W. C., and R. L. Sprinkle. “Contractionary Ef¬
Feltenstein, A., and S. Morris. “Fiscal Stabilization fects of Devaluation in Mexico.” Social Science
and Exchange Rate Instability: A Theoretical Ap¬ Quarterly, 68 (December 1987), 885-893.
proach and Some Policy Conclusions Using Mexi-
CHAPTER
Rational Expectations and

Efficient Markets
hroughout the first half of this book, we have seen that expectations
of consumers, banks, and other businesses have pronounced effects on supply
and demand in financial markets. If inflationary expectations rise, the supply
of and demand for loanable funds shift, and higher interest rates result.
Similarly, we saw in the last chapter that when traders in the foreign
exchange market expect the exchange rate to fall, the result is a rise in the
cost of buying imports. Thus, changes in expectations affect the rate you
pay on car loans and, if your car is an import, the sticker price of your car.
We conclude the first part of the book by taking a closer look at how
participants in financial markets form expectations and what might cause
those expectations to change over time.
We learned in Chapter 9 that risk is present when individuals know the
likelihood that different outcomes will occur. In many instances, however,
individuals do not know the underlying probabilities of different outcomes.
For instance, do you know the probability that the inflation rate will rise to
10 percent or fall to 1 percent next year? In such instances of uncertainty,
you and other individuals must base your expectations on a subjective as¬
sessment of different future outcomes. In the first part of this chapter, we
look at various methods used to form expectations under uncertainty. Then
we deal with a related issue—the efficient markets hypothesis—which has
profound implications for your ability to 6 ‘beat’ ’ financial markets like the
stock market. To put all of this in perspective, let us review the important
role expectations play in financial markets.
The Role of Expectations in Financial Markets
Throughout this book you have seen how expectations affect financial mar¬
kets and, more generally, the economy. Our analysis of the Fisher equation
in Chapter 3, coupled with Chapter 4’s model of loanable funds, showed
how inflationary expectations influence interest rates. Chapters 5 and 6 re¬
vealed that a bank’s portfolio of assets and liabilities depends in part on the
expected return to various assets and the expected cost of various liabilities.
369
370 Chapter 12 Rational Expectations and Efficient Markets
For example, a bank’s amount of excess reserves is determined partly by

the expected return on alternative assets such as loans and bonds and partly
by expectations of the rate of deposit withdrawals. In Chapter 9, we saw that
the price of a corporate bond will depend on investors’ expectations about
the bond’s default risk. Chapter 10 revealed that the term structure of interest
rates is affected by expectations, since long-term interest rates depend on
expectations of future short-term interest rates. Chapter 11 showed how
expectations influence short-term exchange rates.
In later chapters, we examine the implications of expectations on such
factors as money creation (Chapter 14), the supply of and demand for money
(Chapter 15), aggregate supply and demand (Chapter 16), and various aspects
of monetary policy conducted by the Fed (Chapters 18 through 21). In fact,
we will see that increases in the money supply have different effects on
employment and real output depending on whether changes in the money
supply are anticipated by workers or by firms.
As you can see, expectations play a central role not only in money and
banking but in every aspect of economic life. In the next section, we look
at methods people use to form expectations.
Three Theories of Expectation Formation

Suppose you want to forecast the inflation rate for the current year to deter¬
mine the expected real rate of interest on your saving deposits. By the Fisher
equation, your expected real interest rate (r) is the difference between the
nominal interest rate (/) paid by the bank and your inflationary expectations
On*):
r — i — ire.
If your bank pays 3 percent on saving deposits and you expect inflation to
be 2 percent over the next year, your expected real interest rate is only 1
percent. If your inflationary expectations are higher—say, 5 percent—your
expected real interest rate is — 2 percent, meaning your deposits would lose
2 percent of its purchasing power while in the bank.
Obviously your incentives to deposit money in a savings account de¬
pend not only on the rate the bank offers but on your inflationary expectations
as well. The same is true for other savers, including large investors like
mutual funds, insurance companies, and traders on Wall Street. Furthermore,
there is a cost to making errors in your inflationary expectations. If your
inflationary expectations are too low on average during the course of your
life, your errors may lead you to make investment decisions that yield a
negative real return. How, then, do you go about forming expectations of
inflation?
In this section we look at three ways to form expectations: the Markov
expectations, adaptive expectations, and rational expectations hypotheses.
Each approach requires a different level of sophistication on the part of those

forming expectations. To compare and contrast expectation formation under
each theory, we focus on the formation of inflationary expectations, since
they are an important determinant of virtually every financial market we
have studied. As we will see, however, you may use similar methods to
forecast other variables—your grade in this course, for instance.
Markov Expectations
The simplest way to form expectations about future events is to form Markov
expectations. The Markov expectations hypothesis asserts that individuals
expect the future to be like the most recent past. For instance, if the inflation
rate last year was 3.5 percent, you can simply presume the future rate will
be the same as last year’s: 3.5 percent. With Markov expectations, individ¬
uals expect tomorrow to be exactly like today.
An obvious shortcoming of Markov expectations is that they do not take
into account knowable events that might change the future environment. For
instance, if the inflation rate is rising, Markov expectations will continually
miss the mark. Each year you will expect the inflation rate to remain at its
previous level, but each year you will be surprised to find it higher than
expected. As we will see, another shortcoming is that this hypothesis ignores
other information that might be useful in predicting the future. Markov
expectations are based solely on the most recent value of the variable being
forecast.
The Mechanics of Markov Expectations. Under Markov expec¬

tations, the expected value of xt, based on information available at time t —
1, is simply xt-\.
xf = xt_x.
Thus, you simply use the most recent known value of x to forecast future
values of x.
Although Markov expectations suffer from a number of deficiencies,
they are a reasonable method of forming expectations when the forecasting
environment is stable rather than changing over time. For instance, if infla¬
tion tends to be stable at 3.5 percent, as it was during much of the 1980s,
using last year’s rate of 3.5 percent to predict this year’s inflation rate will
not lead to any systematic error in your inflation forecast. In contrast, when
the structure of the economy and economic variables are in a state of flux
so that the inflation rate systematically rises or falls year after year as it did
during the 1970s, Markov expectations tend to yield systematic errors. This
is because the past is a poor predictor of the future during periods of change,
and Markov expectations actually assume the future will be like the most
recent past. In the next section, we take a deeper look at the deficiencies of
Markov expectations during periods of change.
An Example: Markov Inflationary Expectations. If the infla¬

tion rate last year was 3.5 percent (nt-i = -035), the Markov inflationary
expectation is also 3.5 percent (uf = .035). Clearly, forming Markov ex¬
pectations of this year’s inflation rate is a piece of cake: We simply obtain
information about the most recent rate of inflation and use it as the forecast.
Table 12.1 shows that Markov expectations perform well when the
economic environment is stable but can lead to systematic errors in an
environment of change. The second column represents the actual inflation
rate in each of 10 hypothetical periods. The third column gives the Markov
expectation of these inflation rates, which are simply the previous period’s
inflation rate. The final column gives the forecast errors, the difference
between the actual and expected rates of inflation. Notice that an individual
with Markov expectations cannot form a forecast of the inflation rate in
period 1, since there is no previous period on which to base the forecast.
Once the individual observes that the actual inflation rate in period 1 is 5
percent, this value becomes the forecast of the inflation rate for period 2. No
forecast errors occur for periods 2, 3, and 4 because the actual inflation rate
did not change between periods.
In period 5, however, the actual inflation rate rises to 11 percent. This
individual’s expectation of the inflation rate in period 5 is the inflation rate
in period 4, again 5 percent. Since the actual inflation rate in period 5 is 11
percent, the forecast error is 6 percent.
Once the individual observes the inflation rate in period 5 is 11 percent,
the forecast of the inflation rate in period 6 becomes 11 percent. But in
period 6, the actual inflation rate is 25 percent, so the forecast error is 14
percent. In periods 7 through 10, the actual inflation rate remains at 25
percent, so the forecast error is zero. Thus, Table 12.1 illustrates that Markov
expectations work well in environments where inflation does not change
from year to year but leads to systematic forecast errors in environments of
change. In particular, when the inflation rate increases over time, Markov
expectations will lead to systematic underestimates of inflation. Forecasters
never learn from their past mistakes, and errors result each time the inflation
rate rises.
Adaptive Expectations
The adaptive expectations approach assumes expectations evolve over
time in light of past experience. It allows past trends in variables as well as
previous forecast errors to affect future expectations as people slowly learn
from past mistakes. For instance, a person may notice that her expectation
of inflation was too high the previous period and thus revise downward her
expectation of the future inflation rate. Since adaptive expectations look
backward, they take into account the past history of a variable and of fore¬
casts of that variable.
Table 12.1
:
A < X. A,:
,
X","N- ■ sS'K - & A •'-"••.'■'A Markov Expectations of Inflation
Markov expectations of inflation will be correct as long as the inflation rate does not
change. This is the case in periods 2 through 4, where the forecast errors are zero. When
the rate of inflation does change, as in periods 5 and 6, Markov expectations will lead to
large forecast errors.
II
ri
H*
1
Actual Inflation Expected Inflation Forecast Error
Period (t) Rate (7Tt) Rate (irf) (Tlt ~ O
1 5% — —
2 5 5% 0%
3 5 5 0
4 5 5 0
5 11 5 6
6 25 11 14
7 25 25 0
8 25 25 0
9 25 25 0
10 25 25 0
A potential problem with adaptive expectations is that when the future

is continually changing, past history is not always a good forecast of the
future. Moreover, we will see that adaptive expectations do not use all the
information that is useful in forecasting the future.
The Mechanics of Adaptive Expectations. Adaptive expecta¬

tions, like Markov expectations, are based on past experience. Unlike Mar¬
kov expectations, however, adaptive expectations adjust current expectations
based on information about the accuracy of prior forecasts. More specifically,
the current adaptive expectation about x in period t given what is known in
period t — 1, denoted xf, is given by
4 = 4-\ + k(xr_| - xf_j).
This formula states that the current expectation of x, xf, equals last period’s
expectation, xf_b plus a term that adjusts this expectation in light of past
errors, \(x,_! — xf_j), where the expression in parentheses is last period’s
forecast error and X is called the adjustment factor. More generally, adap¬
tive expectations make current forecasts a weighted average of past obser¬
vations.
The speed of adjustment, X, determines how quickly expectations adjust
to past errors. If X is zero, expectations never adjust, and we always have
xet = Xf-\. In this case, the current expectation is identical to that in the
previous period, regardless of how inaccurate the expectation was last period.
In contrast, if X is 1, expectations always adjust instantaneously to the
previous value of the variable, and hence xet = xt_x. This illustrates that
when X — 1, we have a special case of adaptive expectations called Markov
expectations.
An Example: Adaptive Inflationary Expectations. Table 12.2

presents an example of adaptive inflationary expectations when the speed
of adjustment, X, is .5. The data in column 2 are hypothetical, but they
illustrate a situation where inflation permanently increases from 5 to 10
percent in period 5. This is similar to what occurred in the United States
during the last half of the 1970s, when the inflation rate doubled. To read
this table, take as given the row for period 1; it simply says that we are
starting from a point where the expectation of inflation is consistent with the
actual inflation rate. Given that expectations were not in error in period 1,
the adaptive expectation of the inflation rate in period 2 is
tt| = Trf + X(tT] — Trf) = .05 + .5(.05 — .05) = .05.
That is, since the forecast error in period 1 was zero, the inflation rate in
period 2 is expected to be what was expected in period 1: 5 percent. When
the inflation rate is announced in period 2, it is 5 percent—exactly what was
expected. Given this, the forecast of the inflation rate for period 3 is
Trf = Trf + X(tt2 — Trf) = .05 + .5(.05 — .05) — .05,
which is the same as we expected the previous period. Notice this situation
continues through period 4.
In period 5, the forecast of the inflation rate for period 5 is again given
by
Trf = Trf + X(tt4 — Trf) = .05 + .5(.05 — .05) = .05.
Unfortunately, however, in period 5 the actual inflation rate is 10 percent,

and the expectations are off by 5 percent. Consequently, when the forecast
of inflation for period 6 is made, this error in expectation affects the estimate
of inflation:
tt| = tt| + X(tt5 — tt|) = .05 + .5(40 — .05) = .075.
Notice that the expected rate of inflation in period 6 is higher than that in
periods 1 through 5. The reason is that in period 5, the forecast error was
(.10 — .05) = .05. This is multiplied by the speed of adjustment in expec-
Table 12.2
Adaptive Expectations with a Permanent Jump in Inflation: A. = .50
Like Markov expectations, adaptive expectations perform well when the inflation rate in column 2 is
constant (periods 1 through 4). In period 5 the actual inflation rate increases to 10 percent, but the
adaptive expectation of inflation for that period remains at 5 percent. Given the permanent jump in the
inflation rate, the forecast error gradually declines in periods 7 through 10. The adaptive expectation of
inflation is still off, however, even in period 10.
< = <-! + .5(71, _! - 7T?_!)
Expected Inflation Actual Inflation Forecast Error

Actual Inflation Expected Inflation Rate Last Period Rate Last Last Period
Period (t) Rate (tt,) Rate (rif) W-i) Period (tt^_x) (tt,_! — net-\)
1 5% 5% 5% 5% 0%
2 5 5 5 5 0
3 5 5 5 5 0
4 5 5 5 5 0
5 10 5 5 5 0
6 10 7.5 5 10 5
7 10 8.75 7.5 10 2.5
8 10 9.375 8.75 10 1.25
9 10 9.6875 9.375 10 0.625
10 10 9.84375 9.6875 10 0.3125
tations (X = .5), which yields .5(.10 — .05) = .025. When this is added to
the previous year’s expectation of .05, the result is .075—a 7.5 percent
expected inflation rate for period 6. This example illustrates how errors in
past forecasts influence adaptive inflationary expectations as people attempt
to learn from past mistakes.
The estimate of the inflation rate in period 7, given what is known in
period 6, is the period 6 expectation (7.5 percent) plus half of the previous
error in expectations (2.5 percent). Thus, the expected rate of inflation in
period 7, formed in period 6, is 8.75 percent. This process continues, and
with a constant actual inflation rate of 10 percent through period 10, the
expectation in period 10 has reached 9.84375 percent—just slightly lower
than the actual inflation rate in this period.
Figure 12.1 plots the results from Table 12.2 and reveals an important
feature of adaptive inflationary expectations. When the actual rate of inflation
jumps to 10 percent, expectations adjust slowly and only gradually get closer
Here inflation is stable Figure 12.1

until period 5, when it Gradual Adjustment of Adaptive Expectations
permanently jumps to
10 percent. Adaptive
expectations only Actual
gradually adjust to the Actual
and Expected
higher inflation rate, Inflation v
Inflation (%)
leading to systematic Rate \
errors in period 5 and 10
beyond.
\ Adaptive
/ x Inflationary
Expectations
5 10 Period
to the actual rate of inflation. This is because adaptive expectations make

the current forecast a weighted average of past observations. During the
adjustment process, systematic errors occur in inflationary expectations,
since inflationary expectations are continually below the actual inflation rates
in Figure 12.1 after the permanent jump in inflation in period 5.
The speed of adjustment, X, determines just how quickly expectations
adjust to the jump in inflation in period 5. To see this, suppose expectations
adjust faster—X = .75. Given the higher value of X, expectations adjust
much more quickly to the change in the inflation rate, as you can see by
comparing Table 12.3 with Table 12.2. For instance, when X = .75, the
expected inflation rate in period 8 in Table 12.3 is roughly 9.92 percent,
compared with 9.375 percent in Table 12.2.
Figure 12.2 illustrates how changes in the speed of adjustment affect
inflationary expectations. The higher X is, the more rapidly adaptive expec¬
tations converge to the true inflation rate of 10 percent. Similarly, the lower
the value of X, the slower expectations are to adjust. Still, Figure 12.2, on
page 378, illustrates that even with a higher speed of adjustment, adaptive
expectations lead to a systematic bias in inflationary expectations when a
permanent jump in inflation occurs.
IMMlIllISlIlliM
, ■
Table 12.3
■
Adaptive Expectations with a Permanent Jump in Inflation: A. = .75
:
This table shows how the numbers in Table 12.2 change with a faster speed of adjustment in expecta¬
tions. As in Table 12.2, the inflation rate permanently changes in period 5, which leads to a forecast
error. With a faster speed of adjustment, however, the forecast errors are smaller and the gap between
actual and expected inflation closes much more quickly than that with a slower speed of adjustment.
< = <-i + •75('jr, _ i - <-,)

Actual Inflation Expected Inflation Rate Last Rate Last Last Period
Period (t) Rate (tt,) Rate (irf) Period firf-i) Period (7jy_ x) (ttx)
1 5% 5% 5% 5% 0%
2 5 5 5 5 0
3 5 5 5 5 0
4 5 5 5 5 0
5 10 5 5 5 0
6 10 8.75 5 10 5
7 10 9.6875 8.75 10 1.25
8 10 9.921875 9.6875 10 .3125
9 10 9.980469 9.921875 10 .078125
10 10 9.995117 9.980469 10 .019532
Since the speed of adjustment is important to adaptive expectations, you

may be wondering why individuals don’t react by setting X as high as
possible to get very quick adjustment to changes in the actual inflation rate.
The reason is that not all changes are permanent as they are in Tables 12.2
and 12.3, where the increase in the inflation rate persists through time.
Sometimes the inflation rate increases for a period but then returns to its
previous level. This is precisely what occurred in the United States between
1946 and 1948, when the inflation rate more than doubled only to return to
its previous level. Table 12.4, on page 379, shows how a one-time blip in
inflation affects adaptive inflationary expectations when X = .5. In this case,
an error in expectations occurs in period 5, when the actual inflation rate is
10 percent but the expectation is only 5 percent, and another occurs in period
6, when the actual inflation rate returns to 5 percent but the expectation is
7.5 percent. Furthermore, if the speed of adjustment were faster—say, .75
Here inflation is stable * Figure 12.2

until period 5, when it How the Speed of Adjustment Affects Adaptive Expectations
permanently jumps to
10 percent. When the
speed of adjustment is
higher (X = .75),
adaptive expectations
converge to the new
inflation rate more
rapidly than when it is
lower (X = .5).
as it was in Table 12.3—the expectation in period 6 would be 8.75 percent

instead of 7.5 percent, an even larger forecast error. Thus, there is a tradeoff
between quickly responding to permanent changes in the level of the inflation
rate, as in Tables 12.2 and 12.3, and not overreacting to one-time blips in
the inflation rate, as in Table 12.4.
We should note some important features of adaptive expectations. First,
like Markov expectations, adaptive expectations ignore other variables that
might be useful in forecasting the future. For instance, based on what we
learned about the effect of the money supply on the price level in Chapter
3, changes in the money supply might be useful in forecasting the inflation
rate. Yet adaptive expectations (and Markov expectations) use only past
values of the inflation rate to predict future rates of inflation. Second, even
in response to a permanent change in the inflation rate, as in Tables 12.2
and 12.3, the adjustment of expectations is only gradual. After a permanent
change in the actual inflation rate occurs, expectations take time to converge
to the new inflation level. Yet when there is a one-time blip in the inflation
rate, as in Table 12.4, expectations overreact and return to the actual inflation
rate only gradually. Thus, whether a permanent change or a temporary blip
occurs, adaptive expectations will systematically deviate from the actual
inflation rate for several periods after the change.
Table 12.4
Adaptive Expectations with a Blip in Inflation: \ = .5
Even with a temporary change (a one-time blip) in the inflation rate in period 5, adaptive expectations
only gradually lead to more accurate inflation forecasts. In period 5 the inflation forecast is too low, but
in periods 6 through 10 the forecasts are too high.
Tit = + .5(tt,_! - <_i)

Actual Inflation Expected Inflation Rate Last Rate Last Last Period
Period (t) Rate (tt,) Rate (Trf) Period (ttL_i) Period fa-i ~ <-i)
1 5% 5% 5% 5% 0%
2 5 5 5 5 0
3 5 5 5 5 0
4 5 5 5 5 0
5 10 5 5 5 0
6 5 7.5 5 10 5
7 5 6.25 7.5 5 -2.5
8 5 5.625 6.25 5 -1.25
9 5 5.3125 5.625 5 -.625
10 5 5.15625 5.3125 5 -.3125
Rational Expectations
The main criticism of Markov and adaptive expectations is that they may
not be based on variables economic theory suggests are useful in forming
more accurate expectations. This has led to the idea of rational expectations,
which is the most sophisticated of the three methods of forming expectations.
The rational expectations hypothesis asserts that when people form expec¬
tations, they use all knowable information, including variables economic
theory suggests are relevant for making predictions. It is important to realize
that rational expectations do not imply forecasts are always correct, only
that no systematic errors in forecasts occur. Since rational expectations in¬
corporate all relevant information, any deviations from what is expected are
purely random in nature; any forecast errors that arise are just as likely to
be positive as negative, as Box 12.1 shows.
Let us look at an example of rational expectations that is relevant to
your performance in this money and banking course. Suppose you are at-
Errors in the Inflationary Expectations

of "Experts"
In Box 4.1, we looked at the inflationary expec¬ less important than not making huge mistakes
tations of so-called experts who sell their fore¬ or systematic errors in forming inflationary ex¬
casts to businesses and governments. Do these pectations. Since the errors these experts make
experts make systematic errors in their forecasts, are purely random and are correct on average,
or are the forecasts correct, on average, with these forecasters are not systematically surprised
purely random errors? Look at the accompany¬ by high or low inflation. In other words, the ex¬
ing figure, which plots the difference between perts appear to form rational inflationary expec¬
experts' expectations and the actual inflation tations. As we will see later in this chapter when
rate, and see what you think. we look at the efficient markets hypothesis, this
Notice that the forecast errors do seem to is not surprising because inflationary expecta¬
randomly fluctuate around zero, so the experts' tions play an important role in financial markets.
inflationary expectations tend to be correct on
average. Still, the forecasts are seldom exactly
right; that is, even professional forecasters of in¬
Source for data: ASA-NBER Business Outlook Survey, var¬
flation are making predictions that are mostly ious issues, Citibase electronic database, and authors' cal¬
wrong—and they know it. Being exactly right is culations.
10 "1
8 -
6 -
4 -
LLI <T3
-l-> CD
CO ^
£
2 -
S &
£ £
C= O) 0
o t!
V' CD
CD Q_
-2
-4 -
-6 -
-8
'82
tempting to forecast the grade you will make on your final exam. If you
have Markov expectations, your forecast will be the grade you earned on
your last midterm exam. If you have adaptive expectations, your forecast
will depend on what you expected to do on your last exam, as well as on
your forecast error. If you did better on your last exam than you expected,
your forecast of your final exam grade will be your previous forecast plus
an additional amount to correct for your past error (underestimating your
grade). In other words, you will expect to do better on the final than on the
previous midterm.
In contrast, if you use rational expectations, your forecast will depend
on all known variables that could theoretically affect your grade. These
variables would include your propensity to excel in this subject (your past
grades in the course), the number of other finals you have on the same day,
how much time you plan to study, the tendency of your professor to give
finals that are more difficult than midterms, and so on. Clearly each of these
variables is important in forecasting your grade in the course; yet Markov
and adaptive expectations ignore variables other than your past performance
in this course.
The Mechanics of Rational Expectations. In forming a rational

expectation about interest rates, bond prices, or exchange rates, an individual
would list all the determinants of demand and supply. All available infor¬
mation about these determinants would be used to help predict the location
of demand and supply curves to form an expectation of the equilibrium
interest rate, bond price, or exchange rate. In this way, rational expectations
are based on the known variables economic theory views as relevant; thus,
there are no systematic forecasting errors.
An Example: Rational Inflationary Expectations. Since infla¬

tion is a key determinant of demand and supply in financial markets, we will
look at how rational expectations can be used to form inflationary expecta¬
tions. This will not only allow us to compare rational inflationary expecta¬
tions with the Markov and adaptive expectations examined earlier but will
also be useful for our analysis of macroeconomics and monetary policy in
the second half of the book.
The first step in forming rational inflationary expectations is to determine
the relevant theory that explains inflation. For instance, Chapter 3 described
how classical economists developed the simple quantity theory of money
that is based on the equation of exchange. According to this theory, velocity
is constant, and changes in the money supply have no impact on the econ¬
omy’s real output. This theory, along with the equation of exchange, implies
that the inflation rate equals the growth rate in the money supply (gM) minus
the rate of growth in real output (gY):
77 ~ 8 m 8y-
This is a theory, however, and as such it may not be the correct explanation
for inflation (we will take a more careful look at competing theories and
historical evidence in Chapters 16, 17, and 18). But if correct, this theory
implies that a 10 percent increase in the money supply, coupled with a 3
percent increase in real output, will result in a 10% — 3% = 7% rate of
inflation. According to the simple quantity theory, inflation is caused by
increases in the money supply that exceed the growth rate in real output:
Too much money is chasing too few goods.
To make things as simple as possible, we first assume the growth rate
in real output is zero and the inflation rate in the current period equals the
growth rate in the money supply in the previous period. (In Chapter 18, we
will see that it often takes many months for changes in the money supply to
affect the price level). Given this simple economic model of inflation, in¬
cluding our simplifying assumption of no real output growth, the rational
expectation of the inflation rate in period t depends only on the money
growth rate in period t — 1. In fact, the inflation rate today is the money
growth rate of last period. If this information is available, the rational
inflationary expectation under the simple quantity theory is simply the
money growth rate last period:
rrrRE — r,
^7 §Mt- r
For instance, if the money supply increases by 7 percent in period t — 1,

the rational expectation of the inflation rate in period t is 7 percent. Notice
that this rational expectation does not depend on past inflation rates or
previous errors in expectations.
Table 12.5 compares rational and adaptive expectations for a situation
in which money growth is 5 percent for the first three periods and then jumps
to 10 percent in periods 4 through 10. The actual inflation rate is 5 percent
for the first four periods and then jumps to 10 percent in the remaining
periods, just as it does in Tables 12.2 and 12.3.
For the first four periods inflation is stable, and so are everyone’s ex¬
pectations. What does a person forming a rational expectation under the
simple quantity theory expect the inflation rate to be in period 5? Since the
growth rate in the money supply in period 4 was 10 percent, this person’s
forecast of inflation in period 5 is
^ = gM4 -10’
or 10 percent, which coincides with the actual inflation rate that occurs in
period 5. As a point of contrast, Table 12.5 also includes the adaptive
expectations first calculated in Table 12.2 for the same trend in inflation
rates. In period 5, adaptive expectations miss the mark by 5 percent. Fur¬
thermore, they adjust to the permanent increase in the inflation rate only
gradually, because adaptive expectations are not based on the economic
theory that informs us that money growth is related to inflation.
In the example in Table 12.5, rational expectations result in inflation
forecasts that are perfect. It is important to recognize that reality is more
Table 12.5
Rational Versus Adaptive Expectations with a Permanent Jump in Inflation
Since rational expectations of inflation incorporate information not used by adaptive expectations, they eliminate sys¬
tematic errors in inflation forecasts. The simple quantity theory assumes velocity is constant. It predicts that when real
output is also constant, increases in the money supply lead to inflation. Consequently rational expectations of inflation
based on the simple quantity theory are based on the previous period's growth in the money supply. The rational
expectations of inflation in column 4 coincide with the actual inflation rates in column 2, illustrating the absence of any
forecasting errors in this simple case. In more complex environments, rational expectations do not entirely eliminate
forecasting errors, but they do ensure that no systematic errors occur.
re .
■- gMt-1
77/ = <_! + .5(77^! - TtJ-j)
Money Growth Actual Inflation Rational Expectation Adaptive Expectation

Period (t) Rate (gMt) Rate (jTt) of Inflation (77? £) of Inflation (77/)
1 5% 5% 5% 5%
2 5 5 5 5
3 5 5 5 5
4 10 5 5 5
5 10 10 10 5
6 10 10 10 7.5
7 10 10 10 8.75
8 10 10 10 9.375
9 10 10 10 9.6875
10 10 10 10 9.84375
complicated than our simple example and that rational expectations do not
require zero forecasting errors. As we noted earlier, rational expectations
will sometimes lead to incorrect forecasts. The reason is that typically ran¬
dom, unpredictable events could occur that cause the inflation rate to rise or
fall by more than expected based on available information. For instance,
velocity might rise or fall randomly over time, resulting in errors in rational
inflationary expectations that are unavoidable if changes in velocity are
themselves unpredictable.
Part a of Figure 12.3 shows such a situation. Notice that the actual
inflation rate fluctuates randomly about 5 percent, while expectations are
consistently at 5 percent. But deviations between actual and expected infla¬
tion are purely random, and there are no systematic errors. In contrast, part
b shows a situation where systematic errors in inflationary expectations occur
Figure 12.3
Nonsystematic and Systematic Errors in Inflation Forecasts
f i' s l- 'Cl'
Part a shows a situation where inflationary expectations are correct on average and have no systematic
errors. Even though inflation fluctuates around 5 percent, the resulting errors in expectations are purely
random. In contrast, in part b systematic errors in inflationary expectations occur after period t, since
they are too low on average.
Actual Actual
and Expected and Expected
Inflation (%) Inflation (%)
Actual Inflation
maaaaamm/
Inflationary
Expectations
Period
(a) Nonsystemic Errors
after period t. In this case, inflationary expectations beyond period t always

understate actual inflation, reflecting systematic errors in expectations. Even
a person forming rational expectations cannot predict changes in inflation
that are purely random, and therefore might make errors in forecasts like
those in part a of Figure 12.3. But if all available information is being used,
no systematic errors will arise in forecasts like those in part b. Thus, the
advantage of rational expectations over other methods of forming expecta¬
tions is not the elimination of all forecasting errors but the elimination of
systematic forecasting errors.
The ability of rational expectations to eliminate systematic forecasting
errors relative to those that would arise under adaptive expectations can be
seen in yet another way. Consider a one-time blip in inflation, such as the
one that occurred in the United States between 1946 and 1948. During this
period, the inflation rate increased from about 3 to 13 percent and then fell
back to about 3 percent thereafter. The rise in inflation was caused by
increases in the money supply during World War II, and the subsequent
decline in inflation was due to the reduced money growth that followed the
war. Figure 12.4 presents a stylized diagram that illustrates how this type of
change in the economic environment affects rational and adaptive expec¬
tations.
Notice that the inflation rate fluctuated around 3 percent until 1946; then
it began fluctuating around 13 percent until 1948, when it returned to 3
percent. A person forming rational inflationary expectations would have
noticed the increased growth in the money supply during the war and raised
inflationary expectations for the 1946-1948 period accordingly. As part a
of Figure 12.4 shows, this gave rise to inflationary expectations that con¬
tained no systematic errors prior to 1946, when inflation averaged 3 percent,
and no systematic errors for the 1946-1948 period, when inflation averaged
13 percent. Furthermore, inflationary expectations returned back to 3 percent
for 1948 and beyond, as people forming rational expectations fully antici¬
pated inflation would decline due to reduced monetary growth following the
war. Part a of the figure illustrates that people forming rational expectations
formed correct inflationary expectations, on average, and were not surprised
by the short-term blip in inflation between 1946 and 1948 caused by in¬
creased monetary growth.
Contrast this situation to the adaptive inflationary expectations shown in
part b of Figure 12.4. The adaptive expectations worked well when inflation
was stable prior to 1946, but missed the upward blip in inflation between
1946 and 1948 and also overreacted from 1948 and beyond. In effect, people
forming adaptive expectations were surprised by the blip in inflation and
were surprised again when inflation returned to its previous pattern. Notice
that between 1946 and 1948 people systematically underestimated the rate
of inflation, and from 1948 and beyond they systematically overestimated
it. Once again we see that rational inflationary expectations have the advan¬
tage of eliminating these systematic errors—provided, of course, the expec¬
tations are based on the correct economic model. Box 12.2 shows what can
happen when this isn’t the case.
Our previous examples are based on the assumption that the rate of
growth in real output is zero on average, as it was during the period imme¬
diately following World War II. When the growth rate in real output is not
zero, the simple quantity theory implies the inflation rate is given by the
difference between the monetary growth rate and the rate of growth in real
output:
^ = 8m ~ gr¬
in this case, the rational expectation of inflation in period t based on the

simple quantity theory is
RF
~ gMt~ i - 8yt- v
Let us see how those forming rational expectations can use this more general
framework to respond to changes in the structure of the economy.
The Importance of Using

the Correct Model
When Forming Rational Expectations
What happens if the economic model on which (a) Price Level and Lagged Price Level 1970 - 1993
rational expectations are based is incorrect? The
answer is that simpler expectations, like Markov
expectations, may perform better. To illustrate
this, we present two figures based on U.S. data
for the period 1970 through 1993. The first one
graphs the price level in a given period (Pt)
against the price level in the previous period
(Pt-iY, the second graphs the price level in a
given period against the money supply. The idea
is to see whether the past price level or the
money supply better explains the current price
level.
In the first figure, the current price level is
graphed as a function of the price level in the
previous period, as it will be under Markov ex¬
pectations. Notice the relation between these (b) Relationship Between the Money Supply and
two variables lines up almost on a diagonal line, the Price Level 1970- 1993
indicating that last period's price level does a
very good job of explaining the current price
level. Based on this time period, Markov expec¬
tations of the price level forecast the future
price level quite adequately.
In the second graph, the current price level
and the money supply do not line up so nicely,
indicating the money supply by itself is not as
good a predictor of the price level. This may
seem strange to you; after all, shouldn't rational
expectations of price increases incorporate all
relevant information about things that cause
Money Supply (M2, Billions $)
prices to rise, like the money supply? Of course
they should, but this information must be incor¬
porated in a way that is consistent with correct prices; it says that a 1 percent increase in the
economic theory. Even holding other factors money supply will lead to a 1 percent increase
(like real output) constant, the simple quantity in prices. The money supply is relevant for pre¬
theory does not predict that a $1 increase in the dicting future prices, but not in the way postu-
money supply will lead to a $1 increase in Continued on p. 388
388 C hapter 12 Rational Expectations and Efficient Markets
Continued from p. 387 the correct model or choose the wrong one,
lated in the graph. This is why the relationship in Markov expectations may lead to better fore¬
the second graph is less perfect than that in the casts than they would if you incorporated rele¬
first. vant information into the wrong model.
What should you make of all this? To form
rational expectations of future variables, you
Source for data: U.S. Department of Labor, Bureau of La¬
must not only incorporate all relevant informa¬
bor Statistics, The Consumer Price Index, various issues;
tion into the forecast but also make sure the ec¬ Board of Governors of the Federal Reserve System, Statis¬
onomic model into which you are incorporating tical Release and Federal Reserve Bulletin, various issues;
the information is correct. If you are unsure of and Citibase electronic database.
Between 1979 and 1982, real output growth in the United States was
sluggish, averaging zero. Things changed in 1983, when real output growth
increased to about 3 percent per year and hovered around there through
1989. The rate of growth in the money supply declined during this period
as well. Together these changes caused the inflation rate to decline from
about 9 percent prior to 1982 to about 4 percent thereafter. Figure 12.5
presents two stylized diagrams showing this change in the inflation rate,
along with the rational and adaptive inflationary expectations.
Based on the simple quantity theory, an increase in the growth in real
output, coupled with a decline in monetary growth, reduces the inflation
rate. Accordingly, part a of Figure 12.5 shows that those forming rational
expectations lowered their inflationary expectations in response to these
structural changes in the economy. The result is that rational expectations
of inflation were 9 percent through 1982 and then fell to 4 percent beginning
in 1983. In contrast, the adaptive expectations of inflation shown in part b
of Figure 12.5 worked well until 1983, when they began to systematically
overstate the actual inflation rate. These errors gradually diminished over
time, but even by 1987 people with adaptive expectations were still forming
forecasts of inflation that were too high. Thus, adaptive expectations adjust
gradually to structural changes in the economy, just as they do to a change
in the actual inflation rate. In contrast, rational expectations adjust quickly
to structural changes, in this case the increased growth in real output accom¬
panied by a decline in monetary growth.
Finally, it is important to stress that the rational inflationary expectations
examined in this section are based on the simple quantity theory. If the
underlying causes of inflation in the economy are consistent with the theory,
no systematic errors in inflationary expectations will occur. However, one
assumption underlying the simple quantity theory is that velocity is constant
or, at the very least, changes in velocity are unpredictable. If velocity is not
constant and in fact systematically rises or falls over time, there will be
systematic errors in inflationary expectations that are based on the simple
quantity theory. In effect, the errors are due to the fact that “rational”
expectations are being formed based on an incorrect economic model. Ra¬
tional expectations improve forecasts and lead to no systematic forecasting
errors when they are based on the correct economic model. Expectations
based on an incorrect theory, in contrast, may lead to systematic errors in
forecasts. Given the importance of the underlying economic theory in form¬
ing rational expectations of inflation, in the second half of this book we take
a closer look at the theoretical and empirical link among money, output, and
inflation.
Summary 1 Last period, real output increased by 4 percent and the money supply in¬
Exercise 12.1 creased by 5 percent. The inflation rate last period was 10 percent, (a) What
is the Markov expectation of inflation this period? (b) If the adjustment
factor is .3 and no forecast error occurred last period, what is the adaptive
expectation of this period’s inflation rate? (c) What is the rational expectation
of the inflation rate? (Use the simple quantity theory as the relevant economic
model.)
Answer: (a) The Markov expectation of inflation is uf = ttSince last

period inflation was 10 percent, the Markov expectation of inflation this
period is uf = .10, or 10 percent, (b) Since inflation last period was 10
percent and there were no errors in expectations, last period’s adaptive
expectation of inflation was tjf_! = .1. Given this and the fact that X = .3,
the adaptive expectation of inflation this period is
< = 1 + -3(^-1 “ <-i) = .1 + .3(.l - .1) = .1,

or 10 percent, (c) Given the simple quantity theory and information about
last period’s growth rates in money and real output, the rational expectation
of inflation this period is
RF
= gMt-i ~ §Yt-1 .05 - .04 = .01,
or 1 percent.
4 Critique of the Views of Expectations Formation

We conclude our discussion of expectations formation by pointing out some
of the common criticisms levied against Markov, adaptive, and rational
expectations. We do so with an analogy from American football to emphasize
that the criticisms are potentially valid for expectations formed outside the
normal realm of economics (although given the salaries of big-league play¬
ers, economics is relevant even to football).
Consider an American football game in which a team is allowed four
attempts (or downs) to gain 10 yards. If the team is unsuccessful, the op-
A Critique of the Views of Expectations Formation 391
ponent gets the ball. The usual play is to run or pass on the first three downs
and, if unsuccessful at advancing 10 yards, use the fourth down to punt (kick
the ball far down field so the opponent gets the ball far from your goal).
What would you expect a team to do on third down and five yards to go?
With Markov expectations, you would expect the team to do what they
did in the previous third-down-and-five situation; if they ran before, you
would expect them to run again. If you used adaptive expectations, you
would expect them to do what you thought they would do last time, but
adjust your forecast based on your error in expectations last time. If you
used rational expectations, you would form an expectation based on all
available information: the players on the field just prior to the snap of the
ball, their tendencies on previous downs in similar situations, the score, time
left in the game, and so on.
What if the rules were changed to only allow only three attempts to gain
10 yards? Blind use of Markov or adaptive expectations would ignore this
structural change in the rules of the game and form an expectation of the
third-down play based on past experience with third downs. However, with
such a rule change, past experience will be a very poor indicator of what the
team will do given only three downs. With rational expectations, you would
use your knowledge of the new rules of the game and conclude that a punt
is now more likely. This is analogous to using your knowledge of the rules
of the game in economics—economic theory—to sharpen your expectations
about future values of economic variables.
This example highlights the weaknesses of Markov and adaptive expec¬
tations we alluded to in earlier sections, but also allows us to discuss by
analogy some of the criticisms of rational expectations. One common criti¬
cism is that people do not know enough to form rational expectations. This
criticism is actually a two-pronged attack. The claim is that they either don’t
know the relevant economic theory or simply find the required calculations
too difficult. In our football example, this is analogous to saying that people
either don’t know the rules of football or just find it too difficult to figure
out the strategy a coach would use in a certain situation. Critics of rational
expectations would argue that the economy is certainly more complex than
a football game; it is more reasonable to think people don’t understand the
rules of economics than it is to think people don’t understand the rules of
American football (though both may be true for most people).
The response from those espousing rational expectations is that people
are willing to spend a lot of time learning the rules of the economy if doing
so is important to their well-being. This is why coaches spend time gathering
information about football—because it is valuable to them! Thus, those for
whom an accurate expectation is important will devote considerable time,
effort, and resources to learning the rules of the economy. This means not
that they will always be right but that they will not make systematic errors
in their forecasts. If they did make systematic errors, they could profit by
figuring out how to better form expectations.
392 Chapter 1 2 Rational Expectations and Efficient Markets
Another criticism of rational expectations is that people don’t gather the

necessary information. Defenders of rational expectations would argue that
those people who don’t gather the relevant information are the very ones
who don’t find it very important for their well-being. Again, those for whom
the information is important will expend the effort and resources to gather
it. As Box 12.1 showed, those who make a living forecasting inflation rates
aren’t systematically surprised by high or low inflation.
To critics of rational expectations, the clinching argument is that econ¬
omists themselves do not always agree on the relevant economic theory, so
how can noneconomists be expected to both know the relevant theory and
gather and process the needed information? It is this argument that is often
used to justify reliance on the more simplistic but more easily used adaptive
or Markov expectations.
Summary l (a) Suppose money growth, inflation, and the growth in real output remain
Exercise 12.2 constant for a long period of time. Would inflationary expectations depend
on which of the three methods were used to form expectations? (b) How
would your answer to part a change if money growth, output growth, and
inflation varied considerably over time? (c) Would your answer to part a
change if there was a more complicated economic reason for inflation than
the simple quantity theory suggests?
Answer: (a) If the economy started out in a situation where everyone’s

expectations were realized, each of the three methods would work equally
well at forecasting future rates of inflation. However, if initial expectations
were in error, adaptive expectations would only gradually lead to expecta¬
tions of inflation that equaled the actual rate. In contrast, Markov expecta¬
tions would adjust to the actual level of inflation after only one period of
error. Rational expectations would lead to expectations that on average
equaled the actual inflation rate, provided relevant information about past
monetary and real output growth were available and the corresponding model
of inflation was correct, (b) With considerable variation over time in inflation
rates, money growth, and real output growth, Markov and adaptive expec¬
tations would continually have systematic forecast errors. Properly formed
rational expectations, however, would be correct on average, with no sys¬
tematic errors, (c) With a more complicated model of inflation, additional
data would be needed to form rational expectations. If these data were not
available, or if the complicated link between inflation and other variables
could not be successfully modeled, rational expectations might perform more
poorly than adaptive or Markov expectations. Rational expectations are no
better than the validity of the model used to form expectations, whereas the
other two methods are no better than the stability of the economic variable
to be forecast.
Efficient Markets
If you have ever dreamed of making a fortune in the stock market, this
section is written just for you. Before you call your broker in an attempt to
live out your dreams, however, we encourage you to read what we have to
say about efficient markets—something related to but distinct from rational
expectations.
The Efficient Markets Hypothesis

The efficient markets hypothesis states that the current price of an asset
such as a share of stock reflects all available information about the value of
the asset. Consider Figure 12.6, which plots the price of a share of stock
between January 15, 1994, and January 15, 1995. Notice that the trend in
this stock price is upward; the stock started out at $10 in 1994 and trades
for $20 in 1995. A common temptation when reading the financial pages is
to extrapolate beyond the current date to infer that the price of the stock will
surely rise to $40 by next year. The extrapolation in Figure 12.6 is repre¬
sented by the dotted line.
The efficient markets hypothesis asserts that on January 15, 1995, all
available information about the value of the stock is contained in the price
on that date, which is $20. This implies, among other things, that nothing
can be gained by drawing a ‘Trend line” to forecast the future stock price.
To see why, simply notice that other investors also have access to a chart
such as the one in Figure 12.6. If people knew the stock would be selling
for $40 in 1996, do you really think it would be selling for $20 in 1995? If
so, you and other investors would seek to purchase the stock today to earn
a return of
d _ ^1996 F1995 _ $40 $20 _

l"6 “ P1995 _ $20
or 100 percent! But as you and other investors began purchasing the stock,
the price would rise, thus reducing your return. This process would continue
until the hefty return was no longer available.
More generally, according to the efficient markets hypothesis, the risk-
adjusted expected return on all investments will be equal; that is, the return
you should expect to earn on this stock exactly equals the return you could
earn on any other asset with similar risk characteristics. The reason is that
if one asset earns a higher risk-adjusted expected rate of return than another
asset, investors will quickly attempt to purchase that asset, driving up its
price and thus lowering its expected return. Recall from Chapter 10 that the
expectations hypothesis of the term structure of interest rates postulates that
the expected return from holding long-term bonds equals the average ex-
394 Cha pter 12 Rational Expectations and Efficient Markets
Figure 12.6
■
A Bad Method of Forecasting Stock Prices
If you are aware that a particular stock has increased in price from $ 10 to $20 over the past
year, you might be tempted to expect this price to continue to rise to $40 next year. However,
this extrapolation is generally a bad forecasting method, since it ignores economic determi¬
nants of stock prices such as taxes, interest rates, firm earnings, and so on. If markets are
efficient, the price of any stock will be driven to an equilibrium level such that its return
equals the return on other assets of similar risk. No one will be able to make money in the
stock market by extrapolating or charting points using the method shown in this figure.
Stock
Price ($)
_ _.4
40 P
X
Extrapolation .
V
20
10
Jan. 15, 1994 Jan. 15, 1995 Jan. 15, 1996

(Current Date) (Future Date)
pected return from holding short-term bonds during the life of the long-term
bonds. This is a form of the efficient markets hypothesis.
The efficient markets hypothesis has profound implications for your
ability to make a killing in the stock market—or in any other financial
market, for that matter. If the risk-adjusted returns on all financial assets are
equal, you cannot systematically beat the market by picking a stock you
think will outperform the market. All information that is relevant for fore¬
casting the future returns on the stock are already reflected in its price. Of
course, you might get lucky and pick a stock that does better than average,
but you are equally likely to pick one that does worse than average. In
general, you will earn the same return, and at lower risk, by purchasing a
diversified portfolio of many stocks so that stocks that perform better than
average offset those that do worse than average.
How should you determine which stocks to put in your portfolio? Believe
it or not, research suggests that picking stocks by throwing darts at the
financial pages does as well as spending lots of time (or money) researching
particular stocks! Since the stock market is efficient, the only differences in
returns on different stocks are due to risk, so riskier stocks earn higher
returns than less risky ones. Which stocks you pick is less important than
picking enough that you benefit from diversification, as we saw in Chap¬
ter 9.
For small investors like university professors, a more economical method
of diversifying risk is to purchase shares in a diversified no-load mutual
fund. These funds do not charge fees for buying or selling shares in their
funds, and they pool the funds of many investors to purchase shares of
hundreds of stocks. In contrast, broker commissions would bankrupt small
investors who attempted to purchase single shares of hundreds of different
stocks directly.
Rational Expectations and Efficient Markets

The efficient markets hypothesis is closely related to the notion of rational
expectations. In fact, for a market to be efficient, expectations must be based
on all known relevant information, which rational expectations require. Thus,
rational expectations are necessary for efficient markets. Let us see why.
Consider again Figure 12.6. You want to forecast the price at which the
stock will trade in 1996, but all you know is the current price and the past
trend. Is this the only information you need to form an expectation of the
future price? Clearly not. The price of the stock in 1996 will depend on a
number of variables, including the earnings of the issuing firm, tax policy,
interest rates, and the like. A forecast based on a trend line ignores all of
this information.
Participants in the stock market who invest millions of dollars in stocks,
such as pension funds and insurance companies, have a tremendous incentive
to form accurate estimates of the company’s future earnings. They also are
likely to spend considerable time and money learning about government
legislation that would affect the future price of the stock, the quality of the
firm’s management, and the like. An expectation of the stock price based on
a simple trend line is a very poor substitute for this information.
Fortunately, however, if the market is efficient, the current stock price
will already reflect this information. Those who invested considerable sums
to gather the information they needed to form a rational expectation of the
future stock price will, based on their information, buy the stock when it is
underpriced and sell it when it is overpriced. In equilibrium, the price of the
stock will rise or fall to reflect the relative demand for and supply of the
stock given available information. You can hardly beat the professionals by
drawing a line to connect the dots in Figure 12.6!
The idea that the risk-adjusted expected returns on alternative assets are
equalized in equilibrium is based on the notion that market pressures will
ensure that any temporary deviation from this equilibrium is corrected. These
market pressures are based on the expectations of those who use all available
information about the value of financial assets and continually revise their
buy-and-sell decisions when new information becomes available. For in¬
stance, when people who make their living forecasting inflation expect it to
rise, they will immediately attempt to sell off their bond portfolios. This
leads to a very quick rise in interest rates, and this new information is quickly
reflected in lower bond prices. By the time the information trickles down to
you or you learn inflation is higher, bond prices already reflect this infor¬
mation. Similarly, stock prices tend to change very quickly in response to
new information, and deviations from equilibrium are typically corrected in
seconds or minutes rather than days or weeks.
Versions of the Efficient Markets Hypothesis

In its most general form, the idea of efficient markets is that asset prices
tend to change very quickly in response to new information and thus the
risk-adjusted returns on assets tend to be equal. Some economists believe in
more specific forms of efficient markets: the weak form, the semi-strong
form, and the strong form of market efficiency. As their names imply, these
three statements of market efficiency make increasingly strong assumptions
about asset market pricing.
Weak-Form Market Efficiency. Weak-form market efficiency

claims the best predictor of next period’s asset price is this period’s price.
No other past information on the asset price can improve on this prediction.
One implication of weak-form market efficiency is that the so-called tech¬
nical analysis of asset prices—charting or plotting historical data on asset
prices—cannot help predict future values of the asset price. As Box 12.3
shows, it is difficult at the very least to predict returns on stocks or Treasury
bills by looking at historical data.
There is some evidence of minor departures from weak-form market
efficiency in stock markets. For instance, a close look at stock prices some¬
times reveals day-of-the-week effects (stock prices tend to rise on Monday
and fall on Friday), time-of-year effects (stock prices tend to rise in January),
and small-firm effects (small-firm prices typically rise by more than large-
firm prices). But little evidence exists that average investors can use these
effects to earn above-normal profits once transactions costs such as broker
fees are taken into account.
The Random Nature of Stock Returns,

Interest Rates, and Inflation
Think you can make money predicting changes and the inflation rate were usually fairly close to
in stock prices or interest rates? If financial mar¬ each other, although during the middle 1980s,
kets are efficient, stock returns and interest the Treasury bill rate was substantially above the
rates will fluctuate randomly, with no discernible inflation rate. This indicates Treasury bills have
trends in their movements over time. To check mostly stayed just at or slightly above the infla¬
whether this is the case, the accompanying fig¬ tion rate, but in the mid-1980s investors could
ure plots interest rates and returns on stocks for earn a substantial return above the inflation rate
the period 1971-1992. The stock returns are by investing in T-bills. More important, notice
based on a stock portfolio that mimics the Stan¬ there were no discernible trends in interest
dard and Poor's 500 Index and includes price rates; they went up about as often as they went
appreciation and dividends paid by the compa¬ down, with increases and decreases randomly
nies. The interest rate is that on one-year Treas¬ dispersed over the period.
ury bills, a measure of the risk-free interest rate. The return on stocks was considerably more
The figure also plots the inflation rate over this volatile, but also much greater on average than
period to show whether the returns were suffi¬ the risk-free interest rate and the rate of infla¬
cient to offset increases in the cost of living dur¬ tion. While stock returns were negative at times,
ing this time. especially in the middle 1970s, the positive re¬
There are several things to notice in this turns far outweighed the negative returns. No-
graph. First, the interest rate on Treasury bills Continued on p. 398
60 i
'85 '90
398 C hapter 12 Rational Expectations and Efficient Markets
Continued from p. 397 stock market exceeded the return on risk-free

tice too that some of the positive (and negative) investments like Treasury bills; (2) stock returns
returns were quite sizable. The largest positive were much more volatile, with periods of sub¬
return was 50 percent in early 1983, with nu¬ stantial negative returns interspersed among
merous 30 percent returns randomly sprinkled periods of substantial positive returns; and (3)
throughout the 21-year span. There were also there was no discernible trend in stock returns
periods of negative returns, the most sizable or interest rate movements.
one in early 1975. The higher expected returns
on stocks merely compensated investors for the
additional risk arising from the returns' higher
volatility; no predictable patterns in the returns
serve System, Federal Reserve Bulletin, various issues;
on stocks occurred over this period. Standard and Poor, The Outlook, various issues; Bureau
The main conclusion you should draw from of Labor Statistics, The Consumer Price Index, various is¬
this graph is that (1) on average, returns on the sues; and Citibase electronic database.
Semistrong-Form Market Efficiency. Semistrong-form market

efficiency asserts that no publicly available information will help predict
future asset prices or returns better than the last known value of the asset.
This includes not only past prices of the asset in question but also publicly
available information on other asset prices, interest rates, profits, tax rates,
business opportunities, and so on. Again the idea is that any useful publicly
available information will be so quickly reflected in the market price that
the current price is the best predictor of future prices.
However, this form of market efficiency leaves open the possibility that
private information can be useful in predicting future asset prices. For in¬
stance, suppose you are an accountant at a major firm and, during an audit,
learn that XXX Corporation understated its profits last year by $10 billion
due to a bookkeeping error. When this information becomes public, the stock
price will rise. In the meantime, you could buy shares of XXX Corporation
and make a profit on your inside information. Your private information
allows you to predict the future stock price better than the market can. This
said, we note that it is illegal to act on this type of inside information. If you
gain private information while in a fiduciary relationship, you cannot legally
act on it, even though the information would allow you to form a better
forecast of the future stock price than the market could.
The primary reason insider trading is illegal is that asymmetric infor¬
mation can destroy financial markets. Insiders have better information than
outsiders do, and in a market dominated by insiders, outsiders stand only to
lose by participating in the market.
To see why, suppose insiders know a stock is really worth $100 per
share. As an outsider, you don’t know for sure what the stock is worth.
Should you purchase the stock? Insiders will sell you shares only if you
offer them more than $100. The only way you can buy the stock is by paying
more than it is really worth, so you are better off staying out of the market
altogether. If you and other investors fear the market is dominated by insid¬
ers, the market could collapse, leaving only those with the “best” inside
information to buy and sell stocks. This reduces the number of market
participants, decreases the liquidity of assets traded in the market, and po¬
tentially ruins the market for everyone. Laws against insider trading are
designed to alleviate these problems.
Strong-Form Market Efficiency. Strong-form market efficiency

makes a very strong statement: that no current information, either publicly
available or not, can improve on using the most recently known value of an
asset price to predict the future value of that asset price. Thus, the strong-
form version basically claims that even insider information cannot aid in
predicting stock price movements.
There is evidence that the stock market does not meet the stringent
conditions of strong-form market efficiency. For instance, Ivan Boesky
earned millions of dollars year after year by trading on inside information
about corporate takeovers, at least until the Securities and Exchange Com¬
mission (SEC) charged him with trading on such information. While Boesky
ultimately paid $100 million in fines and was sentenced to three years in
jail, his ability to “beat the market” for many years by using insider infor¬
mation makes strong-form market efficiency suspect. The weak and semi¬
strong forms of market efficiency, in contrast, are largely consistent with
what one observes in most financial markets.
Efficient Markets and Expectations

One interesting feature of these last three hypotheses about market efficiency
is that they insist on a version of Markov expectations for predicting asset
prices. This is an example of a situation in which Markov expectations are
rational expectations! In this case, the relevant economic theory of efficient
markets implies that all information that is useful for predicting movements
in asset prices are already contained in the current price. Any changes in the
asset price will be purely random, so the rational expectation of tomorrow’s
asset price is today’s price—its Markov expectation. Thus, we looked up
the price of Microsoft stock in the paper on June 21, 1994, and saw the price
was $53.50. Our best forecast of the price of Microsoft stock for July 3 was
$53.50.
Even though this was our best forecast, it was still subject to a lot of
uncertainty; that is, while we say $53.50 was our best forecast, we can also
say there was a large chance that the price would differ from this number
and that the probability that the actual value would differ from our forecast
would increase over time. However, none of this contradicts the claim that
$53.50 was the best forecast we could make. As with meteorologists, the
best forecast we can make may be subject to a lot of uncertainty, especially
as we forecast far into the future.
Summary Suppose two bonds with identical risks and terms to maturity have different
Exercise 12.3 yields, (a) If the market is efficient, what would you expect to happen to the
yields and prices of these bonds? (b) If you observe two apparently identical
bonds with different yields, should you buy the one with the higher yield?
Answer: (a) If the market is efficient, the prices of the bonds will adjust
very quickly to equalize the bonds’ yields. The view that the bonds have
identical risks must, of course, be based on rational expectations, that is, all
available information about the riskiness of each bond, (b) If you are con¬
fident the bonds are really identical, of course you should purchase the one
with the higher yield. But chances are that by the time you call your broker,
some other investor who makes his or her living looking for such profit
opportunities will have already driven up its price and thus eliminated the
higher yield. It is also possible—and arguably more likely—that the bonds
are really not identical. The higher yield may reflect information by insiders
that the bond involves higher risk. If this is the case, you can rest assured
that the risk-adjusted return on this bond will be the same as those on other
bonds.
Conclusion
In this chapter, we examined alternative methods of forming expectations

and noted the pros and cons of each. Properly formed rational expectations
contain no systematic errors and are correct on average, but they usually
require more information than Markov or adaptive expectations do. We also
saw that Markov expectations are sometimes exactly the expectations that
economic theory suggests are rational expectations. Finally, we looked at
the concept of efficient markets and its implications for asset prices.
The material in this chapter, coupled with that presented in the first 11
chapters, has given you a firm grasp of the workings of the specific financial
markets in which banks, pension funds, insurance companies, and individuals
such as yourself borrow and lend loanable funds. In the next half of the
book, we use what we have learned about these specific markets and rational
expectations to analyze macroeconomics and policy—the study of how all
economic markets, aggregated together, are affected by government and Fed
actions.
KEY TERMS
Markov expectations hypothesis rational inflationary expectations

Markov inflationary expectations efficient markets hypothesis
adaptive expectations approach weak-form market efficiency
speed of adjustment technical analysis
adaptive inflationary expectations semistrong-form market efficiency
rational expectations hypothesis strong-form market efficiency

1. Provide, at an intuitive level, an explana¬ 4. Last year, both real output and the money
tion of Markov, adaptive, and rational ex¬ supply increased by 6 percent. The infla¬
pectations. tion rate was 10 percent. What would you
expect the inflation rate to be this year?
2. Determine whether the following exam¬
Why?
ples most closely conform to Markov,
adaptive, or rational expectations. 5. What are the primary advantages of Mar¬
kov expectation formation? What are the
(a) . When Sue wakes up in the morning, primary disadvantages (i.e., when do they
she expects her husband to fix the same lead to systematic forecast errors)?
breakfast he made yesterday.
(b) . Sam always expects his wife to fix 6. What are the primary advantages of adap¬
pancakes for breakfast. tive expectation formation? What are the
(c) . Joe bases his expectation of dinner on primary disadvantages (i.e., when do they
what he observed thawing out in the sink lead to systematic forecast errors)?
when he left for work. 7. What are the primary advantages of ra¬
3. Last year, both real output and the money tional expectation formation? What are the
supply increased by 6 percent. The infla¬ primary disadvantages?
tion rate was 10 percent. 8. What is the efficient market hypothesis?
What does it imply about using past
(a) . What is the Markov expectation of
trends to “beat the market?”
inflation this year?
(b) . If the adjustment factor is 1, what is 9. An investment banker forms Markov ex¬
the adaptive expectation of this year’s in¬ pectations of inflation for period t based
flation rate? on information available in period t — 1.
(c) . If the adjustment factor is .5 and last Complete the following table for the
year’s forecast error was 6 percent, what banker. Are the expectations ever correct
is this year’s adaptive expectation of infla¬ in periods 2 through 10? Why or why
tion? not?
Actual Expected \ = .5. Complete the following table. Are

Inflation Inflation Forecast the banker’s expectations ever correct in
Period Rate Rate Error periods 2 through 10? Why or why not?
(0 (n,) W) O, - TT(
Actual Expected
1 1% 1% 0 Inflation Inflation Forecast
2 2 Period Rate Rate Error
3 3 (t) (n t) On?) On, - n{
4 4 1 1% 1% 0
5 5 2 2
6 4 3 3
7 3 4 4
8 2 5 5
9 1 6 4
10 0 7 3
.
10 Suppose the investment banker in problem 8 2
9 forms adaptive expectations of inflation 9 1
and the speed of adjustment parameter is 10 0
.
11 Now suppose the investment banker in the following table. Are the banker’s ex¬
problems 9 and 10 forms rational expecta- pectations ever correct in periods 2
tions of inflation based on the simple through 10? Why or why not?
quantity theory, tt = gM — gY. Complete
Money Output Actual Rational

Growth Growth Inflation Expectation Forecast
Period Rate Rate Rate of Inflation Error
(0 (gMt) (gyt) (n,) (n? - tt,
1 2% 0% 1% 1% 0%
2 3 0 2
3 4 0 3
4 5 0 4
5 6 2 5
6 7 4 4
7 8 6 3
8 9 8 2
9 10 10 1
10 11 12 0
12. Complete the following table to illustrate 13. Can it ever be “rational” to use some¬
how the speed of adjustment affects the thing other than rational expectations to
formation of adaptive expectations. De¬ form future expectations? Explain care¬
scribe verbally the difference in adaptive fully.
expectations when X = .2 compared to
14. Suppose you wanted to form rational ex¬
the case where X = .8.
pectations of future short-term interest
Adaptive Adaptive rates. What variables would economic
Expectation Expectation theory suggest are relevant for this under¬
Actual Inflation Inflation taking?
Inflation Rate Rate
Period Rate «) (O 15. Suppose you used your model in problem
(0 fa*) When X = .2 When X = .8 14 to predict future interest rates. Do you
think you could make money by using
1 0% 0% 0%
this forecast in financial markets? Explain.
2 5
3 5
4 5
5 5
6 5
7 5
8 5
9 5
10 5
Balvers, R. J., and T. F. Cosimano. “Actively Learn¬ Cams, F., and R. E. Lombra. “Rational Expectations
ing about Demand and the Dynamics of Price Ad¬ and Short-Run Neutrality.” Review of Economics
justment.” Economic Journal, 100 (September and Statistics, (November 1983), 639-643.
1990), 882-898. Cecchetti, S. G., R. E. Cumby, and S. Figlewski. “Es¬
Balvers, R. J., T. F. Cosimano, and B. McDonald. timation of the Optimal Futures Hedge.” Review
“Predicting Stock Returns in an Efficient Market.” of Economics and Statistics, 70 (November 1988),
Journal of Finance, 45 (September 1990), 1109— 623-630.
1128. Cho, D. W. “Formation of Inflationary Expectations
Blenman, L. P. “Price Forecasts and Interest Rate by Business Economists.” Business Economics, 21
Forecasts: An Extension of Levy’s Hypothesis.” (April 1986), 34-39.
Journal of Futures Markets, 10 (December 1990), Chung, Pham. “A Note on Policy Evaluation and Ra¬
605-610. tional Expectations.” Public Finance, 41 (1986),
Bonham, C. S., and D. C. Dacy. “In Search of a 139-146.
‘Strictly Rational’ Forecast.” Review of Economics Darrat, A. F., and F. A. Lopez. “Price Instability and
and Statistics, 73 (May 1991), 245-253. Inflation: Some Tests Based on Rational Expecta¬
Brandon, C., R. Fritz, and J. Xander. “Econometric tions Models.” Eonomic Letters, 26 (1988), 111-
Forecasts: Evaluation and Revision.” Applied Eco¬ 119.
nomics, 15 (April 1983), 187-201. Continued on p. 404
Driskill, R. and S. McCafferty. “Spot and Forward Performance.” Journal of Futures Markets, 8
Rates in a Stochastic Model of the Foreign (April 1988), 167-184.
Exchange Market.” Journal of International Eco¬ Mankiw, N. G., J. A. Miron, and D. N. Weil. “The
nomics, 12 (May 1982), 313-331. Adjustment of Expectations to a Change in Re¬
Driskill, R., and S. McCafferty. “Exchange Market gime: A Study of the Founding of the Federal Re¬
Intervention under Rational Expectations with Im¬ serve.” American Economic Review, 77 (June
perfect Capital Substitutability.” Exchange Rate 1987), 358-374.
Management under Uncertainty. Cambridge, Pearce, D. K. “Short-Term Inflation Expectations:
Mass., and London: MIT Press, 1985, 83-95. Evidence from a Monthly Survey: A Note.” Jour¬
Hall, A., and R. J. Rossana. “Estimating the Speed of nal of Money, Credit, and Banking, 19 (August
Adjustment in Partial Adjustment Models.” Jour¬ 1987), 388-395.
nal of Business and Economic Statistics, 9 (Octo¬ Wible, J. R. “An Epistemic Critique of Rational Ex¬
ber 1991), 451-453. pectations and the Neoclassical Macroeconomic
Joutz, F. L. “Information Efficiency Tests of Quar¬ Research Program.” Journal of Post Keynesian
terly Macroeconomic GNP Forecasts from 1976 to Economics, 7 (Winter 1984-85), 269-281.
1985.” Managerial and Decision Economics, 9 Wrightsman, D. “Forecasting with Velocity.” Chal¬
(December 1988), 311-330. lenge, 28 (July/August 1985), 58-60.
MacDonald, S. S., R. L. Peterson, and T. W. Koch.
“Using Futures to Improve Treasury Bill Portfolio
PART FOUR
The Creation of Money

CHAPTERS
13
The Federal Reserve System: History,
Modern Structure, and Policy Tools
14
The Money Supply Process
15
The Demand for Money and Equilibrium
in the Money Market
CHAPTER
The Federal Reserve System:

V
History, Modern Structure,

and Policy Tools
February 4, 1994, Fed chair Alan Greenspan released a written

statement to the press saying the Fed would tighten credit to keep inflationary
pressures in check. This news shook financial markets. Interest rates rose,
bond prices fell, and the stock market plummeted, posting its largest daily
decline in almost two years. The Fed’s announcement marked the end of the
steady decline in interest rates that began with Fed actions in 1989. These
events highlight the Fed’s power over the economy.
In the first part of this book, we focused largely on how financial markets
function and how they are affected by changes in interest rates, expectations,
and various types of risk. In the second half, we examine the overall func¬
tioning of the economy and in particular how actions by the Fed influence
it. Why does the Fed tighten credit or raise interest rates? Do the Fed’s
efforts to keep inflation in check affect other aspects of the economy like
investment, employment, or real output? We answer these and related ques¬
tions in the remaining chapters. Before we do this, however, we need a broad
overview of why and how the Fed functions as a central bank, how the Fed
is structured, and the tools it uses to control the money supply.
We begin this chapter with a close look at the evolution of the central
bank of the United States, the Federal Reserve System. In contrast to Chapter
7, in which we focused on how the Fed and other regulatory bodies like the
FDIC regulate the banking industry, here we examine the Fed’s role as the
central banker and regulator of the money supply. Our look at the structure
of today’s Fed, both as central banker for the United States and as a quasi¬
independent policymaker, introduces you to the tools the Fed uses to influ¬
ence the economic landscape that extends far beyond the banking system.
In later chapters, we will see how the Fed uses these tools to affect the
money supply (Chapter 14), interest rates (Chapter 15), and the overall
economy, including inflation, unemployment, investment, and real GDP
(Chapters 16 through 18).
406
A Brief History of Central Banking in the United States
A central bank is a bank for the government and a bank for banks. A central
bank holds deposits for and loans reserves to private commercial banks like
Citibank. It also holds deposits for the government and takes actions that
facilitate government borrowing. In its role as a bank for banks and govern¬
ment, the Fed also regulates the money supply—the topic of the next chapter.
Throughout its history, the United States has had three central banks:
the First Bank of the United States, the Second Bank of the United States,
and the Federal Reserve system (the current central bank). As we will see,
there have also been several periods during the past 200 years in which no
central bank existed. In this section, we briefly review the history of banking
and central banking in the United States.
Banking in the United States Before 1791

Before 1791, the United States had no central bank and relatively few
commercial banks. The first institution that functioned like a modern com¬
mercial bank was established in Philadelphia in 1782. The Bank of North
America was a private bank created primarily by merchants, who used it to
pool their funds and provide short-term financing for commercial transac¬
tions. No central bank existed at this time, nor even a banking industry as
such. The Revolutionary War had just ended, and the federal government
had not yet formed.
In these early days, banking developed as an activity carried out by
private banks with state charters. These banks were primarily in the business
of making short-term commercial loans, which financed purchases of goods
and were liquidated when the goods were sold.
One key difference between banking then and banking today was that
the issuance of banknotes—that is, the issuance of currency by private
banks—was the primary means by which early banks made loans. Today,
when a bank makes a loan, it commonly gives the borrower a checking
account containing the loaned amount or makes some equivalent transfer of
funds. With a mortgage loan, for example, a bank might provide the loaned
funds directly to the seller of the house, again in the form of a checkable
account or a check written against such an account. Rarely does a bank today
make a loan by actually providing a borrower with a substantial sum of
currency.
The use of banknotes instead of checkable deposits in the early years of
the United States can be explained by several features of that period. First,
the acceptability of checks in trade requires facilities for quickly clearing
checks among banks. These facilities were not readily available in a nation
with a quickly expanding frontier and poorly developed transportation and
communication routes. Second, the acceptability of checks requires confi-
408 Chapter 13 The Federal Reserve System
dence in the honesty of people writing the checks or an efficient means of

policing the use of bad checks. These elements too were absent in the early
years of the United States. Thus, banknotes, usually redeemable in gold at
the bank of issue, were a superior form of money during that period. As
long as the issuing bank was sound, the holder of a banknote was certain of
redemption in gold. These notes passed from hand to hand, serving as money
in various transactions, before eventually being returned to the issuing banks
for redemption in gold.
There was a problem, however, in maintaining banknotes’ convertibility
into gold. First, banks were tempted to overissue notes, since this was their
way of making loans and hence profits. The result of overissue was that
these banks could not meet the promise to redeem notes for gold on demand.
Second, the fractional reserve banking system meant the gold reserves banks
held did not equal the value of notes in circulation. (This occurs even today,
although modem banks hold reserves as vault cash and deposits at the Fed
instead of gold). Fear of inconvertibility caused bank runs. In the early United
States, this meant noteholders would rush to the bank in an attempt to redeem
their notes for gold before the bank exhausted its reserves. These problems
ultimately led to the development of the first central bank in the United
States.
The First U.S. Central Bank:

First Bank of the United States
The first institution to perform some of the roles of a central bank was the
First Bank of the United States, chartered by Congress in 1791. The First
Bank of the United States acted as the bank for the U.S. government, with
a main office in Philadelphia and branches in the major cities of the day:
New York, Boston, Baltimore, Washington, Norfolk, Charleston, Savannah,
and New Orleans. The government owned stock in the bank, but British
investors owned a greater than two-thirds share of the bank. This foreign
ownership disturbed many Americans, but the bank nonetheless acted as
fiscal agent for the federal government and in particular for the U.S. Treas¬
ury. The First Bank of the United States issued its own banknotes, which
circulated side by side with banknotes issued by a number of state-chartered
banks. The First Bank of the United States not only issued its own notes; it
also played a part in discouraging excessive note issue by state banks, since
it forced state-chartered banks to redeem their notes in gold by promptly
presenting any such notes in its possession to the issuing banks. If these
banks refused, the First Bank of the United States refused to accept their
banknotes for U.S. Treasury transactions.
These actions were not popular among state banks. The political oppo¬
sition they created, along with concerns about the constitutionality of a
federal charter, led to the early demise of the First Bank of the United States.
Its original charter expired in 1811 and was not renewed. The first U.S.
experience with central banking was thus short-lived.
The Second U.S. Central Bank:

Second Bank of the United States
After the demise of the First Bank of the United States, banking was left for
a time in the hands of state-chartered banks. By 1814, banks across the nation
were refusing to redeem their notes for gold. This contributed to the creation
of the Second Bank of the United States in 1816. This bank too served as
fiscal agent for the U.S. government. It issued its own notes and, like the
First Bank of the United States, forced state banks to redeem their banknotes
in gold by presenting any such notes it received directly to the issuing banks
for redemption. This bank also encountered severe political opposition from
state banks, and it too was attacked on constitutional grounds. In the early
1830s, President Andrew Jackson took actions that substantially weakened
the bank’s powers, and its charter was not renewed when it expired in 1836.
The Absence of a Central Bank: 1836-1913

After the demise of Second Bank of the United States, the United States
lacked a central bank until the creation of the Federal Reserve System in
1913. Between 1836 and 1913, the United States experimented with two
major banking systems: the free banking system and the national banking
system.
The Free Banking System. After 1836, banking in the United States,
was once again a business left to the states, which could charter banks and
decide how to regulate them. The free banking system was an innovation
that began in Michigan in 1837 and in New York in 1838. The free banking
movement had two main goals. The first was to make entry into the banking
industry “free,” or at least less heavily regulated than before. Prior to free
banking, the procedure for obtaining a state charter was complicated and
highly politicized, requiring action by the state legislature. The second goal
was to make banknotes safe. This was attempted by requiring collateral for
note issues. Under the free banking system, anyone could open a bank and
issue notes as long as that person provided the requisite collateral. This
commonly meant a bank had to deposit with the state government bonds
equal in value to the notes issued. Furthermore, banks had to redeem notes
for gold on demand. Failure to do so resulted in the closure of the bank,
with the sale of the bonds held as collateral to compensate the noteholders.
In addition, most states stipulated that if the collateral was not sufficient,
noteholders had first claim on the bank’s remaining assets. Finally, some
states made bank owners legally liable to noteholders to the extent of their
ownership share in the bank. The idea was to make banknotes a safe form
of money. Even in the case of bank failure, sufficient assets had to be

deposited with the state to redeem the notes of the failed bank.
Free banking spread rapidly among the states, although it was not uni¬
versally adopted. The first goal of the free banking system was successfully
obtained: The number of banks soared. In New York, for example, the
number of banks almost doubled within three years, and other free banking
states saw similar rapid increases.
The second objective, to provide a safe issue of banknotes, was much
less successful, and the reasons are a matter of current debate among econ¬
omists. Conventional wisdom has it that the free banking system was marked
by fraud, wildcat banking, bank failures, and widespread losses to notehold¬
ers. The term wildcat bank was coined during this period to describe a bank
that would purposely issue excessive notes and locate in remote places (near
the ‘ ‘wildcats”) to make redemption difficult. If substantial numbers of notes
were presented for payment, the bank would close up and disappear, leaving
noteholders with notes of dubious value at best. Conventional wisdom is
supported by personal anecdotes and by a look at some statistics on bank
failures.1 For example, Michigan, a pioneer of free banking in 1837, saw 40
banks established in the first year, but only 4 remained by the end of 1839.
Furthermore, some estimates place the losses to noteholders as high as 45
percent of Michigan’s annual income! Minnesota had a similar experience.
Minnesota allowed free banking in 1858; 16 new banks were formed, but
only 5 survived through 1863. In addition, in seven of the bank closings,
noteholders received 35 cents or less on the dollar.
Other states had more pleasant experiences, but still saw a number of
bank failures. In Indiana, 57 out of 72 free banks went out of business within
two years, although at least 14 managed to fully compensate noteholders.
Wisconsin saw 79 out of 140 free banks go out of business by the early
1860s, and in 37 of these failures noteholders experienced losses. Even in
New York, one-fourth of the free banks established in the first year were out
of business by 1841, and noteholders experienced losses in about half of
those cases.
The conclusion from such statistics and from personal anecdotes in the
historical record is that the free banking era failed to maintain a safe banking
system and a sound currency. The failure of free banking is often blamed
on fraudulent practices and wildcatting. Counterfeiting also played a role. It
was perhaps easier to engage in counterfeiting due to the large number of
different notes in circulation; by the end of the 1860s, almost 5,400 types of
counterfeit notes were in circulation. Finally, as Box 13.1 shows, some
problems during the free banking era can be attributed to volatile bond prices
and to a lack of diversity in the portfolios of the free banks.
1 Statistics from Arthur J. Rolnick and Warren E. Weber, “Free Banking, Wildcat Banking, and
Shinplasters,” Federal Reserve Bank of Minneapolis Quarterly Review (Fall 1982), 10-19.
Declining Bond Prices and Bank Failures

During the Free Banking Era
Arthur Rolnick and Warren Weber, economists bank owners would lose the value of their in¬
at the Federal Reserve Bank of Minneapolis, vestment in the bank—but nothing more—and
have provided evidence that declining bond the noteholders would be partially compensated
prices, rather than fraud and theft, were the pri¬ from the sale of the bank's collateral. In this sit¬
mary cause of bank failures during the free uation, bank owners and noteholders shared in
banking era. Free banks were required to hold the loss caused by the decline in bond prices.
state bonds as collateral against their note is¬ Rolnick and Weber point out that during
sues. Some states valued bonds put up as collat¬ the period from 1852 to 1863, most bank clos¬
eral at face value, while others required bonds ings occurred during times when bond prices
to be valued at market value. In either case, a declined sharply. For example, bond prices fell
substantial fall in the market price of those more than 20 percent in the last half of 1854,
bonds made a bank's collateral worth less than dropped more than 15 percent in the middle of
its note issue. In that case, a bank run could be¬ 1857, and again fell more than 15 percent from
gin. At least some noteholders, aware of the de¬ late 1860 through the summer of 1861. Such
cline in bond prices—and hence in the collateral large declines in the value of collateral could
backing the bank notes—would attempt to re¬ well have placed banks in a situation that would
deem their bank notes for gold. At the same trigger noteholders to redeem their notes, and
time, if bond prices had fallen far enough that a banks in such a position could have opted for
bank's assets were less than its liabilities, bank closure.
owners would have to raise additional funds
from their personal wealth or elsewhere to be
able to fully redeem the notes. The alternative
was to refuse to redeem the notes, triggering a Source: Federal Reserve Bank of Minneapolis Quarterly
bank closure and the sale of the bonds held as Review, Fall 1982, pp. 10-19. Copyright© 1982 Federal
collateral to compensate the noteholders. The Reserve Bank of Minneapolis. Used with permission.
The National Banking System. During the Civil War, the U S.

government found itself in need of funds. These funds were provided in part
by the issue of a fiat currency called greenbacks (recall that fiat currency is
not backed by gold or any other commodity). Greenbacks and related U.S.
government notes made up about 75 percent of the money in circulation by
the end of the Civil War and had supplanted state banknotes as the primary
currency in circulation.
The experience of the free banking system and the problems of financing
the expenditures of the Civil War led Congress to establish a sound national
currency with passage of the National Banking Act of 1863. This act, along
with several subsequent revisions and extensions, established the national
banking system. The federal government again began regulating both the
banking system and the circulation of banknotes and imposed a 10 percent
tax on any bank or individual using state banknotes for transactions. As
discussed in Chapter 7, this policy did not drive state banks out of existence
as intended. Instead it led to the dual banking system of today because state
banks innovated by introducing demand deposits, a close substitute for bank¬
notes issued by national banks.
The national banking system, like the free banking system, had some
successes. National banknotes succeeded in that they provided a sound and
stable national currency. However, the national banking system was created
at a time when deposits were replacing banknotes as the main method of
holding money. In 1860 deposits and banknotes in circulation were roughly
equal in amount, but by 1880 deposits were more than three times larger
than bank notes. In a sense, the national banking system concentrated on
providing a sound and stable currency, but it paid too little attention to
deposits.
The main criticism of the national banking system was that it did not
prove flexible in dealing with the seasonal variations in business activity,
especially in agriculture. Periodic increases in the demand for loans and in
the holding of currency occurred during the spring planting season and
especially during the fall harvesting periods.
For example, rural banks held not only their legal reserves but also a
portion of their extra reserves in correspondent banks in reserve cities. These
reserve city banks treated these deposits as they did all other deposits,
holding the legally mandated level of reserves against them but otherwise
lending out a substantial portion. During periods of seasonal activity in
agriculture, demand for currency increased at the rural banks, because many
transactions were still made with currency, and demand for loans rose as
well. Rural banks found themselves in need of funds to lend and thus
withdrew the extra reserves they had deposited in reserve city banks. The
reserve city banks then had to recall loans to meet their own reserve re¬
quirements.
At times this situation became more than just an inconvenience. When
periods of seasonal demand coincided with periods of financial strain, the
call of bank loans in reserve cities was often unsuccessful. The loans called
in were actually called call loans and were overnight loans secured with
common stock as collateral. These loans could be called on demand. How¬
ever, if the common stock used as collateral had fallen in price, calling the
loan might not have yielded full repayment. Also, calling the loan forced
the sale of the stock, and if this happened on a large scale the result was an
additional decline in the price of the common stock, further reducing the
chance of repayment. Reserve city banks thus found themselves in a rather

precarious situation. They could either respond to the withdrawal of deposits
requested by rural banks and violate their reserve requirements or fail to
honor the request for deposits and trigger a bank run.
If bank runs spread to many banks, a banking panic resulted. There
would be a widespread demand for currency by the public, a demand the
banking system could not meet. The result would be restrictions on the
withdrawal of currency, called a suspension of payments. Such restrictions
occurred in 1857, 1861, 1873, 1893, and 1907. During these times, banks
remained open and processed checks much as usual, but they restricted
payment of currency in exchange for deposits. Some customers were able
to obtain small amounts of currency, or even receive currency in the usual
amounts for business purposes, while other customers found the restrictions
less flexible, and large-scale withdrawals were rarely permitted. (During this
period, currency could actually be purchased, but at a premium. In New
York City this premium was as high as 5 percent, but often less.)2
Other attempts were made to provide the demanded currency. The U.S.
Treasury tried to handle some of the seasonal demand by depositing funds
each autumn in certain chosen banks to give them additional funds to lend
or to meet the demand for currency. In response to specific episodes in 1873,
1893, and 1907, the U.S. Treasury made large purchases of bonds in an
attempt to disburse currency to the public. This action was a precursor of
open market operations, discussed later in this chapter.
The national banking system was unable to provide an elastic banking
system, one that could expand or contract currency and loans in response to
seasonal and other variations in the demand for currency and loans. In the
national banking system, no bank was able to issue additional notes in a
time of high seasonal demand for currency and reserves. Thus, support grew
for the creation of a central bank, one that would be able to supply reserves
and loans elastically. The result was the birth of the Federal Reserve System.
The Third U.S. Central Bank: The Federal Reserve System

The Federal Reserve System (or simply the Fed) was established as the
central bank for the United States by the Federal Reserve Act of 1913, during
the presidency of Woodrow Wilson, and began operations in 1914. Chapter
7 provided a quick overview of the Fed’s history; now we take a more
detailed look.
The Fed was established as a central bank to ensure that commercial
banks indeed kept reserves (which were required to be on deposit at the
central bank) sufficient to cover withdrawals by depositors. The Federal
Reserve System was to stand willing to meet systemwide increases in the
2 See Gerald P. Dwyer, Jr., and R. Alton Gilbert, “Bank Runs and Private Remedies,” Federal
Resen’e Bank of St. Louis Review (May/June 1989), 43-61.
demand for currency by supplying currency to the banking system in the

quantity desired. The Fed would make it possible for the public to always
be able to change its desired ratio of currency to demand deposits without
severely disrupting the banking system. In other words, one main goal in
establishing the Fed was to provide an elastic currency, a money supply
that could be easily converted from demand deposits to currency. Moreover,
it was expected that the very availability of currency on demand would quell
the fears that fueled bank runs or bank panics, especially those that became
widespread. Thus, the very provision of an elastic currency was expected to
vastly reduce or even eliminate the occasional bank panics that occurred
under the national banking system.
In addition to providing an elastic currency, the Federal Reserve would
serve as the nation’s central bank, regulating money markets in general and
having day-to-day regulatory oversight on some commercial banks, as well
as managing the nation’s money supply for macroeconomic policy purposes.
The Federal Reserve Act of 1913. The Federal Reserve Act of

1913 set up the Federal Reserve System as a set of 12 separate institutions,
the district Federal Reserve banks. The actions of these 12 separate entities
were to be coordinated by the Federal Reserve Board in Washington, D.C.
Each Federal Reserve bank was set up to administer to a specific geographic
area or district, with branch banks to aid in administering the large areas in
some districts.
The district Federal Reserve banks were established by the sale of stock
to commercial banks that became members of the Federal Reserve System.
Member banks were required to purchase stock in the Federal Reserve bank
equal to 3 percent of their own net worth. These stock certificates paid a
dividend restricted by law to be 6 percent or less. Earnings in excess of these
dividends were remitted to the federal government.
In terms of membership, all national banks—which meant all banks that
were part of the national banking system—were compelled to join the Fed¬
eral Reserve System or give up their national charter. However, state-
chartered banks were allowed to choose to be or not be members of the
Federal Reserve System. All member banks were required to hold reserves
at their district Federal Reserve bank or branch bank.
Benefits of membership included access to check-clearing services and
aid in times of stress. Member banks could get the Fed to loan them funds
using bills of exchange or promissory notes as collateral. This practice, called
discounting, was the forerunner of the modern discount window.
The original organization of each of the district Federal Reserve banks
was as follows. There were nine directors, six elected by the member banks.
Three of these directors were to be bankers, and three were to be business-
persons not in the banking business. The remaining three directors were
appointed by the Federal Reserve Board, and one of them was appointed as
chair of the board and executive officer of the bank.
The Federal Reserve Board itself was composed of seven members

appointed by the president and approved by the Senate. One member was
named governor and one deputy-governor. The secretary of the Treasury
and the comptroller of the currency were made members. Also established
was a group called the Federal Advisory Council, composed of one banker
from each district, with the role of offering advice to the board.
The basic aim in establishing the Fed—an idea that can be seen in its
structure—was that of cooperative regulation. The idea was that central
banking could be accomplished fairly automatically. Cooperation among the
Board of Governors, the Federal Reserve banks, and the member banks
would ensure that the tasks of the central bank were successfully accom¬
plished. Moreover, the gold standard of the time required fairly automatic
responses to economic events. The job of the central bank was to maintain
a constant ratio between gold reserves and the sum of deposits plus notes.
The other task of the Fed was to provide an elastic banking system.
Indeed, this was a primary motivation in establishing the Federal Reserve
System. Elasticity was to be achieved by following what has been called the
real bills doctrine. According to this doctrine, the central bank would supply
liquid reserves to the member banks whenever they needed funds to make
productive business loans. These loans from the Fed to member banks were
temporary in nature and were made only if the banks could show the funds
would be used to increase loans to productive businesses. These loans were
to be discount loans, in which the collateral was various forms of short-term,
high-quality commercial paper.3 They were discount loans because the Fed
would purchase the commercial paper at a discount from its face value and
the member bank would buy back the paper at face value. When business
activity increased and there was a greater demand for liquidity as evidenced
by an increase in commercial paper and bills of exchange, the central bank
would supply the liquidity, lending notes and reserves to member banks that
presented these real bills as collateral.
The creators of the Fed intended that reserves would be created by
discounting real bills. Such actions would produce short-term increases in
reserves that would end when the discount loan was repaid by the borrowing
bank. However, during the 1920s the district banks found another way to
increase banking system reserves: by purchasing government securities for
their own portfolios. This method was discovered by accident, as the banks’
intent was to increase earnings on otherwise idle balances. (This was a
forerunner of open market operations, which are discussed in detail later in
the chapter.) To coordinate this activity, the Conference of Governors, which
included the governors of the 12 district banks, formed the Open Market
Investment Committee. This committee acted independently of the Federal
Reserve Board.
3 Recall that commercial paper is a short-term (nine months or less) debt instrument that firms
issue to obtain funds. Commercial paper is negotiable, meaning it can be sold to someone else.
How did the Fed function as a whole? Basically the district Federal
Reserve banks acted with a large degree of autonomy. They pursued their
own independent discounting policies. If anything, they followed the lead
of the Federal Reserve Bank of New York, which held the largest percentage
of system reserves. Indeed, the New York Fed assumed responsibility for
relationships with all foreign central banks and sometimes acted without
informing even the Board of Governors. The Board of Governors found
itself with little real power.
The Great Depression. Perhaps the defining event in the history of

the Federal Reserve System was the debacle of the Fed’s handling of the
onset of the Great Depression. In 1929 the stock market crashed, and sub¬
sequently a tremendous wave of bank failures swept the nation. From 1929
to 1932 more than 5,000 banks failed, which amounted to one-third of all
commercial banks. Banks were unable to supply the currency depositors
demanded, and this failure only fueled the ongoing runs on banks across the
nation. In essence, the situation was like that faced at times during the
national banking system, only magnified many times over. The Federal
Reserve System, which had been created to provide an elastic banking
system—one that would stand able to provide currency during a crisis—
failed miserably in this test, refusing or neglecting to provide the necessary
reserves. In fact, the money supply actually fell by about one-third over this
period. The effect on the economy was severe, as the nearly total collapse
of the financial system led business to a standstill from which recovery came
only slowly.
The cause of the Great Depression is an issue of ongoing debate. Whether
the banking panics that followed the stock market crash were a cause or a
symptom is one element of that debate. However, condemnation of the Fed’s
inaction over this period is nearly unanimous. Whether that inaction merely
increased the severity of the Great Depression or directly caused it also
remains a matter of debate. There seems little question, however, that the
Fed failed to take the necessary and appropriate action during that crisis.
The Banking Act of 1935. In response to the Great Depression and

the failure of the Fed to supply the reserves needed to maintain stability in
the banking system and in the money supply, Congress moved to reform the
Fed. As we learned in Chapter 7, the Glass-Steagall Act of 1933 prohibited
commercial banks from dealing in corporate securities and empowered the
Fed to set margin requirements on stock purchases. This act allowed the Fed
some indirect regulation of securities markets and kept commercial banks
from investing in corporate bonds and stocks, which fluctuated too widely
in value and were considered too risky for a safe banking system.
However, it was the Banking Act of 1935 that really made the Fed what
it is today. It transformed the Fed from a cooperative with 12 independent
A B R IE F H ISTO RY O F CENTRAL BANKING IN THE UNITED STATES 417
offices to an institution with a real central authority located in the Board of

Governors in Washington, D.C. The Federal Reserve banks were no longer
able to set policy independently of the wishes of the Board of Governors.
The board was given complete power over relations with foreign central
banks, setting discount rates, and even the internal affairs and budgets of the
Federal Reserve banks. Monetary policy would now be determined in Wash¬
ington. The Federal Open Market Committee, derived from the governors’
conference’s Open Market Investment Committee, was created to buy and
sell government securities.
There were structural reforms as well. The chief executive officers of
the Federal Reserve banks were renamed from governors to presidents. They
were still chosen by the directors but were subject to the veto power of the
Board of Governors. The Board of Governors also was reformed. The sec¬
retary of the U.S. Treasury and the comptroller of the currency were no
longer allowed to serve on the board, and members’ terms were lengthened
from 10 to 14 years.
We describe the function and structure of the modern Federal Reserve
System in more detail later in this chapter. For now we simply note that the
institutional structure of the Fed today was determined in most of its essential
elements by the Banking Act of 1935.
The Accord of 1951. In the early 1940s, the Federal Reserve System
acted to keep the interest rate on government securities low and almost
constant to help the U.S. Treasury finance the war effort at a low interest
rate. The Fed could do this by buying or selling securities at the pegged
interest rate. This practice continued for some time after the war ended, but
the Fed was concerned it was becoming just an arm of the Treasury instead
of an independent monetary policymaker. Thus, the Fed and the Treasury
reached a friendly agreement in March 1951, known as the Federal Reserve-
Treasury Accord, that ended the practice of the Fed fixing the interest rate
to facilitate financing of the debt by the Treasury.
Between 1951 and 1980, the main problem the Fed faced was declining
membership. As we learned in Chapter 7, declining membership meant the
Fed’s control over the money supply was diminishing.
The Depository Institutions Deregulation and Monetary

Control Act of 1980. The Depository Institutions Deregulation and
Monetary Control Act of 1980, or DIDMCA, had a profound effect on the
structure of the Federal Reserve System—not by changing the board of the
Federal Reserve banks but by altering the membership status of commercial
banks and other depository institutions. Prior to 1980, commercial banks
could choose whether or not to be members of the Fed. Nationally chartered
banks were members, but state-chartered banks could choose not to be
members. Membership had both costs and benefits. Costs included purchas-
ing shares in the Federal Reserve bank, being subject to Fed oversight, and
holding reserves at the Federal Reserve bank. Benefits included access to
check clearing at subsidized prices and access to the discount window. State-
chartered banks, savings and loans, thrifts, and other depository institutions
had no direct access to Fed services. DIDMCA changed all that. Basically
it said that all federally insured depository institutions must meet Federal
Reserve System reserve requirements. In exchange, it also granted these
institutions access to the discount window and Fed check clearing. The act
also stipulated, however, that the Fed could no longer subsidize the price of
check-clearing services, so the value of this benefit lessened considerably.
As a result of DIDMCA, by 1987 the Fed’s authority over required
reserves extended to virtually all depository institutions, including commer¬
cial banks, savings and loans, credit unions, and any others accepting federal
deposit insurance. This was a huge increase in the Fed’s direct authority
over the banking system. Since 1987, legislation has focused largely on
reforms directed at the S&L crisis rather than at the Fed’s direct authority
over the money supply. Chapter 7 details these legislative actions.
Summary Throughout banking history, there have been various calls for an end to
Exercise 13.1 fractional reserve banking. The rationale for 100 percent reserve banking is
that a run on the deposits of a bank would never occur since the bank would
actually have reserves in currency, gold, or some other very liquid form
equal to the quantity of deposits. This idea was circulated as an alternative
to the national banking system and to the Federal Reserve System and, in a
revised form, has recently been suggested as a way to reform the current
U.S. banking system by Nobel prize-winning economist James Tobin, In
what ways did the free banking system and the national banking system
differ from a system of 100 percent reserves?
Answer: Both the free banking system and the national banking system
required banks to deposit with a government agency bonds equal in value
to the notes they issued. However, these bonds were not reserves usable in
the day-to-day operation of a bank; in fact, they served merely as collateral
against the issue of notes. Especially in the free banking system, the value
of these bonds fluctuated so widely that they might not have equaled the
value of the circulating bank notes. Thus, unlike 100 percent reserves in
currency (or 100 percent reserves in gold under a gold standard), the collat¬
eral of free banks was neither readily available for redeeming notes nor
necessarily in a form that would always maintain a value equal to the value
of the deposits. The main point is that having collateral equal to the value
of bank notes, or equal to the value of deposits, is not the same as having
100 percent reserves.
The Structure and Role of the Modern Federal Reserve System 419
The Structure and Role of the Modern

Federal Reserve System _
From its inception, the Fed has been largely independent of both the legis¬
lative and executive branches of government, an independence shared by
only a few other central banks worldwide. For example, the Bank of England,
the world’s oldest central bank, serves as an arm of the British government.
The independence of the Fed is the source of ongoing controversy, and
scarcely a session of Congress passes without someone introducing a bill to
somehow limit or abolish this independence.
To understand the functioning of the Federal Reserve System and such
issues as the debate over its independence, it is important to know how the
Fed is structured today.
The Institutional Structure

Today the Federal Reserve System consists of four main elements: (1) the
Board of Governors, (2) the 12 regional Federal Reserve banks, (3) the
Federal Open Market Committee, and (4) the member banks and other
depository institutions. We look at the first three elements in turn; we looked
at the banking system in detail in earlier chapters.
The Board of Governors. The Board of Governors consists of seven

members appointed by the U.S. president and confirmed by the Senate. Each
member serves a 14-year term. Moreover, no member is allowed to serve
more than one full term, although reappointment after a partial term (to
replace a resignation, for example) is possible. The terms are staggered so
that one position is appointed every two years. This policy has several
implications. First, the members know they can serve well beyond the term
of the president who appoints them. Second, a president would have a
difficult time appointing a majority of members to the Board of Governors,
since presidential terms are limited to four years (or to eight with reelection).
Thus, it is not easy to stack the board with sympathetic members in an
attempt to influence policy. Members are often chosen from the banking or
business community, and a fair number of economists have served among
the governors.
One member of the Board of Governors is appointed by the president to
serve as chairperson, again with Senate confirmation. The chair has a four-
year term that begins near the midpoint of a presidential term. Thus, newly
elected presidents inherit a chair of the Board of Governors for the first half
of their administration. The chair’s term can be renewed, and a chair whose
term is not renewed can continue to serve on the board as a regular member,
although this is rarely done.
It is fair to say that most of the real power at the Board of Governors—
and indeed in the entire Federal Reserve System—is concentrated in the
hands of the Fed chair (more formally known as the chair of the Board of
Governors). The chair of the Board of Governors sets the agenda for meet¬
ings on monetary policy and has substantial control over the staff serving
the board. Governors who are out of favor with the chair may find their staff
allocations cut significantly. Without competent staff support, a governor
would have a difficult time researching and keeping informed about current
issues and therefore would suffer a loss of power and influence. The chair
of the Board of Governors as of July 1994 is Alan Greenspan, who replaced
Paul Volcker in 1987.
The institutional structure of the Board of Governors is supposed to
insulate that body from political considerations, and to some extent it does.
The Fed does not rely on any arm of government for paying its bills and is
not audited by the Office of Management and Budget. This fact, coupled
with long appointments of members that technically are not renewable,
reduces the Fed’s tendency to cower to political pressures. Short of major
legislative changes or extreme presidential pressure, members of the Board
of Governors are relatively free of partisan considerations.
However, the board is responsible to Congress, and indeed the entire
Federal Reserve System is a creation of Congress. Even major changes
in the structure of the Fed require only a majority vote in each house
and the approval of the president (or a two-thirds vote in each house in the
event of a presidential veto). It is fair to say that the Fed is keenly aware of
this and is careful not to place itself at odds with Congress unnecessarily.
As long as the economy and the banking system function well, Congress
tends to leave monetary policymaking to the Fed. However, when things
have not gone well, such as during the Great Depression, Congress and the
president have been willing to greatly modify the structure and operation of
the Fed.
Despite its independence, the Federal Reserve works closely with the
U.S. Treasury on macroeconomic policy issues and on issues involving the
foreign exchange market. The Board of Governors is also responsible for
setting discount rate policy, setting reserve requirements, and supervising
the entire Federal Reserve System, including the budgets and internal affairs
of the Federal Reserve banks. To aid in these tasks, the board has a large
staff, including many economists who advise the governors on monetary
policymaking, regulation of the banking industry and the financial industry
in general, and other topics that may arise.
The Federal Reserve Banks. The structure of the district Federal

Reserve banks is nearly the same as that determined at the creation of the
Fed, but the Banking Act of 1935 stripped these banks of much of their
autonomy in policy matters. Let us briefly review this current structure.
The Federal Reserve System divides the country into 12 districts, and
each district contains a Federal Reserve bank. The boundaries of these dis¬
tricts have changed little from those in effect in 1917 and are shown in
Figure 13.1. Today the Federal Reserve banks are in Boston, New York,
Philadelphia, Cleveland, Richmond, Atlanta, Chicago, St. Louis, Minneap¬
olis, Kansas City, Dallas, and San Francisco. The Federal Reserve banks are
actually owned by commercial banks in their districts that are members of
the Federal Reserve System. This is accomplished by requiring each member
bank to buy shares in the district Federal Reserve bank, which pay a flat
Source: Federal Reserve Bulletin, Board of Governors of the Federal Reserve System, June
1993, A80.
dividend. Each Federal Reserve bank has nine directors, six chosen by the
commercial banks in the district that are members of the Federal Reserve
System and three appointed by the Board of Governors. The Board of
Governors also names a director to be chair of the board at each regional
bank. These directors nominate the presidents of the regional Federal Reserve
banks, but their choices must be approved by the Board of Governors.
The regional banks are technically private corporations owned by com¬
mercial banks, but in reality they operate under the direction of the Board
of Governors. In addition, any profits made over and above operating costs
are remitted to the U.S. Treasury rather than to the commercial banks’
shareholders. Most of these profits represent interest earnings on the Federal
Reserve System’s substantial holdings of U.S. government securities. The
remittance to the Treasury is small by U.S. budget standards but not insig¬
nificant, being on the order of $30 billion per year.
The regional banks serve as regulatory agencies for banks in their dis¬
tricts and as check-clearing centers. Sometimes they have two or three branch
offices in other cities in their region. The Federal Reserve Bank of New
York serves as the primary location for carrying out the Federal Reserve
System’s monetary policy actions. It also carries out foreign exchange in¬
terventions on behalf of the U.S. Treasury. Finally, as we shall see, the
presidents of the regional banks attend the Federal Open Market Committee
meetings and sometimes vote on monetary policy issues. The regional banks
provide research staffs of economists and others to advise their presidents
on both monetary policy issues and issues involving the regulation and
supervision of commercial banking.
The Federal Open Market Committee. The Federal Reserve Sys¬

tem’s policymaking arm is the Federal Open Market Committee, or
FOMC. The voting members of the FOMC consist of the seven governors,
the president of the Federal Reserve Bank of New York, the president of the
Federal Reserve Bank of either Chicago or Cleveland, and three presidents
from the other nine regional Federal Reserve banks (see Figure 13.2). Note
that five members of the FOMC—the presidents of regional Fed banks—are
not appointed by elected officials or elected by the general population. This
has led to controversy over the years, as Box 13.2 (page 424) explains.
Monetary policy actions, called open market operations and discussed
later in this chapter, are carried out primarily at the Federal Reserve Bank
of New York, which is why the president of that bank is always a voting
member of the FOMC. The other bank presidents serve on a rotating basis,
and the special positions accorded the Chicago and Cleveland banks are due
to historical and political factors. All 12 regional bank presidents attend and
discuss monetary policy issues at the FOMC meetings. Currently these meet¬
ings occur eight times per year, but only five of the bank presidents are
allowed to vote at any one time.
The Roles of the Fed

Originally the Fed was to provide an elastic currency and perform the fairly
automatic tasks required to maintain the price of gold and keep the United
States on the gold standard. Today its role has both expanded and become
more complex. If nothing else, monetary policy is more difficult than it was
under the gold standard.
The roles of the Fed are varied. They include
Bank for bankers

Bank for the federal government
Regulator of banks
Guardian of financial markets
Keeper of the currency
Gold depository
Clearinghouse for checks and wire transfers
Monetary policymaker
The Fed functions as the banker’s bank in several ways. It holds the
deposits of member banks, or reserves, and it makes loans to these banks at
Making the Fed More Responsive to the

Electorate: The Rebirth of an Old Idea
As recently as 1994, U.S. representative Henry B. monetary mismanagement in the pursuit of

Gonzales proposed legislation that would make political objectives; but I consider it prefer¬
members of the Fed, and particularly members able for such mismanagement to be a clear
of the FOMC, more accountable to the elector¬ responsibility of the administration, and ac¬
ate. The idea of making those responsible for countable to the electorate.
monetary policy accountable to those affected
Johnson recognized that a potential danger
by their policies is not new. In 1964, for in¬
of making the FOMC completely composed of
stance, the late Harry Johnson, then economist
presidential appointees is that it might lead the
at the University of Chicago, testified before the
Fed to pursue short-term political objectives (like
Subcommittee on Domestic Finance, Committee
cranking up the money supply just before an
on Banking and Currency, House of Representa¬
election to stimulate the economy and win
tives. According to Johnson,
votes). Largely for this reason, past attempts to
I believe that monetary policy should be make all members of the FOMC appointed have
brought under the control of the Executive failed.
and Legisature in the same way as other as¬
pects of economic policy, with the adminis¬
Source: Harry G. Johnson, "Should There Be an Inde¬
tration bearing the ultimate responsibility
pendent Monetary Authority?", in W. L. Smith and R. L.
for monetary policy as part of economic Teigen, eds., Readings in Money, National Income, and
policy in general. In making this recommen¬ Stabilization Policy, 3rd ed. (Homewood: Irwin Press,
dation, I must admit there is a danger of 1974), 277.
the discount window. The Fed is also the bank of the federal government.
The U.S. Treasury and other government agencies hold their accounts at the
Federal Reserve banks and make deposits and withdrawals from these ac¬
counts just as a depositor in a commercial bank would. The Fed also handles
the sale of U.S. savings bonds and U.S. Treasury bills for the government.
The Fed has regulatory power over depository institutions, including the
ability to set reserve requirements. The job of providing auditing oversight
is split between the Fed and other regulators, such as the FDIC. The Fed
also has the broader responsibility of maintaining stability in financial mar¬
kets, especially in the money markets.
The Fed is in charge of maintaining the supply of Federal Reserve notes,
acting as a gold depository, and clearing checks. Maintaining the supply of
Federal Reserve notes means providing the supply of currency demanded by

the public and keeping that supply in good condition. To this end, the Fed
removes worn notes from circulation and adds new ones as required. The
Fed, and in particular the Federal Reserve Bank of New York, serves as
gold depository for the Federal Reserve System and holds gold for both the
United States and other nations. Gold flows among nations are often no more
than transfers of gold bars from a room or shelf holding the gold of one
nation to a room or shelf holding the gold of another. Finally, the Fed
provides check-clearing services. Under the DIDMCA of 1980, these serv¬
ices are priced at cost, so the Fed must compete with check-clearing services
provided by private banks. Private check-clearing services can pick and
choose the most profitable locations in which to provide these services.
However, much as the post office must provide mail service to all locations
in the United States, the Fed must provide check-clearing services to all U.S.
banks.
Finally, the most important role of the Fed is that of the nation’s mone¬
tary policymaker. The Fed is in charge of the money supply, which consists
of not only Federal Reserve notes but also checkable deposits and other
forms of money we looked at in Chapter 2. Moreover, the Fed is responsible
for using its ability to change the money supply in ways that will benefit the
economy. The Fed’s overall policy goals are often listed as price level
stability, full employment, economic growth, interest rate stability, stability
of financial markets, and exchange rate stability. In our discussion of mone¬
tary policy in Chapters 19 through 21, we will see how the Fed might use
monetary policy to help achieve some of these goals.
Figure 13.3 shows the informal structure of the Fed and how it uses its
power to influence such things as discount rates, reserve requirements, and
open market operations. The next section provides a broad overview of these
monetary policy tools and how they affect the banking system. In Chapter
14, we examine in more detail how they affect the money supply.
Summary Ij The structure of the Fed is unique among central banks worldwide. The
Exercise 73.2| system of 12 Federal Reserve banks was originally set up to avoid concen¬
trating too much power in Washington, although the Banking Act of 1935
changed this by centralizing much of the Fed’s power in the Board of
Governors. What do you think is the benefit of continuing to have 12 distinct
Federal Reserve banks?
Answer: The Federal Reserve System needs branch offices to facilitate

check clearing, regulation, discount window lending, and other routine tasks.
The Federal Reserve banks and their branches fulfill this requirement. Thus,
for most purposes the Federal Reserve banks serve a function that would
still have to be performed in some way under an alternative Fed structure.
In monetary policy, however, the Federal Reserve banks play an important

role. The presidents of five of the reserve banks participate in Federal Open
Market Committee meetings and get 5 of the 12 votes on monetary policy
decisions. Thus, they can be independent voices at the FOMC meetings.
Tools of Monetary Policymaking 427
Moreover, the Federal Reserve banks have their own research staffs that are
independent of the board staff, allowing the bank presidents to more easily
receive advice and keep informed about monetary and financial policy issues.
Thus, the structure of the Fed, and particularly that of the FOMC, gives the
Federal Reserve bank presidents the ability to present opinions on policy
from outside the Board of Governors. When the Fed was first set up, there
was a fear of concentrating too much power in Washington. Today the
structure of the Fed shows the last vestiges of that fear in the provisions that
voices from outside Washington be heard and sometimes vote on monetary
policy actions.
Tools of Monetary Policymaking
As Figure 13.3 suggests, there are three main tools or instruments of mon¬
etary policy and a few minor ones. The major tools are open market opera¬
tions, discount window policy, and reserve requirements. Minor policy in¬
struments are the ability to enact selective credit controls, including margin
requirements, and the ability to use persuasion (sometimes called moral sua¬
sion) to lead banks and other depository institutions to act in a particular way.
Open Market Operations

Today the primary day-to-day means of conducting monetary policy actions
is via open market operations. The Fed continually engages in open market
operations, in which it buys U.S. government debt from securities dealers
or sells government debt to dealers. We will see in the next few chapters
that these open market operations affect, among other things, interest rates
and the money supply. In an open market purchase, the Fed purchases
U.S. government securities from a securities dealer, paying for the purchase
with an electronic funds transfer or a check that promises to pay Federal
Reserve notes.
When the Fed engages in an open market purchase, its assets increase,
since it has additional government securities. But its liabilities also increase,
since it has an outstanding check on which it will have to pay. The securities
dealer, meanwhile, has merely traded one asset—government bonds, for
instance—for another, the check from the Fed. Figure 13.4 illustrates this
process.
When the securities dealer deposits the check in his or her bank, the
dealer is again trading one asset, a check from the Fed, for another, deposits
at the bank. Meanwhile the bank experiences an increase in both assets and
liabilities. The check from the Fed is an asset, while the deposit account of
the securities dealer is a liability.
Finally, the bank will itself transform its assets, since it will present the
check to the Fed, depositing it in the bank’s reserve account. For the bank,
V

this means exchanging one asset, a check, for another, a reserve account.
For the Fed, this transaction transforms one of its liabilities from a check to
a reserve account.
The bottom of Figure 13.4 shows the net effect on the three parties. As
a result of the open market purchase, the Fed has gained an asset, the U.S.
government bonds, and incurred a liability, reserve deposits of the bank. The
securities dealer has merely traded one asset, the government bonds, for
another, a deposit account at a bank. The bank has gained an asset, an
increase in its reserve account, and incurred a liability, an increase in the
securities dealer’s deposit account.
How does this operation affect the money supply? An open market
purchase affects the reserves of a bank, which in turn affects the amount of
money it can loan out. In particular, the bank is required to hold reserves
against deposits, but these reserves are only a small fraction of the size of
the deposit account. With equal increases in reserves and deposits due to an
open market purchase, the bank will have excess reserves that it can use to
make loans. We will look at this process in much more detail in the next
chapter; here an example will highlight how it works.
Suppose required reserves are 10 percent of deposits. The bank, which
has just received an additional $100 worth of deposits, is required to hold
only $10 in reserves and may loan out up to $90 of the other reserves. Thus,
the bank’s position shown in Figure 13.4 will change again, since the bank
will seek to make loans of up to $90 with its new reserves. (We will see in
Chapter 14 how the initial increase in loans and deposits actually gets mul¬
tiplied by the actions of the banking system so that the increase in deposits
and loans is much greater than just the initial $100 increase in deposits and
$90 increase in loans.) In this way, open market purchases directly increase
deposits and also increase excess reserves. When banks lend out these excess
reserves, further increases in the money supply will occur. As we will see
in Chapter 16, increases in the money supply can have a stimulative effect
on the economy. Thus, open market purchases are considered an expansion¬
ary policy, a policy that tends to increase aggregate economic activity.
In an open market sale, the Fed sells government securities (bonds,
bills, or notes) to a securities dealer. In Figure 13.5, for instance, the secu¬
rities dealer gains an asset, government bonds, but loses another asset, its
deposit account. The Fed loses an asset, the government bonds it sold, but
it also loses a liability, since the bank’s reserve account at the Fed decreases.
The bank loses an asset, since its reserve account is reduced, and also loses
a liability, because the securities dealer’s deposit account decreases.
Again, however, this decrease in the bank’s assets and liabilities has
other effects. If the bank were holding reserves just equal to 10 percent of
all deposits, the equal reduction in its reserves and deposits would mean the
bank has too few reserves. For example, in Figure 13.5 the bank has lost
$100 in deposits and reserves. Since it lost $100 in deposits, it can lose $10
of reserves with no consequence to its reserve position. But the loss of $100
of reserves means the bank is $90 short of its reserve requirement. It must
either recall some outstanding loans or sell some of its holdings of U.S.
government bonds to receive sufficient reserves to cover the $90 loss. Thus,
the open market sale has resulted in a decrease in deposits—that is, a decrease
in the money supply—and a reduction in loans. This decrease in deposits,
loans, and reserves at one bank reverberates throughout the banking system,
leading to total decreases in deposits and loans that are multiples of the
initial effects, as we shall see in Chapter 14. Thus, open market sales contract
the money supply and exert a contractionary force on the economy. Box
13.3 on page 432 shows the magnitude of open market purchases and sales
carried out by the Fed.
There are a few important institutional details regarding open market
operations. First, the Fed and the securities dealers are dealing in a secondary
market for securities. The U.S. government securities being bought and sold
are not new issues by the Treasury; they are bonds that have been previously
bought by someone other than the Fed. The Fed is prohibited from actually
bidding on new issues of U.S. government securities.
Second, the purchase or sale of government securities from or to secu¬
rities dealers is a voluntary activity; dealers do not have to buy or sell them
at a set price. If the securities dealers don’t want to sell at the offered price,
they will simply refuse the Fed’s offer. Of course, if the Fed wants to buy
securities badly enough, it will merely offer a higher price, and eventually
it will find a price at which someone will sell. Similarly, if it wants to sell
securities badly enough, it will offer a lower price.
Reserve Requirements
The Board of Governors sets reserve requirements on all federally insured
depository institutions within limits set by Congress. Reserve requirements
are both an extremely powerful and a seldom used tool of monetary policy.
The Fed has found other policy tools, especially open market operations, to
be much more effective and controllable ways to change the money supply.
When changes in reserve requirements do occur, they usually do so in
conjunction with other structural changes in the banking industry, such as
the DIDMCA of 1980, which imposed a new structure of reserve require¬
ments across financial institutions. Such changes in reserve requirements are
not conceived of as monetary actions per se but are made for purposes of
regulation and stabilization of the banking industry. Still, changes in reserve
requirements are a potential policy tool, and even when they are made for
reasons not connected with monetary policy, they exert strong effects on the
money supply.
How does an increase or a decrease in reserve requirements change the
money supply? Consider a decrease in reserve requirements. A bank has
432 C hapter13 The Federal Reserve System
Open Market Operations and Churning
The accompanying table indicates the dollar vol¬ intentional obscuring of the thrust of monetary
ume of open market operations from May policy by continual buying and selling of securi¬
through November 1992. These transactions in¬ ties. According to this view, frequent buying and
clude what are called outright transactions, in selling by the Fed hides the true intent of its
which the Fed simply buys or sells securities; monetary policy. This may indeed explain part of
matched transactions, in which the Fed sells a the large-scale use of temporary purchases and
security to a dealer and agrees to buy it back at sales.
a specified future date and price; and repur¬ However, there are also good reasons for
chase agreements, in which the Fed buys a se¬ temporary changes in reserves. Part of the job of
curity from a dealer and the dealer agrees to ensuring an elastic currency is to provide for the
repurchase it in the future. Neither matched increased demand for currency during the
transactions nor repurchase agreements create Christmas season by increasing bank reserves
permanent changes in reserves; rather, both and then reducing those reserves when the de¬
change reserves only temporarily. The table mand for currency returns to more normal levels
shows, however, that the vast majority of open after the holidays. One way to do this is through
market purchases and sales are matched trans¬ Fed repurchase agreements. The Fed buys secu¬
actions or repurchase agreements. rities in late November, and dealers agree to re¬
Why the preponderance of temporary as purchase them a few weeks later, when the
opposed to permanent open market sales and Christmas shopping rush is over, to increase
purchases? Nobel Laureate Milton Friedman and banking system reserves for a short period of
other critics argue the answer is churning—the Continued on p. 433
Outright Transactions Matched Transactions Repurchase Agreements
Month Purchases Sales Purchases Sales Purchases Sales
May $4,310 0 $118,972 $117,524 $38,777 $38,533

June 3,836 0 126,977 129,216 10,792 11,036
July 0 0 127,051 126,137 12,224 12,224
August 866 0 104,873 102,575 39,484 31,868
September 5,927 0 116,331 115,579 68,697 59,628
October 4,272 0 116,024 114,917 18,698 35,383
November 7,820 0 115,020 117,020 42,373 39,117
Figures in millions of dollars.

Sources: Federal Reserve Bulletin, Board of Governors of the Federal Reserve System (March 1993), A10; National
Economic Trends (April 1993), 21.
Continued from p. 432 Fed buys and sells an amount of securities equal
time. However, critics point out that the Fed en¬ to roughly half of its portfolio in any given
gaged in matched purchases on the order of month! Critics like Friedman claim this volume
$115 billion in November 1992 compared with of trading is not necessary to deal with seasonal
total Fed holdings of government securities, changes in the demand for currency.
which were $291 billion. More generally, the
deposits against which it holds required reserves; it generally makes loans

with the remainder. Thus, a bank might be in the position shown in part a
of Table 13.1. The required reserve ratio is 10 percent. The bank has deposits
of $1,000. It holds reserves equal to 10 percent of these deposits and has
loaned $900. What if the required reserve ratio is reduced to 5 percent, as
in part b of Table 13.1? Now the bank still has reserves of $100, but required
reserves are only $50. There is $50 in extra reserves, called excess reserves.
These reserves are now available for lending, and the bank will increase
loans by $50. Furthermore, this happens at every bank. Not only do loans
increase at every bank, but the interactions of these banks will produce a
change in loans and deposits that is a multiple of this initial effect, as we
will see in Chapter 14. The reduction in reserve requirements causes an
expansion in the money supply and has an expansionary influence on the
economy.
What about an increase in reserve requirements? A rise in reserve re¬
quirements will have a similar but opposite effect on each bank and on the
banking system. Consider the situation in part a of Table 13.2. The bank
initially faces a 10 percent reserve requirement. It has deposits of $1,000,
reserves equal to the required $100, and loans of $900. An increase in reserve
requirements to 20 percent creates the situation in part b of Table 13.2. The
bank’s required reserve amount increases to $200. Now excess reserves are
— $100. In this situation, the bank’s only recourse is to reduce loans by $100
or sell $100 in securities to obtain the required reserves. This causes a
problem, of course, because calling in loans is always painful, and more so
because this will happen at all banks. (A general sale of securities would
also be painful, since it would cause the prices of those securities to decline.)
In the end, not only will the initial reduction in loans occur, but the effect
on loans and deposits will be magnified by the interactions of banks in the
banking system. Thus, an increase in reserve requirements has a contrac¬
tionary effect on the money supply and on the economy.
Discount Window Policy

We have seen that in the history of the Fed, discounting was once the primary
means of conducting monetary policy. This has long since changed, as have
(a) Required Reserve Ratio = 10%
Assets Liabilities
Reserves: $100 Deposits: $1,000
Loans: $900
(b) Required Reserve Ratio = 5%
Assets Liabilities

Required: $50
Excess: $50
Loans: $900
the procedures at the discount window. Originally the purpose of the dis¬
count window was to make loans against specific assets offered by banks,
first commercial paper and later other assets. Today the discount window
operates in somewhat the same way under procedures set by the Fed. Banks
and other depository institutions can “go to the window” to borrow from
the Fed. What the banks actually borrow are reserves. These reserves, called
borrowed reserves, function in the same way as any other reserves of a
bank. They can be used to meet reserve requirements, and excess reserves
can be used to make loans. The only difference is that borrowed reserves
must be repaid to the Fed, usually within a few weeks or months.
The interest rate the Fed charges is called the discount rate. It is not a
market rate but a rate set by the Fed. Changes in the discount rate are widely
publicized. Increases tend to discourage borrowing, reducing borrowed re¬
serves. This reduction in borrowed reserves has the same effect on the
banking system as any other reduction in reserves, such as that caused by
an open market sale. Deposits and loans contract, and the money supply
Table 13.2
Impact of an Increase in Required Reserves on a Bank's Balance Sheet
An increase in the required reserve ratio from 10 to 20 percent leads to a decrease in

excess reserves of - $100; that is, it increases the amount of required reserves. This typi¬
cally leads to a decrease in loans.
(a) Required Reserve Ratio = 10%
Assets Liabilities
Loans: $900
(b) Required Reserve Ratio = 20%
Assets Liabilities

Required: $200
Excess: -$100
Loans: $900 -y;'', \ Ai'A'y’
falls. Thus, increases in the discount rate are considered a sign of contrac¬
tionary monetary policy. Reductions in the discount rate tend to encourage
borrowing, thus increasing borrowed reserves. This rise in reserves increases
deposits and loans of the banking system and is considered a sign of an
expansionary monetary policy.4
An important point to remember in evaluating discount window policy
is that the Fed sets not only the discount rate but also the amount a bank
can borrow. It is not the case that a bank can borrow all the reserves it wants
at the discount rate. The actual effect of changes in the discount rate depends
on the resulting change in reserves, so a reduction in the discount rate that
4 The above analysis has an important qualification. The discount rate is set by the Fed and might
be above or below other market-determined rates. If the discount rate is below other market rates,
an increase in the discount rate may be seen as a “technical adjustment” to keep the discount rate
in line with market rates and not as a sign of contractionary policy. Similarly, if the discount rate
is above other market rates, a reduction in the discount rate may also be seen as a “technical
adjustment” and not as a sign of expansionary policy.
does not lead to an increase in reserves will not be expansionary. There must
be a reduction in the discount rate, accompanied by increased Fed lending
at the window.
The discount window has traditionally been intended for short-term
borrowing of reserves—what the Fed calls adjustment credit and seasonal
credit. Adjustment credit is loans made to depository institutions to allow
them to more gradually adjust their reserves and loan portfolios in response
to changes in deposits. Seasonal credit is loans made to depository institu¬
tions facing strong seasonal variations in deposits and loans, such as banks
in agricultural areas. A third category of borrowing is called extended credit
and represents long-term lending to troubled depository institutions. In recent
years, the dollar amount of extended credit has at times rivaled adjustment
and seasonal credit as the Fed made long-term loans to troubled savings and
loans and other depository institutions. In 1988 extended credit exploded to
more than $3 billion, as Figure 13.6 shows, and, after returning to near $500
million in the second half of 1989, again peaked to more than $2 billion in
1990. By 1992 extended credit had returned to about $150 million, the same
amount as the sum of seasonal plus adjustment credit.
Other Monetary Policy Tools

In addition to setting the discount rate, establishing reserve requirements,
and directing open market operations, the Fed has other monetary policy
tools. These include selective credit controls and moral suasion.
Selective Credit Controls. The Fed is authorized to enact selective

credit controls. It can, for example, set the margin requirement on stock
purchases. Stocks are purchased on margin when the purchaser borrows a
portion of the funds needed to buy the stock. A margin requirement of 40
percent would require that a buyer provide funds equal to 40 percent of the
price of the stock. The Fed can raise margin requirements if it so chooses to
prevent buyers from borrowing the entire amount of the purchase price. If
margin purchases were widespread, a large reduction in the stock price could
cause problems in the repayment of such loans and thus cause problems for
the entire financial industry.
The Fed is also authorized to set selective credit controls on other loans.
An example occurred in 1980, when President Carter asked the Fed to restrict
unsecured consumer credit. This took the form of a 15 percent reserve
requirement on credit card loans and other unsecured consumer loans and is
widely blamed for reducing consumer spending and contributing to the
recession of 1981-1982.
MoraI Suasion. As monetary policymaker and regulator of the banking

industry, the Fed has considerable clout in persuading depository and other
Source: Federal Reserve Bank of St. Louis Monetary Trends (October 1993), 12.
financial institutions to behave in what it regards as the public’s best interest.

Moral suasion is the term used for this persuasion, and the term itself
indicates the response by banks is voluntary. The Fed can use moral suasion
in attempts to lower interest rates or to get banks to impose looser or tighter
standards on loan applicants. It is often difficult, however, to distinguish
moral suasion from more direct pressure that the Fed can bring to bear on
depository institutions. For example, depository institutions are well aware
the Fed can use its discount window policies and even its auditing powers
to punish noncooperative behavior.
Summary Figure 13.4 showed the effects of an open market purchase on a bank’s
Exercise 13.3 balance sheet. Now consider a bank that goes to the discount window to
borrow $100 from the Fed, thus incurring a liability of $100—its IOU to the
Fed. The bank also gets an increase in its reserve account, an asset, of $100.
Show the balance sheet of this bank, and compare it to the balance sheet
after the open market operation shown in Figure 13.4. Is there any difference?
Why might the Fed choose to use discount window lending instead of open
market operations?
After the discount window borrowing, the bank has additional

reserves of $100 and an additional liability, the $100 owed to the Fed. Thus,
the change in its balance sheet looks like this:
Discount Window Borrowing
Assets Liabilities
Reserves: + $100 Loan from Fed: + $100
Recall that after the open market operation, the change in the balance sheet
of the bank looked like this:
Open Market Purchase
Assets Liabilities
Reserves: + $100 Deposits: + $100
The only difference is in the type of liability incurred. Discount window

borrowing leaves the bank with a loan it owes to the Fed of $100, whereas
the open market operation leaves the bank with a deposit account of $100.
The loan to the Fed is usually short term, meaning the bank knows the loan
will have to be repaid, reversing the process and draining away the $100 in
reserves. Thus, the bank will not make long-term loan commitments with
the additional excess reserves. The open market operation, in contrast, has
no such self-liquidating feature, giving the bank more flexibility in lending
out its excess reserves. Of course, discount window borrowing can mimic
open market operations if the borrowing is extended credit. In that case, the
loan is much longer term, and the bank can be more flexible in using the
reserves it obtains.
Why doesn’t the Fed use open market operations instead of the discount
window? If the Fed wants the change in reserves caused by an open market
operation to be short term, it can always conduct an open market sale
The Debate over Federal Reserve Independence 439
in the future.5 However, open market operations are nonselective, affecting

banks and the banking system but not targeted at any particular bank. The
discount window is a way for the Fed to increase or even decrease reserves
at a particular bank and as such is an instrument it can use to aid troubled
banks or just banks that want to stretch out the process of adjusting to
changes in reserves.
The Debate over Federal Reserve Independence
Our description of the structure and history of the Fed demonstrates how the
Fed is able to act at least somewhat independently of both the legislative
and executive branches of government. However, it does not tell us whether
or not that independence is a good thing. Is the Fed too independent, or is
it not independent enough?
Two issues arise in any discussion of Fed independence. The first is
whether or not the Fed should be allowed to pursue an independent monetary
policy. By this we mean a policy independent of the dictates of fiscal policy.
Fiscal policy is determined by the combined actions of the legislative and
executive branches of government. The issue—whether the Fed should be
able to pursue a monetary policy that directly contradicts or counteracts the
government’s fiscal policy—is discussed in detail in Chapter 19.
The second issue is whether monetary policy should be set in a discre¬
tionary way, meaning in each period the Fed decides what it currently
determines as best for the economy, or be more automatic, following some
set rule that limits discretion. More simply, should the Fed be able to set
monetary policy independently of the stance of fiscal policy? Those an¬
swering yes usually base their arguments on the idea that Fed independence
insulates the Fed from political pressure to inflate. The reasoning is as
follows. Politicians are often shortsighted, worrying only about the next
election. If they were allowed to determine monetary policy, their shortsight¬
edness would prevent the pursuit of long-run price stability. Instead politi¬
cians would pursue a policy of inflation to gain a short-run benefit to the
economy and enhance their reelection chances. According to this view, an
independent Fed is a stalwart opponent of inflation and will maintain price
stability if politicians are kept from having direct control of the Fed.
There is some evidence that the political manipulation of the Fed by
elected officials has resulted in inflationary monetary policy by the Fed. In
particular, some argue that during presidential election campaigns, the ex¬
ecutive branch has successfully influenced the Fed to pursue an inflationary
policy, temporarily stimulating the economy before the election to help the
5 It could also use a repurchase agreement. In a repurchase agreement, the Fed buys securities but
agrees to sell them back at a set price at a specified point in time. This creates a temporary increase
in reserves that ends when the repurchase occurs.
incumbent’s chances but resulting in longer-term inflation. Box 13.4 pro¬

vides some international comparisons of inflation and the level of central
bank independence.
The second argument in favor of Fed independence is that independence
allows the Fed to resist pressure from the government to monetize the
government debt. The Fed can monetize the debt by printing money and
using it to buy outstanding government bonds. This action would reduce the
size of the debt held by the public, but it would also be inflationary. Ac¬
cording to proponents of independence, an independent Fed is again a bul¬
wark against the inflationary impulses of the government.
Finally, some argue that an independent Fed is actually a useful public
scapegoat for the executive and legislative branches of government; that is,
the Fed is useful because it can be publicly blamed for a host of evils that
might actually have their origin in the actions of others in the government.
If the Fed were more clearly a branch of the executive, the president would
have greater difficulty blaming the Fed for various ills.
These arguments view the Fed as the “good guy” in the battle against
inflation and elected politicians (the government) as all too ready to inflate
the money supply for their own short-run purposes if only the independence
of the Fed didn’t get in their way.
Of course, there are arguments on the other side. The need for policy
coordination between fiscal and monetary policymakers is one argument for
linking the Fed more closely to policymakers in the executive and legislative
branches. An independent Fed certainly might choose to cooperate with the
fiscal authority, but it might also choose the converse. By making the Fed a
branch of the Treasury, for instance, there would be less potential for con¬
flicting policy actions, whether inadvertent or otherwise.
Summary What is there in the structure of the Fed to explain why the Fed would be
Exercise 13.4 more opposed to inflation than a member of Congress would?
Answer: The main feature is the fact that the members of the FOMC,
from the Board of Governors to the Federal Reserve bank presidents, do not
have to run for reelection. Thus, they can take a longer-term view of policy
actions. They will be less enamored with the short-run benefits of an expan¬
sionary monetary policy and more concerned with the long-run costs in
terms of inflation.
Conclusion
In this chapter, we took a detailed look at the structure of the modem Fed,
along with its historical roots. The creation of the Fed in 1914 was brought
on by the failure of the free banking and national banking systems in the
Conclusion 441
The Benefits of Central Bank Independence:

A Look at Inflation Around the World
An independent central bank has little reason to period for 17 nations and compared it to the
use monetary policy to stimulate the economy degree of independence. The degree of inde¬
during elections. For this reason, proponents of pendence was determined by taking account of
central bank independence argue that this re¬ the formal institutional relationship between the
sults in a more stable money supply and less in¬ central bank and the national government, such
flation. Is this actually the case? as the presence of government officials on the
It is hard to tell by looking just at the United central bank's governing board, whether the
States, since it has never had a central bank that head of the central bank was appointed by the
did not have a large degree of independence. government, and the existence of rules requiring
One way to get information on this question is the central bank to help finance the deficit (i.e.,
to look at the experience of other countries, es¬ requiring monetization of part of the deficit).
pecially other industrialized countries. Recently Alesina constructed four categories of inde¬
this was done by Alberto Alesina, who charted pendence and judged the most independent
the average inflation rate for the 1973-1986 Continued on p. 442
Switzerland
Germany
Japan
More
United States
Independent
Netherlands
Belgium
Canada
Norway
Sweden
Denmark
France
Finland Less
United Kingdom Independent
Australia
New Zealand
Spain
Italy
i
0 5 10 15 20
Inflation (%)
Continued from p. 441 a buttress against inflation. A danger of reduc¬

central banks to be those of West Germany and ing the Fed's independence, as current propos¬
Switzerland. Their average inflation rate over als to make all members of FOMC appointed
this period was 4 percent. The United States and would do, is that doing so could lead to high
Japan were determined to be in the category of inflation rates like those in Spain and Italy,
the second most independent central banks, which have less independent central banks.
with average inflation rates of about 7 percent.
Countries with the least independent central
Source: Alesina, Alberto, "Politics and Business Cycles in
banks—Italy, New Zealand, Australia, and
Industrial Democracies," Economic Policy, 8 (April 1989),
Spain—had average inflation rates of 12.5 per¬
p. 81 and W. Michael Cox, "Two Types of Paper: The
cent. As the accompanying figure shows, the Case for Federal Reserve Independence," The Southwest
more independent central banks do seem to be Economy, Federal Reserve Bank of Dallas, Nov-Dec 1992.
late 1800s to provide an elastic and stable currency. The Fed’s role has
changed over the past 50 years, largely in response to problems that arose
during financial crises like the Great Depression. Today the Fed is charged
with regulating the banking industry and controlling the nation’s money
supply. In the next chapter, we will see how Fed policies like reserve require¬
ments and open market transactions affect the money supply.
Key Terms
central bank open market sale
wildcat bank excess reserves
greenbacks discount window
discounting borrowed reserves
national banking system discount rate
Federal Reserve System adjustment credit
elastic currency seasonal credit
real bills doctrine extended credit
Board of Governors margin requirement
Federal Open Market Committee (FOMC) moral suasion
open market purchase

1. Describe the Board of Governors of the Governors could a two-term president ap¬
Federal Reserve System. With no resigna¬ point?
tions, how many members of the Board of
2. Describe a Federal Reserve System re¬ ing Fed independence. Which do you
gional bank. How is it like a private cor¬ agree with, and why?
poration? How is it like a government
8. What were some benefits of the First and
agency?
Second Banks of the United States?
3. Describe the Federal Open Market Com¬
mittee. How many voting members are on
9. Describe the chief features of the free
banking system. Compare them with those
this committee? How often does a mem¬
of the national banking system.
ber of the Board of Governors get to vote
on the FOMC? How often do the presi¬ 10. What were the goals of advocates of the
dents of the Federal Reserve banks get to free banking system? Did the free banking
vote on the FOMC? system achieve those goals? Explain.
4. What features of its design tend to make 11. What were the goals behind setting up the
the Federal Reserve System independent national banking system? Did the national
of the elected branches of government? banking system achieve those goals?
What features detract from its independ¬ Explain.
ence?
12. The free banking system is used as an ex¬
5. Obtain a copy of the current Economic ample of why banking needs to be regu¬
Report of the President and find out the lated. What were the problems with the
current U.S. government deficit and tax free banking system? What is the evi¬
receipts. Then obtain the Bulletin of the dence that insufficient regulation caused
Federal Reserve System and find out the those problems? Can one argue that the
remittances from the Federal Reserve Sys¬ problem was inappropriate regulation
tem to the U.S. Treasury. What fraction of rather than too little or too much regu¬
U.S. government revenue is made up of lation?
remittances from the Federal Reserve Sys¬
tem? What fraction of the U.S. govern¬ 13. What were the goals of those who origi¬
ment deficit? What fraction of the U.S. nally set up the Federal Reserve System in
government debt does the Fed hold? 1914? Were those goals achieved?
Expain.
6. Considering question 5, what do you think
of a proposal to raise more government 14. During the era of the national banking
revenue by instructing the Federal Reserve system, there continued to be occasional
System to remit more money to the U.S. periods of bank runs in which deposits
Treasury? Since the Federal Reserve Sys¬ could not be redeemed for currency. Did
tem earns most of its profit on holdings of the Federal Reserve Act of 1913 eliminate
U.S. government bonds, it would have to this problem?
purchase more such bonds. Would pur¬ 15. What were the main reforms contained in
chasing more bonds be possible? Would it the Banking Act of 1935?
be desirable? On what do you base your
answer? Consider carefully what it takes 16. What were the main changes made to the
to purchase more bonds and the likely Fed by the DIDMCA of 1980?
effect on the economy.
17. Do banks like high reserve requirements?
7. First, state the case favoring a less inde¬ Why or why not? What about the U.S.
pendent Fed. Second, state the case favor¬ Continued on p. 444
Continued from p. 443 ure. Would this liability be sufficient to

government? Ignoring the effects on the prevent noteholders from suffering losses
financial health of banks, does a revenue- in the above cases?
maximizing government prefer a banking
system with high or low reserves? In an¬
swering this question, consider how these 20. Assume you are a banker in the time of
reserves come into existence. free banking. You start with $50,000 in
gold, which you use to purchase govern¬
18. Is an open market operation that injects ment bonds for deposit in the state treas¬
$ 1 million in reserves into the banking ury. This transaction authorizes you to
system more or less expansionary than a issue bank notes. You use these bank
discount window operation that injects $1 notes to purchase another $50,000 in gov¬
million in borrowed reserves? ernment bonds, which you also deposit at
the state treasury. This second deposit
19. Assume you are a banker in the time of gives you the right to issue another
free banking. You start with $50,000 in $50,000 in bank notes. With this second
gold, which you use to purchase govern¬ $50,000, you use $10,000 to buy gold re¬
ment bonds for deposit in the state treas¬ serves and lend out the remaining
ury. This transaction authorizes you to $40,000. What does your balance sheet
issue bank notes, and you use $10,000 of look like? How safe is your bank? Con¬
these notes to buy gold reserves and lend sider the following.
out the remaining $40,000. What does (a) . If the value of bonds at the state
your balance sheet look like? How safe is treasury falls to zero but your loans re¬
your bank? Consider the following. main sound, will your bank have to close?
(a) . If the value of bonds at the state Will noteholders be at risk of suffering a
treasury falls to zero but your loans re¬ loss?
main sound, will your bank have to close? (b) . If the value of bonds at the state
Will noteholders be at risk of suffering a treasury stays at $100,000 but the value of
loss? the loans you have made falls to zero, will
(b) . If the value of bonds at the state your bank have to close? Will noteholders
treasury stays at $50,000 but the value of be at risk of suffering a loss?
the loans you have made falls to zero, will (c) . If both the vaue of bonds at the state
your bank have to close? Will noteholders treasury and the value of your loans fall to
be at risk of suffering a loss? zero, will your bank have to close? Will
(c) . If both the value of bonds at the state noteholders be at risk of suffering a loss?
treasury and the value of your loans fall to (d) . In many states, owners of free banks
zero, will your bank have to close? Will were liable not only for their initial invest¬
noteholders be at risk of suffering a loss? ments but also for an additional sum equal
(d) . In many states, owners of free banks to their initial investments. That is, the
were liable not only for their initial invest¬ bank owner in our example would be lia¬
ments but also for an additional sum equal ble for another $50,000 in personal funds
to those investments. That is, the bank to cover bank notes in the event of bank
owner in our example would be liable for failure. Would this liability be sufficient to
another $50,000 in personal funds to prevent noteholders from suffering losses
cover bank notes in the event of bank fail¬ in the above cases?
Bradley, M. D. and D. W. Jansen, “Federal Reserve bridge; New York, and Melbourne: Cambridge
Operating Procedures in the Eighties: A Dynamic University Press, 1989, 290-328.
Analysis.” Journal of Money, Credit, and Bank¬ Rockoff, H. “New Evidence on Free Banking in the
ing, 18 (August 1986), 323-335. United States.” American Economic Review, 75
Cosimano, T. F., and D. W. Jansen. “Federal Reserve (September 1985), 886-889.
Policy, 1975-1985: An Empirical Analysis.” Jour¬ Rockoff, H. “Institutional Requirements for Stable
nal of Macroeconomics, 10 (Winter 1988), 27-47. Free Banking.” Cato Journal, 6 (Fall 1986), 617—
Fenstermaker, J. V., and J. E. Filer. “Impact of the 634.
First and Second Banks of the United States and Selgin, G. A. “Why Does Europe Want a Federal Re¬
the Suffolk System on New England Bank Money: serve System?” Cato Journal, 10 (Fall 1990),
1791-1837.” Journal of Money, Credit, and Bank¬ 449-453.
ing, 18 (February 1986), 28-40. Selgin, G. A., and L. H. White. “The Evaluation of a
Jansen, D. W., “Ranking Federal Reserve Research Free Banking System.” Economic Inquiry, 25
Departments by Publications in Professional Jour¬ (July 1987), 439-457.
nals.” Journal of Macroeconomics, 13 (Fall 1991), Wells, D. R., and L. S. Scruggs. “Free Banking: The
733-742. Ultimate Deregulation of the Financial System.”
Lawson, C. L., and L. L. Lawson. “Financial System Atlantic Economic Journal, 12 (September 1984),
Restructuring: Lessons from Veblen, Keynes, and 80.
Kalecki.” Journal of Economic Issues, 24 (March Wicker, E. “Colonial Monetary Standards Contrasted:
1990), 115-131. Evidence from the Seven Years’ War.” Journal of
Luckett, D. G. “A Pedagogical Note on the Use of Economic History, 45 (December 1985), 869-884.
Margin Credit.” Journal of Economic Education, Wicker, E. “Interest Rate and Expenditure Effects of
19 (Fall 1988), 337-340. the Banking Panic of 1930.” Explorations in Eco¬
Miron, J. A. ‘ The Founding of the Fed and the De- nomic History, 19 (October 1982), 435-445.
stabilization of the Post-1914 U.S. Economy.” In Yohe, W. P. “The Intellectual Milieu at the Federal
A European Central Bank? Perspectives on Mone¬ Reserve Board in the 1920s.” History of Political
tary Unification after Ten Years of the EMS. Cam¬ Economy, 22 (Fall 1990), 465-488.
V
CHAPTER
14 The Money Supply Process
the beginning of this book, we have directly or indirectly discussed

topics related to money and the money supply process. In Chapter 1 we
defined money as the medium of exchange, something that is widely accepted
in exchange for other goods. In Chapter 2 we looked at the practical defi¬
nitions of money adopted by the Federal Reserve System, such as Ml, M2,
or M3. In the last chapter we discussed the Fed’s policy instruments for
controlling the money supply.
The nation’s money supply has a vital impact on the performance of the
overall economy. We have already seen that the rate of growth in the money
supply affects the economy’s rate of inflation. In the next few chapters, we
will see that money can also affect interest rates, investment, real GDP, and
employment.
Before we explore these important aspects of money, however, we need
to look more closely at how money is created. We will see in this chapter
that money creation is not simply the government’s printing of the dollar
bills you use to buy hamburgers and CDs. In fact, we will see that the entire
banking system creates money with the aid of the Fed, which acts as the
nation’s central bank.
Our look at the Fed in the previous chapter gave you a broad overview
of its central banking activities and pointed out that its primary function is
to manage the nation’s money supply. In this chapter we examine the quan¬
titative impact on the money supply of Fed actions such as open market
operations. Of course, if the Fed takes actions that change the nation’s money
supply too much or too little, it can have adverse effects on the economy,
causing a rapid rise in the price level or even a recession. The techniques
developed in this chapter, however, show that the Fed can reasonably ascer¬
tain how much its policies will change the money supply.
We begin our study by extending what we learned about money creation
in Chapter 1 to our modern economy, in which the Fed actively manages
the money creation process by supplying reserves to the banking system.
Then we develop a graphical framework for analyzing monetary policy that
will be very useful in the remainder of the book for analyzing the effects of
the Fed’s monetary policy on economic activity.
446
The Balance Sheet of Banks and

Multiple Deposit Creation
In Chapter 6, we saw that individual banks use deposits as an input to produce
loans. In this chapter, we look at the maximum amount of loans the entire
banking system can create given policies enacted by the Fed. Because new
bank loans can affect the total amount of money in the hands of the public
(in the form of either currency or deposits at other banks), the banking
system’s production of loans affects the amount of money in the economy.
We will see in this section, however, that new loans issued by a bank create
money only if the reserves used to create the loans come from outside the
banking system.
Our basic tool of analysis in this chapter is the T-account, such as that
shown in Table 14.1. Notice that total assets (the left-hand side of the
T-account) equal the bank’s liabilities plus net worth, since net worth is
defined as the difference between assets and liabilities. Table 14.1 shows
that our hypothetical bank, Bank One, has assets of $510 million consisting
of $100 million in reserves, $400 million in loans, and a building worth $10
million. At the same time, the bank has liabilities of $500 million to depos¬
itors with accounts at the bank and $10 million of net worth.
Reserves consist of legally mandated reserves, called required reserves,
and any additional reserves, called excess reserves. Required reserves can
be held in two ways: as vault cash or as a deposit at the district Federal
Reserve bank. (In fact, at least a portion of required reserves must be held
With deposits of $500 Table 14.1

million and a required Balance Sheet of a Bank with a Required Reserve Ratio of 20 Percent
reserve ratio of 20
percent, the bank
must hold $100 mil¬
Balance Sheet of Bank One
lion of its assets as re-
serves. This bank is
not holding any excess Assets Liabilities
reserves.
Reserves $100,000,000 Deposits $500,000,000

Required $100,000,000
Excess 0
Loans 400,000,000 Net Worth 10,000,000
Building 10,000,000
Total $510,000,000 Total $510,000,000

448 Chapter14 The Money Supply Process
on deposit at the Fed.) Both vault cash and reserve deposits at the Federal
Reserve banks are called legal reserves. A bank may also hold reserves not
recognized by the Federal Reserve System as legal reserves such as deposits
in other banks. Table 14.1 assumes the required reserve ratio is 20 percent.
With deposits of $500 million and a required reserve ratio of 20 percent,
required reserves are $100 million, so the bank has $100 million of required
reserves and zero excess reserves. We would say the bank is fully “loaned
up,” because it has no excess reserves available to increase loans. Its reserves
just equal the 20 percent of deposits specified by the reserve requirement.1
For simplicity, the T-account in Table 14.1 combines many of the assets
and liabilities that comprise the balance sheets of real-world banks. As Box
14.1 shows, banks in the banking system own a variety of types of loans
and hold deposits in forms that include savings deposits, small-denomination
time deposits and CDs, large-denomination CDs, and borrowings from other
banks or from the Fed. In Table 14.1, however, loans are the category of
assets we use to include all financial instruments owned by the bank. As
such, it includes consumer loans, mortgages, and government debt obliga¬
tions (U.S. government securities). The category labeled deposits includes
demand deposits and other forms of deposits.
Our simplified T-accounts will allow us to see how the entire banking
system responds to a change in its reserves. In conducting this analysis, we
will make the following assumptions (we will relax them in the next section):
1. All deposits are checkable deposits.

2. The reserve requirement is 20 percent of deposits.
30 Deposits are the only form of money (the public does not hold currency).
4JLoans are made only to private individuals and companies.
50 Banks do not want to hold any excess reserves.
How a Bank Responds to a Change in Reserves

from Within the Banking System
Suppose your dear Aunt Bertha gives you a gift—a check for $1 million
that is drawn from her account at Bank Two. You deposit the check in your
account at Bank One. Your new deposit affects Bank One’s balance sheet
as shown in Table 14.2. Part a indicates how the balance sheet looks com¬
pared to the initial balance sheet shown in Table 14.1. Part b shows the
changes in the balance sheet between Tables 14.1 and 14.2 to point out
1 In Chapter 6, we learned that with given bank facilities and a fixed number of loan officers, a
bank may be unable to immediately transform the reserves created from deposits into loans. In the
short run, loan applications may stack up on the desks of loan officers. In the long run, however,
banks can generally adjust to new deposits by increasing the use of other inputs such as labor and
physical capital, allowing them to lend out funds to their legal maximum. Henceforth we concern
ourselves with this long-run scenario where banks keep only those reserves prescribed by law.
The Balance Sheet of the Commercial

Banking Sector
The accompanying table shows the balance billion. Assets included reserves of $56 billion,
sheet of the U.S. commercial banking sector in which amounted to 7.8 percent of demand de¬
October 1992. Liabilities of the commercial posits. Liquid assets were $148 billion, invest¬
banking sector included demand deposits ($718 ment securities were $822 billion, loans were
billion), savings deposits ($739 billion), time de¬ $2,261 billion, and other assets accounted for
posits ($1,022 billion) and other liabilities, and the remaining $287 billion.
net worth ($1,095 billion) for a total of $3,574
Balance Sheet of All Commercial Banks, October 1992

Assets (billions of dollars) Liabilities (billions of dollars)
Reserve Assets Demand Deposits 718.2
Reserves with Federal
Reserve Banks 24.5 Savings Deposits 738.9
Vault Cash 31.7 Small Time Deposits 645.0
Total Reserves 56.2 Large Time Deposits 376.7
Liquid Assets 147.5 Other Liabilities 1,095.2
Investment Securities 822.3
Loans 2,260.8
Other Assets 287.2
Total 3,574.0 Total 3,574.0
Source: Federal Reserve Bulletin, January 1993, p. A19.
exactly which categories have changed. The $ 1 million check has increased
the liabilities of Bank One, since it now owes depositors $501 million for
their deposits. At the same time, the $1 million check will be cleared through
the Federal Reserve System, resulting in an increase in Bank One’s reserves
of $1 million. (The $1 million will be subtracted from the issuing bank’s
[Bank Two’s] reserve account at a Federal Reserve Bank, so total banking
system reserves are unchanged. We will examine the impact of this reduction
in reserves on Bank Two in a moment; for now we focus on how Bank One
responds to the additional $1 million of reserves.)
Part a shows the bal¬ 7;;

ance sheet of Bank Effect of a $1 Million Deposit on a Bank
One after it receives
$1 million in new de¬
posits. Part b shows
(a) Balance Sheet of Bank One
the changes in the
Immediately after Receiving a $1 Million Deposit
bank's deposits, re¬
serves, and loans after
the bank receives the Assets Liabilities
$1 million deposit.
Note that the bank
has yet to lend out Reserves $101,000,000 Deposits $501,000,000
any of the $1 million Required $100,200,000
deposit, so it has ex¬ Excess 800,000
cess reserves of
Loans 400,000,000 Net Worth 10,000,000
$800,000.
Building 10,000,000
Total $511,000,000 Total $511,000,000
(b) Changes in Balance Sheet of Bank One

Immediately after Receiving a $1 Million Deposit
Assets Liabilities
Reserves + $1,000,000 Deposits + $1,000,000

Required + $200,000
Excess + 800,000
Loans 0 Net Worth 0
Building 0
Total + $1,000,000 Total + $1,000,000
The additional reserves Bank One receives are not all required reserves.
Since deposits increased by $1 million required reserves increased by 20
percent of $1 million, or $200,000. Since total reserves increased by $1
million the difference—$800,000—represents excess reserves, reserves held
in excess of the legally prescribed minimum.
Of course, Bank One will not be satisfied with $800,000 in excess
reserves. It earns no interest on cash sitting in the vault or on reserve deposits
at the Fed. Thus, Bank One will eventually lend out the $800,000 of excess
reserves, reducing reserves but increasing loans. This leads to the situation
shown in Table 14.3. Again, part a indicates the balance sheet after the bank
adjusts its loan and reserve holdings, while part b indicates the change in
the balance sheet between the initial balance sheet shown in Table 14.1 and
the balance sheet shown in part a of Table 14.3. Bank One now has reserves
This table shows Bank Table 14.3

One's balance sheet Effect of a $1 Million Deposit and the Resultant Expansion
after the bank has lent of Loans at a Bank
out the $800,000 of
excess reserves. Part a
shows the balance
sheet adjusted for the (a) Balance Sheet of Bank One
deposit, the additional after Receiving a $1 Million Deposit and Making Loans
required reserves, and
the additional loans of Assets Liabilities
the bank. Part b
shows the changes in
the bank's balance Reserves $100,200,000 Deposits $501,000,000
sheet after lending out Required $100,200,000
the $800,000. Excess 0
Loans 400,800,000 Net Worth 10,000,000
Building 10,000,000
Total $511,000,000 Total $511,000,000

after Receiving a $1 Million Deposit and Making Loans
Assets Liabilities
Reserves + $200,000 Deposits + $1,000,000

Required + $200,000
Excess +0
Loans + 800,000 Net Worth 0
Building 0
Total + $1,000,000 Total + $1,000,000
of $100,200,000, which just equals 20 percent of its deposits of $501 million.

Excess reserves are zero, and loans are $400,800,000. Loans plus reserves
equals deposits, and net worth still just equals the value of the physical bank
building.
Notice that in going from Table 14.2 to Table 14.3, Bank One still holds
$501 million of deposits. It has simply rearranged the way it holds its assets,
transferring $800,000 from non-interest-bearing reserves to $800,000 of
interest-bearing loans. At this point, Bank One is again completely “loaned
up,” since it is holding the absolute minimum level of legal reserves. Any
additional reduction of reserves to fund additional loans will bring the bank
into violation of its legal reserve requirement. In this situation, it will be
unable to make any additional loans without first attracting new deposits.
The preceding analysis pertains to a bank in the happy position of

receiving a new deposit of $ 1 million. Remember, however, that the check
was drawn on another bank, Bank Two. How does Bank Two respond when
it faces the loss of $1 million of deposits?
Let us suppose that before the check was cashed, Bank Two’s balance
sheet was identical to the one in Table 14.1. When Aunt Bertha’s check
clears Bank Two, its deposits are reduced by this amount, as are its reserves.
This loss of reserves occurs when the check clears against Bank One’s
reserve account at the Fed. When this happens, $1 million is transferred
from Bank Two’s reserve to Bank One’s reserve account. For the bank
losing reserves—Bank Two in this example—reserves decline by the same
amount deposits decline, $1 million. Table 14.4 depicts this situation.
Part a shows the bal¬ Table 14.4

ance sheet of Bank Effect of a $1 Million Withdrawal on a Bank
Two after a with- 1
drawal of $1 million in
deposits. Part b shows
(a) Balance Sheet of Bank Two
the changes in the
bank's deposits, re¬
Immediately after Losing a $1 Million Deposit
serves, and loans after
it has lost the $1 mil¬ Assets Liabilities
lion in deposits. Note
that the bank has not
yet adjusted its loans Reserves $99,000,000 Deposits $499,000,000
so that it will have Required $99,800,000
enough reserves to Excess - 800,000
meet its reserve re¬
quirement; that is, it
Loans 400,000,000 Net Worth 10,000,000
has negative excess Building 10,000,000
reserves, or excess re¬
serves of - $800,000. Total $509,000,000 Total $509,000,000
(b) Changes in Balance Sheet of Bank Two

Immediately after Losing a $1 Million Deposit
Assets Liabilities
Reserves -$1,000,000 Deposits -$1,000,000

Required - $200,000
Excess - 800,000
Loans 0 Net Worth 0

Building 0
Total -$1,000,000 Total -$1,000,000

Deposits at Bank Two are now $499 million, and reserves are $99
million. This causes a problem for the bank, however, since at 20 percent
of deposits required reserves are now $99,800,000 but actual reserves are
only $99 million. Bank Two actually has negative excess reserves, meaning
it is short of reserves. The only way to obtain additional reserves is to call
in loans or borrow reserves from the Fed (at the discount rate) or from other
banks (at the federal funds rate). We will concentrate here on the impact of
calling in loans.
How does a bank call in loans? We tend to think of loans as long-term
loans, such as 15- or 30-year mortgage loans or 4- or 5-year automobile
loans. In fact, however, such loans make up only part of a bank’s assets.
Banks also make short-term loans to businesses, some of which are due
“upon demand.” Thus, a bank in a position of negative excess reserves can
cover its reserve position by calling in the appropriate amount of loans, that
is, by demanding payment of principal and interest. In addition, a large bank
may have a whole series of short-term loans coming due each day and can
simply decline to renew the appropriate amount of those loans to cover its
shortage of reserves.
In Table 14.4, Bank Two needs $800,000 of additional reserves, which
it obtains by calling in $800,000 in loans. After doing this, the bank finds
itself in the position shown in Table 14.5. Reserves have increased to
$99,800,000, just equal to 20 percent of deposits. Loans have fallen by
$800,000 and now stand at $399,200,000.
Needless to say, this contraction of loans is painful for the bank. Calling
in loans, or even failing to renew loans, is not a good way to make customers
happy. The alternative is to borrow reserves, either from other banks in the
federal funds market or from the Federal Reserve System, by going to the
discount window.
We can summarize the effect of a change in deposits and reserves at a
single bank very simply. An increase in deposits and reserves leads to excess
reserves, which a bank will use to increase loans. A decrease in deposits and
reserves leads to a shortfall of reserves, which a bank will meet by reducing
loans or borrowing reserves.
What is the effect of these transactions on the money stock? Recall that
the money stock is the sum of currency and checkable deposits, and we have
been assuming people use only checks to make transactions. In this case,
the total money stock is the amount of checkable deposits available. Since
the $1 million check was written from a private account in one bank to a
private account in another, one bank’s loss is another’s gain, and there is no
net change in total banking system reserves, deposits, or loans. Thus, the
total money stock remains constant when a check written at one bank is
deposited in another bank. As we will see in the next section, the effect on
the money stock differs when the initial deposit comes from outside the
banking system rather than from another bank.
454 Chapter 1 The Money Supply Process
This table shows Bank

Two's balance sheet
after it has decreased
its lending to gain re¬
serves. Part a shows
the balance sheet af¬
ter the bank has re¬
(a) Balance Sheet of Bank Two
called $800,000 in after Losing a $1 Million Deposit and Adjusting Loans
loans and used this : , . v
amount to increase re¬ Assets Liabilities

serves. Part b shows
the changes in the
bank's balance sheet Reserves $99,800,000 Deposits $499,000,000
after decreasing its Required $99,800,000
lending by $800,000. Excess 0
Loans 399,200,000 Net Worth 10,000,000
Building 10,000,000
Total $509,000,000 Total $509,000,000
(b) Changes in Balance Sheet of Bank Two Immediately

after Losing a $1 Million Deposit and Adjusting Loans
Assets Liabilities
Reserves - $200,000 Deposits -$1,000,000

Required - $200,000
Excess 0
Loans - 800,000 Net Worth 0
Building 0
Total - $ 1,000,000 Total -$1,000,000
Summary What is the effect on a bank’s balance sheet of a withdrawal of $10,000?

Exercise 14.1 Assume the required reserve ratio is 5 percent, the bank held no excess
reserves at the time of the withdrawal, and the bank is unable to borrow
reserves.
The withdrawal of $10,000 reduces both bank reserves and bank

deposits. Deposits fall by $10,000, which reduces required reserves by 5
percent of $10,000, or $500. Thus, the initial effect is to reduce reserves by
$10,000, of which only $500 is a reduction in required reserves. The re¬
maining $9,500 reduction is a reduction in excess reserves, indicating the
bank has a shortfall of this amount of required reserves. The bank must call
in loans of this amount to meet its reserve requirements. After calling in
those loans, the bank’s balance sheet will show that deposits declined by
$10,000, reserves by $500, and loans by $9,500.
How the Banking System Responds to Infusions

and Contractions of Reserves
As we just saw, a single bank that receives a new deposit will transfer the
new reserves into loans to earn interest on its assets. We also saw that when
the deposit comes from another bank, no change occurs in banking system
deposits, reserves, or loans, since one bank’s gain is another’s loss. In this
section, we will see that when a bank receives a new deposit from outside
the banking system, such as from the Federal Reserve System, the additional
deposit will increase the deposits, reserves, and loans of the entire banking
system. We will also see that deposits and loans increase by a multiple of
the initial infusion of reserves. Our focus on the entire banking system
distinguishes this chapter, and indeed the rest of the book, from the micro
analysis of individual banks in Chapters 5 and 6.
We emphasize that in contrast to the previous section, in which the
initial change in deposits and reserves was merely a reshuffling of existing
deposits and reserves among banks, this section examines the implications
of an infusion (increase) or contraction (decrease) in banking system deposits
and reserves. Such infusions or contractions are caused by the day-to-day
operations of the Fed. For example, the Fed can alter total banking system
reserves and deposits by (1) lending reserves to banks through the discount
window or (2) engaging in an open market operation. (Recall that an open
market operation is the purchase or sale of U.S. government securities by
the Federal Reserve System in the open market.) Either of these actions will
change a bank’s reserves and deposits without creating a corresponding
change in the opposite direction at another bank. As we will see next, such
changes in reserves from outside the banking system will affect the total
money stock; these transactions do not represent a reshuffling of reserves
among banks.
Open Market Purchases: Multiple Deposit Expansion. In an

open market purchase, the Federal Reserve System—more specifically, the
Open Market Desk at the Federal Reserve Bank of New York—contracts to
purchase a set amount of U.S. government securities. The Federal Reserve
System pays for these securities with a check or an electronic transfer into
the account of the securities dealer, and the dealer delivers the securities to
the Fed. To be concrete, suppose the Fed purchases $1 million in bonds and
pays by electronically transferring funds to the security dealer’s bank account
in a bank we will call Bank One. Then Bank One has an increase in both
reserves and deposits of $1 million. The Fed has an increase in assets of $1
million and an increase in liabilities, since the bank’s reserve account is a
liability of the Fed.
The change in both the Fed’s and Bank One’s balance sheets is shown
in Table 14.6. The Fed increases its holding of U.S. government securities,
an asset, and its reserve deposits, a liability. Bank One increases both its
deposits and its reserves. With a required reserve ratio of 20 percent, the
increase in required reserves is $200,000, so the remainder of the increase
in reserves, $800,000, is excess reserves.
Notice the similarity between the change in Bank One’s balance sheet
in Table 14.6 and that in part (b) of Table 14.2. In both cases, the bank
received a deposit and new reserves of $1 million. The effect on the indi¬
vidual bank is the same. What is special about the situation in Table 14.6 is
Part a shows the changes in the Fed's balance sheet after purchasing $1 million worth of
government securities. Its assets and liabilities rise by $1 million. The Fed pays the securities
dealer, who deposits the payment in Bank One, where deposits increase by $1 million.
Bank One itself sends the check to the Fed, where it is deposited in Bank One's reserve
account, increasing its reserves by $1 million. Part b shows the changes in the balance
sheet of Bank One after the Fed's $1 million open market purchase. Notice these changes
occur before Bank One has lent out any of its $800,000 in excess reserves.
(a) Changes in Balance Sheet of the Fed

Immediately after a $1 Million Open Market Purchase
Assets Liabilities
U.S. Government + $1,000,000 Reserves + $1,000,000

Bonds
(b) Changes in Balance Sheet of Bank One Immediately

after Gaining $1 Million Through a Fed Open Market Purchase
Assets Liabilities
Reserves + $1,000,000 Deposits + $1,000,000

Required + $200,000
Excess + 800,000
Loans 0 Net Worth 0
Building 0
Total + $1,000,000 Total + $1,000,000
that the increase in reserves at the bank shown there is not countered by a
decrease in reserves at some other bank. Instead, the increase in reserves is
due to a net infusion of new reserves into the banking system, an infusion
caused by an action of the Federal Reserve System. This infusion of reserves
creates a large adjustment throughout the banking system as banks scramble
to adjust to the new level of reserves.
To see these adjustments, look again at part (b) of Table 14.6, and note
that Bank One has $800,000 of excess reserves. The bank will lend out these
reserves as it transforms non-interest-bearing reserves into interest-bearing
loans. When it does so, it ends up in the position illustrated in part a of
Table 14.7.
What happens to the $800,000 of excess reserves that Bank One just
Part a shows the Table 14.7

changes in Bank One's Effect of the Expansion of Loans at Bank One
balance sheet after and Initial Effect on Bank Two
the bank has lent out
all of its $800,000 of
excess reserves. This
(a) Changes in Balance Sheet of Bank One
$800,000 ends up be¬
ing deposited in Bank
after a $1 Million Open Market Purchase and Adjusting Loans
Two, where deposits
and reserves increase Assets Liabilities
by this amount. Bank
Two now has
$640,000 in excess re¬ Reserves + $200,000 Deposits + $1,000,000
serves, as shown in Required + $200,000
part (b). Excess +0
Building 0
Total + $1,000,000 Total + $1,000,000
(b) Changes in Balance Sheet of Bank Two

Immediately after Receiving an $800,000 Deposit
Assets Liabilities
Reserves + $800,000 Deposits + $800,000

Required + $160,000
Excess + 640,000
Loans 0 Net Worth 0
Building 0
Total + $800,000 Total + $800,000

lent out? Consider in particular the borrower, for example, a real estate
developer who borrows the $800,000 to buy real estate in Florida. When she
does so, she gains property worth (she hopes) $800,000 and the seller gains
the $800,000. The seller himself will want to deposit the $800,000 in his
bank, which we will call Bank Two. Thus, Bank One’s loan ends up as a
new deposit and new reserves in Bank Two. The change in Bank Two’s
balance sheet after this new deposit is shown in part b of Table 14.7. Note
that the deposit increased both deposits and reserves at Bank Two by
$800,000. Of course, with a required reserve ratio of 20 percent, the increase
in required reserves is only $160,000, leaving the remaining $640,000 of
new reserves as an increase in excess reserves.
Again, however, we note that excess reserves earn no interest and Bank
Two will want to turn the $640,000 of excess reserves into loans. Once this
is accomplished, the change in Bank Two’s balance sheet is as illustrated in
part a of Table 14.8. Notice that Bank Two behaved similarly to Bank One.
Both received an influx of deposits and reserves, and since they are required
to hold only 20 percent of deposits as reserves, both moved to lend out the
influx of excess reserves. The only difference is that Bank One received a
new deposit of $1 million, while Bank Two received a new deposit of
$800,000, which was the amount lent out by Bank One.
Of course, the story doesn’t end here. Bank Two has just lent $640,000
to someone. These funds will be spent—after all, few people would borrow
such a sum without a specific purpose in mind—and, once spent, will be
redeposited in someone else’s bank account at, say, Bank Three. This will
lead to excess reserves at Bank Three of $512,000, as shown in part b of
Table 14.8. Bank Three will keep a fraction of the money as reserves and
lend out the remainder, which will be deposited in yet another bank. This
process will continue, and at each stage the bank receiving an increase in
deposits and reserves will hold the required 20 percent of the additional
deposits as required reserves and lend out the remaining 80 percent. Table
14.9 shows the final situation after all adjustments have been made in the
entire banking system: Total reserves in the banking system increase by $1
million, total loans increase by $4 million, and total deposits increase by $5
million.
To summarize, the initial $1 million infusion of reserves into the banking
system, caused by the Fed’s open market purchase, leads to multiple deposit
expansion—an increase in total banking system deposits that is a multiple
of the initial $1 million. This expansion is due to the simple deposit multiplier
process, also called the simple money multiplier process. Both terms are
appropriate because in this simple case, all money is held in the form of
deposits. (Since we assumed the public holds no currency, money and de¬
posits are synonymous.) The terms indicate that this is the process by which
the commercial banking system creates money. The raw material in the
money creation process is reserves, and these are supplied by the Federal
Reserve System. But the entire banking system transforms an injection of
1.; "
Part a shows the Table 14.8
changes in Bank Two's Effect of the Expansion of Loans at Bank Two
balance sheet after and the Initial Effect on Bank Three
the bank has lent out
all of the $640,000 of
excess reserves. This
(a) Changes in Balance Sheet of Bank Two
$640,000 ends up be¬
ing deposited in Bank
after an $800,000 Deposit and Adjusting Loans
Three, where deposits
and reserves increase Assets Liabilities
by this amount. Bank
Three now has
$512,000 in excess re¬ Reserves + $160,000 Deposits + $800,000
serves. Required + $160,000
Excess +0
Building 0
Total + $800,000 Total + $800,000
(b) Changes in Balance Sheet of Bank Three

Immediately after a $640,000 Deposit
Assets Liabilities
Reserves + $640,000 Deposits + $640,000

Required + $128,000
Excess + 512,000
Loans 0 Net Worth 0
Building 0
Total + $640,000 Total + $640,000

fiisillllllllilla
reserves into new loans that ultimately increase banking system deposits by
a multiple of the increase in reserves, as summarized at the bottom of the
last column in Table 14.9.
The Simple Deposit Multiplier, How do we know the $1 million

increase in reserves will lead to the final situation in Table 14.9, where total
banking system deposits rise by $5 million? Very simply, the lending process
will end only when all banks are fully loaned up so that excess reserves are
zero. Since the banking system has received an inflow of $1 million of new
reserves, all reserves will be required reserves when deposits grow by enough
that $1 million is 20 percent of deposits. Using rr to signify the required
reserve ratio, AD for the change in deposits, and AR for the initial change
This table summarizes Table 14.9

how the balance Summary of Deposit Expansion Process Across Banks
sheets of banks are af¬
fected after each
round of lending. It Change in
assumes banks hold Change in Assets Liabilities
no excess reserves, the
required reserve ratio Change in Change in Change in
is 20 percent, the Bank Reserves Loans Deposits
public does not hold
currency, and the ini¬ One + $200,000 + $800,000 + $1,000,000
tial $1 million deposit Two + 160,000 + 640,000 + 800,000
at Bank One was due
Three + 128,000 + 512,000 + 640,000
to an open market
Four + 102,400 + 409,600 + 512,000
purchase.
• • • •
• • • •
• • • •
Total for entire + $1,000,000 + $4,000,000 + $5,000,000

banking system
in reserves, the banking system will no longer be able to make additional

loans when
AD X rr = AR.
Solving this equation for AD gives us
AD = — X AR. (14.1)
rr
In this case, AR = $1,000,000 and rr = .20, so
AD = - X $1,000,000
.2
= 5 X $1,000,000 (14.2)
= $5,000,000.
That is, when deposits have increased to $5 million, the banking system will
be in equilibrium, with no further lending activity possible.
Equation 14.1 incorporates the simple deposit multiplier, 1/rr, which
is the multiple by which deposits increase in response to any increase in
reserves. The simple deposit multiplier is the inverse of the required reserve
ratio, so increases in the required reserve ratio will lower the deposit mul¬
tiplier, while decreases in the ratio will increase the deposit multiplier. In
our example, the simple deposit multiplier is 5, meaning total banking system
deposits increase by five times the initial change in reserves. A higher
required reserve ratio—say, rr = .5—would lead to a lower simple deposit
multiplier of 2 (= 1/.5). In other words, had the required reserve ratio been
.5 instead of .2, total banking system deposits would have increased by only
two times the initial change in reserves, or by $2 million. In case you’re
wondering, Box 14.2 shows the actual reserve requirements set by the Fed
as of January 1993.
An alternative way to verify the final situation shown in Table 14.9 is
to add up the changes in deposits at all banks. Notice in Table 14.9 that
deposits at Bank One increase by $1 million, deposits at Bank Two increase
by $800,000 (which is [1 — rr], or 80 percent of $1 million), deposits at
Bank Three increase by $640,000 (80 percent of $800,000), and so on. In
each round of lending, deposits at the next bank increase by 80 percent of
the increase at the previous bank. If we write out the change in deposits at
the first four banks and use ellipses to indicate the change in deposits at all
remaining banks, we obtain
AD = $1,000,000 + $800,000 + $640,000 + $512,000 + . . .
= $1,000,000 + (.8)$ 1,000,000 + (.8)$800,000 + (.8)$640,000 + . . .
= $1,000,000 + (.8)$ 1,000,000 + (.8)(.8)$ 1,000,000
+ (.8)(.8)(.8)$ 1,000,000 + . . .
This may be further simplified as
AD (1 + .8 + .82 + .83 + • • •) X $1,000,000
1
X $1,000,000
1 - .8
= 5 X $1,000,000
= $5,000,000.2
This approach thus tells us deposits increase by $5 million—the same answer

we obtained using equation 14.2.
How do we know total banking system loans in Table 14.9 increase by
$4 million? Since each bank has a balance sheet, and we can aggregate these
balance sheets to construct a balance sheet for the entire banking sector, we
know the change in assets must equal the change in liabilities. For the entire
banking system, liabilities in the form of deposits have grown by $5 million.
2 To calculate the total change in deposits, we need to know the value of the series (1 + .8 + .82
+ .83 + . . .), where the ellipses signify that the series continues forever, with successively larger
exponents on the fraction .8. This is called an infinite series, and there is a mathematical formula
for calculating the sum of this series given by
(1 + x + x2 + x3 + • • •) = .
1 — X
This formula is valid for values of x less than 1. In our example, x = .8.
Reserve Requirements in the United States
The Board of Governors of the Federal Reserve The accompanying table shows the reserve re¬
System establishes reserve requirements in the quirements in effect in early 1993. The reserve
United States within limits set by Congress, and requirements on checking accounts require
these reserves must be held as deposits with small banks (those with less than $46.8 million
one of the 12 Federal Reserve banks or as vault in deposits) to hold 3 percent of deposits as re¬
cash. Under provisions of the Depository Dereg¬ quired reserves, while larger banks (those with
ulation and Monetary Control Act of 1980, de¬ more than $46.8 million in deposits) must hold
pository institutions that must meet these re¬ 10 percent of deposits in reserve. Time deposits
serve requirements include commercial banks, and other forms of deposits are not subject to
mutual savings banks, savings and loan associa¬ reserve requirements.
tions, credit unions, and agencies and branches
of foreign banks.
Reserve Requirements
Reserve Requirement
Type of Deposit (percent of deposit) Effective Since
Checking accounts at banks 10% April 4, 1992

with more than $46.8 million
Checking accounts at banks 3 December 15, 1992
with less than $46.8 million
Other deposits (time deposits 0 December 27, 1992
and Eurocurrency liabilities)
Source: Federal Reserve Bulletin, January 1993, p. A9.
Assets in the form of reserves have increased by $1 million. This means

assets in the form of loans must have grown by $4 million. Indeed, after the
initial increase in reserves due to the open market purchase, the growth in
loans is what leads to the growth in deposits. Thus, apart from the initial $1
million, the change in loans should be the same as the change in deposits.
Open Market Sales: Multiple Deposit Contraction. An open

market sale has the exactly opposite effect from an open market purchase:
It leads to a contraction of banking system reserves. During an open market
sale of $ 1 million in government bonds, the Fed sells securities to a securities

dealer, who likely pays by check. When the Fed cashes that check, it reduces
reserves and deposits at Bank One. This causes Bank One to have negative
excess reserves of $800,000, which can be covered temporarily by borrowing
but will eventually lead the bank to recall $800,000 of its outstanding loans.
Table 14.10 illustrates the positions of the Fed and the bank after the open
market sale.
How does this open market sale affect the entire banking system? When
Bank One recalls $800,000 in loans, these loans are paid by borrowers
writing checks on accounts at, say, Bank Two. Bank Two will thus see a
Part a shows the changes in the Fed's balance sheet after it has sold $1 million in govern¬
ment bonds. Its assets are reduced by this amount, as are its liabilities. Part b shows the
effect on Bank One, the securities dealer's bank. The dealer writes a check on her deposit
account to purchase the bonds from the Fed, and the Fed cashes this check by reducing
Bank One's reserve account. Thus, Bank One has a reduction in deposits and reserves of $1
million. Notice that with this reduction in reserves, the bank now has negative excess
reserves, or excess reserves of - $800,000.
(a) Balance Sheet of the Fed

Immediately after a $1 Million Open Market Sale
Assets Liabilities
U.S. Government —$1,000,000 Reserves -$1,000,000

Bonds

Immediately after a $1 Million Open Market Sale
Assets Liabilities
Reserves -$1,000,000 Deposits -$1,000,000

Required - $200,000
Excess - 800,000
Loans 0 Net Worth 0
Building 0
Total -$1,000,000 Total -$1,000,000

reduction in its deposits and reserves of $800,000, meaning this bank has a
$640,000 shortfall of required reserves that it must meet by recalling loans.
This process continues and is exactly the opposite of what happened when
reserves were added to the banking system. In fact, the same deposit mul¬
tiplier holds in reverse; that is, when the banking system loses $1 million in
reserves, we can still calculate the change in deposits as
AD = — X AR.
rr
In this case, AR = —$1,000,000 and rr = .2, so
AD = X -$1,000,000
= 5 X -$1,000,000
= -$5,000,000.
The $5 million contraction of deposits is driven by the contraction in reserves

thanks to the open market sale and the subsequent reduction in loans through¬
out the entire banking system.
An important feature of this analysis is that the reduction in reserves
engineered by the Fed is not eliminated by the actions of any one bank.
Indeed, this reduction in reserves stays at $1 million throughout the adjust¬
ment process. The adjustment process simply spreads this reduction through¬
out the banking system.
Summary What is the total contraction of deposits, loans, and reserves in the entire
Exercise 14.2 banking system if the required reserve ratio is 5 percent and the Fed makes
an open market sale of $150,000? What happens to the money stock? Re¬
member, we are assuming banks do not want to hold any excess reserves,
the public does not hold cash, and banks do not borrow reserves.
Answer: The open market sale decreases reserves by $150,000. As we

have seen, this will be the total change in banking system reserves. We can
calculate the change in deposits from our simple deposit multiplier formula:
AD = — X -$150,000
.05
= 20 X -$150,000
= -$3,000,000.
Notice that the simple deposit multiplier is 20, much higher than it was when
the required reserve ratio was .2.
To calculate the change in loans, we simply note the change in loans
plus the change in reserves must equal the change in deposits to keep the
balance sheet in balance. The change in reserves is —$150,000 and the
change in deposits is — $3,000,000, so the change in loans equals the change

in deposits minus the change in reserves, or —$2,850,000. Finally, recall
that the money stock is the sum of currency holdings and checkable deposits.
Since deposits are the only form of money in this example, the money stock
decreases by the same amount as deposits, or by $3 million.
A General Model of Money Creation
In the previous section, we derived the simple deposit multiplier as the

inverse of the required reserve ratio. We did this under the assumptions that
all bank reserves are required reserves and the public does not hold currency.
In this very simple setting, the money stock is simply the amount on deposit
at banks; the simple deposit multiplier could equivalently be thought of as
the simple money multiplier.
In reality, the public holds currency as well as deposits, so the total
money stock (M) is the sum of currency (C) and deposits (D):
M = C + D. (14.3)
In this section, we examine money creation in this more general context and
see how the level of deposits and the amount of currency holdings are
determined. Ultimately this will allow us to examine how actions of the Fed,
including open market operations, affect the total stock of money in the
economy.
We first relax the assumptions used in the previous section. In particular,
we now assume the public desires to hold currency in proportion to deposits
so that people who decide to hold more money would increase their cash
and deposit holdings proportionally. This is in contrast to the previous sec¬
tion, where we assumed people held money only in the form of deposits.
We indicate the desired currency to deposit ratio, cd, so that
C = cd X D. (14.4)
We also allow banks to hold excess reserves in addition to reserves required

by law. Thus, total reserves (R) equal required reserves (RR) plus excess
reserves (ER):
R = RR + ER. (14.5)
Required reserves still equal the required reserve ratio, rr, times deposits:
RR = rr X D. (14.6)
Desired excess reserves are assumed to be proportional to deposits so that

desired excess reserves are equal to the desired excess reserve ratio, ed,
times deposits:
ER = ed X D. (14.7)
Together equations 14.5, 14.6, and 14.7 imply that total bank reserves are
R = RR + ER
= rr X D + ed X D (14.8)
= (rr + ed) X D.
Equation 14.8 indicates that when the total banking system holds a constant
fraction of banking system deposits as required and excess reserves, total
banking system reserves are proportional to deposits.
The Monetary Base

Since the public holds currency, the Fed cannot supply funds that will be
earmarked as commercial bank reserves. Instead, funds supplied by the Fed
get divided among two uses: as commercial bank reserves and as currency
in the hands of the public.
Because funds supplied by the Fed are split among these two uses, it is
not correct to view all the funds the Fed supplies as reserves, since a portion
winds up in currency held outside of banks. Instead these funds are called
the monetary base (MB). The monetary base (sometimes also referred to
as high-powered money) is currency in the hands of the public (C) plus
commercial bank reserves (R):
MB = C + R. (14.9)
An example will illustrate why it is convenient to use the monetary base

to analyze Fed open market operations and how currency holdings and excess
reserves affect money creation. Suppose a Fed open market purchase leaves
a securities dealer holding a check for $1 million, which he deposits in his
bank. If the securities dealer did not wish to hold cash, this deposit would
increase the bank’s reserves by $1 million. Furthermore, if the required
reserve ratio is 20 percent and the bank holds no excess reserves, it can lend
out $800,000 of the new deposit. In the previous section, we saw how this
is multiplied throughout the banking system through subsequent rounds of
deposits to create money in the form of deposits.
Now suppose the securities dealer wants to keep his currency holdings
equal to one-quarter of his checkable deposits (i.e., cd = .25) so that when
he deposits the $1 million check, he takes $200,000 out in currency.3 Due
to this cash withdrawal, the bank has gained reserves (deposits) of $800,000.
This illustrates that the initial effect of an open market purchase of $1 mil¬
lion results not in $1 million in new reserves at the bank (as it did in the
absence of currency holdings) but in new reserves of $800,000 and new
cash holdings by the public of $200,000. For this reason, it is more
3 Notice that by withdrawing $200,000, he has $800,000 left on deposit. Thus, his currency to
deposit ratio is $200,000/$800,000, or .25.
convenient to think of the open market purchase as increasing currency

plus reserves (the monetary base) than to figure out how much of the ini¬
tial deposit is converted into currency and how much is in the form of
reserves.
After forking over $200,000 of currency to the securities dealer, how
much of the remaining $800,00 can the dealer’s bank lend out? The required
reserve ratio of 20 percent means $160,000 of this $800,000 must be kept
on reserve, which leaves the bank only $640,000 to lend out. Furthermore,
if the bank’s policy is to keep excess reserves of 5 percent in case of an
emergency, it will keep another $40,000 as reserves, resulting in even less
available to lend out—$600,000. Moreover, in each subsequent round of
lending and deposits of the loans in other banks, currency holdings and
excess reserves continue to diminish the additional loans banks can make.
This example suggests that a $1 million open market purchase will
increase bank loans (and hence deposits at other banks), but by less than
would be the case in the absence of currency holdings and excess reserves.
It also illustrates that such actions by the Fed will have the additional effect
of increasing currency in the hands of the public. In the remainder of this
section, we determine the precise impact of Fed open market operations on
(1) total banking system deposits, (2) total currency holdings by the public,
and (3) the total stock of money.
The Complete Deposit Multiplier

We now derive the relationship between the monetary base and total banking
system deposits, being careful to take into account how currency holdings
and excess reserves drain the banking system of reserves. First, we substitute
equations 14.4 and 14.8 into equation 14.9 to express the monetary base as
MB = C + R = cdD + (rr + ed)D,
which may be rewritten as
MB = (rr + ed + cd) X D. (14.10)
That is, the monetary base is total banking system deposits (D) multiplied
by the sum of the currency to deposit ratio (cd), the required reserve ratio
(rr), and the desired excess reserve ratio (ed).
Next, we solve equation 14.10 for D to obtain the total amount of
deposits that are generated with a given monetary base:
1
D = X MB. (14.11)
rr + ed + cd
Equation 14.11 completely determines total banking system deposits given

the parameters of the banking system. It says that the total amount of banking
system deposits is actually a multiple of the monetary base. For instance,
suppose the Fed sets the monetary base at $350 billion and required reserves
at 20 percent of deposits. If banks hold 5 percent of deposits as excess

reserves and the public holds 25 percent of deposits as currency, total bank¬
ing system deposits are
1
D = X $350,000,000,000
.2 + .05 + .25
= 2 X $350,000,000,000 (14.12)
= $700,000,000,000.
Notice that total banking system deposits are two times the monetary base.
How would total banking system deposits change if the Fed increased
or decreased the monetary base through open market operations? An open
market purchase would increase the monetary base, allowing banks to issue
more loans. This effect would multiply throughout the banking system to
create additional deposits according to the relation
1
AD X AMB. (14.13)
rr + ed + cd
The term in brackets in equation 14.13 is called the complete deposit

multiplier and indicates the multiple by which a given change in the mone¬
tary base will increase total banking system deposits. It is a generalization
of the simple deposit multiplier introduced in equation 14.1. To see this,
notice that when the public does not hold currency and banks do not hold
excess reserves, cd = ed = 0. In this case, the complete deposit multiplier
in equation 14.13 reduces to the simple deposit multiplier, 1 /rr, presented in
equation 14.1.
We can use the complete deposit multiplier to determine how much total
banking system deposits increased when the Fed purchased $1 million in
bonds from our securities dealer. Recall that this open market purchase
increased the monetary base by A MB = $1,000,000. Plugging rr = .2, ed
= .05, and cd = .25 into equation 14.13, we find total banking system
deposits changed by
1
AD X $1,000,000
.2 + .05 + .25 (14.14)
= 2 X $1,000 = $2,000,000.
Here the complete deposit multiplier is 2, and thus total banking system
deposits increase by two times the change in the monetary base, or by $2
million. Notice this is considerably less than the increase in deposits we
found in equation 14.2, which was relevant when cd = ed = 0. (In that case,
recall deposits increased by a whopping $5 million.) Why the difference?
First, we now assume banks hold excess reserves in addition to the
reserves required by law. This means banks keep more of their deposits on
reserve and thus have less to lend out when the Fed increases banking system
reserves through an open market purchase. The less lent out, the less will be
deposited after each subsequent round of lending, leading to fewer deposits
than was the case when banks held only required reserves.
Second, individuals now wish to hold currency as well as deposits. In
particular, when the $1 million check is deposited at the securities dealer’s
bank, the securities dealer withdraws $200,000 to hold as currency. This
reduces the reserves of the bank, thus allowing it to make fewer loans than
it would otherwise be free to make. Excess reserves and currency held by
the public drain banking system reserves, resulting in fewer total deposits
than would otherwise be the case.
More generally, the complete deposit multiplier decreases with increases
in either the currency to deposit ratio, the required reserve ratio, or the
desired excess reserve ratio. This implies the complete deposit multiplier—
which takes into account currency holdings and excess reserves—is less than
the simple deposit multiplier.
The Complete Currency Multiplier

When the public holds currency, an open market purchase leads not only to
an increase in banking system deposits (as shown above) but also to more
currency in circulation. Just how much currency is put in the hands of the
public when the Fed makes a $1 million open market purchase? To answer
this question, we first derive a relationship between the monetary base and
currency holdings.
To do this, we solve equation 14.4 for D and substitute the result into
equation 14.11 to obtain the following relationship between the monetary
base and currency:
C = X MB. (14.15)
rr + ed + cd
Given the parameters of the banking system, equation 14.15 completely

determines the currency holdings of the public. It says the total amount of
currency held by the public is a fraction of the monetary base. For instance,
if MB = $350 billion, rr = .2, ed — .05, and cd = .25, total currency
holdings by the public are
.25
C = X $350,000,000,000
.2 + .05 + .25
= ^ X $350,000,000,000 (14.16)
$175,000,000,000.
In other words, the Fed must maintain a $350 billion monetary base to keep
$175 billion in currency in the hands of the public. Remember, some of the
monetary base ends up as reserves at banks (and at the Fed), which is why
the currency held by the public is actually less than the monetary base.
A change in the monetary base will lead to a change in currency holdings
that is a fraction of the initial increase in the monetary base:
AC X A MB. (14.17)
rr + ed + cd
The relationship between the change in the monetary base and the change
in currency held by the public in equation 14.17 involves the complete
currency multiplier, cd/(rr + ed + cd). We can use this multiplier to
determine by how much currency holdings increase due to a change in the
monetary base.
Suppose the Fed wants to increase the amount of currency in circulation.
By how much would currency in the hands of the public increase if the Fed
engaged in an open market purchase that raised the monetary base by $1
million? If rr = .2, ed = .05, and cd = .25, then
.25
AC = X $1,000,000
.2 + .05 + .25 (14.18)
1
= - X $1,000,000 = $500,000.
In other words, the $ 1 million increase in the monetary base has resulted in
an increase in total currency holdings of $500,000. Since the complete
currency multiplier is 1/2, currency holdings increased by less than the
increase in the monetary base. It is important to note that this $500,000
increase in currency holdings is in addition to the $2 million increase in
deposits shown in equation 14.14 and thus is an additional increase in the
money stock.
The Complete Money Multiplier

How is the money stock related to the monetary base? Since the money
stock consists of currency in the hands of the public plus checkable deposits,
we can use the preceding results to find the link. Recall that total banking
system deposits are
1
D = X MB (14.19)
rr + ed + cd
and total currency in the hands of the public is
C = X MB. (14.20)
rr + ed + cd
Since the money stock is simply the sum of currency and deposits (M
= C + D), we can add equations 14.19 and 14.20 together to get
1 + cd
M = X MB. (14.21)
rr + ed + cd
Equation 14.21 is very important because it shows how the total stock of
money is a multiple of the monetary base. The multiple (the term in brackets)
is called the complete money multipier and depends on the desired currency
to deposit ratio, the required reserve ratio, and the desired excess reserve
ratio. For instance, suppose the Fed maintains a monetary base of $350
billion and requires banks to keep 20 percent of deposits as required reserves.
If banks keep 5 percent of their deposits as excess reserves and the public
holds 25 percent of their deposits as cash, the total stock of money in the
economy is
1 + .25
M = X $350,000,000,000
.2 + .05 + .25
- 2.5 X $350,000,000,000
- $875,000,000,000
This shows how actions of the Fed, banks, and the public determine the total
stock of money in the economy. Since the money stock consists of currency
plus deposits, it is no coincidence that this amount exactly equals the sum
of currency and deposits we calculated earlier in equations 14.12 and 14.16:
M = C + D = $175,000,000,000 + 700,000,000,000 = $875,000,000,000.
In other words, by maintaining a monetary base of $350 billion, the Fed

actually ensures that the economy has $875 billion in money (currency plus
checkable deposits). Box 14.3 discusses how the complete money multiplier
we just derived is related to the actual Ml and M2 money multipliers for
the U.S. economy.
Naturally, if the Fed changes the monetary base, the amount of currency
and checkable deposits will change, since banks will alter their reserves and
the public will change their currency holdings. The change in the money
stock that results from a change in the monetary base is given by
1 + cd
AM X A MB. (14.22)
rr + ed + cd
Since the change in the monetary base is multiplied by the complete money
multiplier, the money stock (currency plus deposits) changes by a multiple
of the change in the monetary base.
Not only does this example demonstrate the power the Fed has in chang¬
ing the total amount of money in the economy, we can also use this formula
to calculate the total amount of money created when the Fed purchased $1
The M2 Multiplier
Our analysis of the money multiplier illustrates tutions), overnight repurchase agreements, and
that the total supply of money in the economy is overnight Eurodollars. As we learned in Chap¬
actually a multiple of the monetary base. For ters 7 and 13, the DIDMCA blurred the distinc¬
simplicity, the analysis in the text is based on a tion between banks and thrifts. This fact,
very narrow definition of the money supply, one coupled with the volatility of Ml during the
that resembles Ml. Recall from Chapter 2 that 1980s, led the Fed to largely abandon Ml and
Ml is defined as currency plus demand deposits use its policy tools to target M2. The multiplier
and other forms of checkable deposits, as well analysis described in the text also applies to M2,
as traveler's checks. The complete money multi¬ but the M2 multiplier is larger because of the
plier discussed in the text is, for all practical pur¬ additional components it contains.
poses, the Ml multiplier, since it includes all of We can gain a feel for how the M2 multi¬
the components of Ml except traveler's checks plier differs from that for Ml by lumping to¬
(which tends to be small amounts). gether several components in M2 to write
A broader monetary aggregate is M2,
M2 = C + D + T + MM,
which adds to Ml savings deposits and time de¬
posits, money market funds (not owned by insti¬ Continued on p. 473
Continued from p. 472 are no reserve reguirements on these compo¬

where, as before, C represents currency, D is nents of M2, financial institutions create loans
demand deposits, T represents time deposits, with these types of deposits that expand more
and MM is the remaining components of M2, readily through the banking system, resulting ulti¬
including money market funds. If we let mately in a much larger money supply as meas¬
cr = 77D and c™ = MM/D, the broader M2 ured by M2.
measure of the money supply is given by the M2 The accompanying figure shows the M2
multiplier times the monetary base: multiplier for the 1972-1993 period. Notice it
generally exceeded 10 during this time, indicat¬
1 + <ri + cf + cm ing that a given monetary base is multiplied
M2 x MB.
rr + ed + ^ throughout the banking system to create more
than 10 times that amount in money in the
The term in brackets is the M2 multiplier. Notice form of M2. In contrast, the Ml multiplier
it is considerably larger than the Ml multiplier shown in part a of Figure 14.1 (page 475) is
discussed in the text, because time deposits and much smaller, and in fact is less than 3.
other components give rise to two additional
terms in the numerator of the M2 multiplier.
Furthermore, the ratio of money market mutual Sources for data: Board of Governors of the Federal
funds and time deposits to demand deposits Reserve System, Federal Reserve Bulletin, various issues;
tends to be much greater than 1. Since there Citibase electronic database; authors' calculations.
million in securities through open market operations. In this case, A MB =

$1,000,000, rr = .2, ed = .05, and cd = .25. Thus, we see the total change
in the money stock is
1 + .25
AM = X $1,000,000
.2 + .05 + .25
= 2.5 X $1,000,000 = $2,500,000.
Since the complete money multiplier is 2.5, the total stock of money in¬
creased by a multiple of the increase in the monetary base, or by $2,500,000.
In equations 14.14 and 14.18, we saw that $2 million of this increase was
in the form of deposits and $500,000 was in the form of currency holdings.
Notice that if you add up these changes in deposits and currency holdings,
you get $2,500,000—the total change in the money stock. This is a general
result, since it is always the case that
AM = AC + AD.
Summary What are the changes in deposits, currency holdings, and the money stock
Exercise 14.3 for an open market sale of $100,000? The required reserve ratio is 10 percent,
the desired excess reserve ratio is 5 percent, and the desired currency to
deposit ratio is 25 percent.
Answer: To answer this question, we first note that an open market sale
is a reduction in the monetary base of $100,000. Thus, we will be looking
at reductions in currency holdings, deposits, and the money stock. Next, we
will use the complete multiplier formulas derived earlier:
1 -
AD X AMB
rr + ed + cd
c
AC X AMB
rr + ed 4- cd
and
1 +
AM = X A MB
rr + ed + cd
Using the figures given in the problem, we find the complete deposit mul¬
tiplier is 2.5, the complete currency multiplier is .625, and the complete
money multiplier is 3.125. It follows that
AD = 2.5 X -$100,000 = -$250,000,

AC = .625 X -$100,000 = -$62,500,
and
AM = 3.125 X -$100,000 = -$312,500.
Thus, the open market operation causes deposits to decline by $250,000,

currency holdings to decline by $62,500, and the money stock to decrease
by $312,500. In fact, this example illustrates a general property of the
complete money multiplier: It is the sum of the complete deposit multiplier
and the complete currency multiplier (in this case, 2.5 + .625 = 3.125).
This observation provides a useful way for you (and us!) to double check
the arithmetic.
Determinants of the Money Supply
Based on our previous analysis, it is natural to define the money supply (Ms)
as
1 + cd
Ms X MB. (14.23)
rr + ed + cd
This relationship is called the money supply equation, because it indicates

how much money is created (produced) in the economy for a given monetary
base. In effect, the banking system and the public transform the monetary
base into specific amounts of currency holdings and deposits. It is customary
to refer to this transformation as the money supply process and the resulting
amount of money as the amount of money supplied. The term money stock
is usually reserved for the specific amount of money supplied, although the
terms money supply and money stock are sometimes used interchangeably.
Equation 14.23 indicates the major determinants of the money supply
are (1) the required reserve ratio (rr), (2) the currency to deposit ratio (cd),
(3) the desired excess reserve ratio (ed), and (4) the monetary base (MB). A
change in any of these determinants will change the amount of money—that
is, currency plus checkable deposits—available in the economy.
Part a of Figure 14.1 shows recent movements in the monetary base and
the Ml money multipier; part b shows actual movements in the money stock
as measured by Ml. Notice that Ml increased substantially from the middle
of 1991 to 1993, reflecting growth in the money stock of almost 13 percent.
Part a shows that this increase in the money stock was caused by two factors.
First, the Fed increased the monetary base by about 10.5 percent over this
The changes in the Figure 14.1

money stock shown in The Monetary Base, Ml Multiplier, and Ml, 1987-1993
part b are a result of
the changes in the
money multiplier, the r 3.0
monetary base, or
both in part a.
- 2.8
■g_
2.6 2
2.4
Source: Copyright © Federal Reserve Bank of St. Louis. Used with permission.
period, which, as we saw earlier, results in an increase in the money stock.

Second, notice the money multiplier also increased during this period, which
accounts for the additional increase in the money stock.
This shows that changes in the monetary base and the money multiplier
actually affect the amount of money in the economy. We will see next why
the money supply might change due to changes in the behavior of (1) the
Fed, (2) banks, or (3) the public.
Federal Reserve Determinants of the Money Supply

The Fed has two primary ways of influencing the money supply: (1) through
the monetary base (MB) and (2) through the required reserve ratio (rr).
During a period of economic expansion, the Fed sometimes takes actions
that reduce the money supply to counter inflationary pressures. In contrast,
the Fed often increases the money supply during economic downturns in the
hope of stimulating the economy. In Chapter 16, we will examine why
changes in the money supply affect the macroeconomy. Here we focus on
the tools the Fed uses to change the money supply.
The Fed can increase (decrease) the monetary

base by engaging in open market purchases (sales). An increase (decrease)
in the monetary base leads to an increase (decrease) in the money supply
that is a multiple of the increase (decrease) in the monetary base, thanks to
the complete money multiplier.
To illustrate this, let us calculate the total money supply for two different
levels of the monetary base. Suppose rr = .2, ed = .05, cd = .25, and MB
= $350 billion. In this case, the total money supply is
1 + .25
Af = X $350,000,000,000
.2 + .05 + .25
- 2.5 X $350,000,000,000
= $875,000,000,000.
If the monetary base were lower—say, $300 billion—but the other param¬
eters remained the same, the total money supply would be
1 + .25
Ms = X $300,000,000,000
.2 + .05 + .25
= 2.5 X $300,000,000,000
= $750,000,000,000.
This example illustrates that the Fed can decrease the money supply by
reducing the monetary base through open market sales.
The Required Reserve Ratio. The Fed can also change the money
supply by changing the required reserve ratio. By changing reserve require-
ments, the Fed alters the money multiplier, thereby changing the amount of
money generated from a given monetary base. An increase in the required
reserve ratio reduces the money multiplier and therefore leads to a reduction
in the money supply, whereas a decrease in required reserves has the opposite
effect.
To illustrate this, we calculate the total money supply for two different
levels of the required reserve ratio. Suppose MB = $350 billion, rr = .2,
ed = .05, and cd = .25. In this case, the total money supply is
1 + .25
Af = X $350,000,000,000
.2 + .05 + .25
= 2.5 X $350,000,000,000
= $875,000,000,000.
If the required reserve ratio were lower—say, rr = .1—but the other vari¬
ables remained the same, the total money supply would be
1 + .25
Ms — X $350,000,000,000
.1 + .05 + .25
- 3.125 X $350,000,000,000
= $1,093.75 billion.
This example illustrates that a reduction in the required reserve ratio leads
to an increase in the money supply. The reason is that it increases the
complete money multiplier, in this example from 2.5 to 3.125.
As a matter of practice, the Fed seldom uses this tool to influence the
money supply; reserve requirements tend to remain fixed for long periods
of time. Prior to the DIDMCA, the Fed lacked the power to set reserve
requirements at S&Ls and other thrift institutions. Consequently the Fed
could not affect the reserves of all depository institutions by changing the
reserve requirements. Although the Fed now has the authority to set reserves
at all depository institutions, it has little experience in using this tool to
influence the money supply. A small change in reserve requirements leads
to a large change in the money multiplier. Therefore, even a small mistake—
setting reserve requirements a few percentage points too high, for instance—
could drastically reduce the money supply, possibly creating a recession. For
these reasons, the Fed relies primarily on open market operations to change
the monetary base on a day-to-day basis.
Banking System Determinants of the Money Supply

The banking system helps determine the money supply by its choice of the
ratio of excess reserves to deposits, ed. Since higher excess reserves reduce
the amount of loans the banking system creates from a given monetary base,
increases in ed lead to reductions in the money supply, whereas decreases in
ed lead to increases.
Figure 14.2 plots the actual excess reserve ratio of the banking system
for the period 1972 to early 1993. Notice the ratio of excess reserves to
deposits was very small, averaging less than .1 percent (or .001) during the
1970s but increasing to between .15 and .20 percent from the mid-1980s
and into the 1990s. Thus, banks held very few excess reserves. In addition,
the ratio of excess reserves to deposits showed a lot of volatility, with large
upward and downward movements in only a few months.
What factors might cause a bank to raise or lower its desired excess
reserve ratio? The major factors are changes in the market interest rate on
loans, the risk of deposit withdrawals, and the interest rates on sources of
borrowed reserves. We explain the link between these factors and the desired
excess reserve ratio next.
Market Interest Rates on Loans. The market interest rates on loans

influence the desired ratio of excess reserves to deposits because these in¬
terest rates are the opportunity cost of holding excess reserves. Recall that
Sources: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, various
issues; Citibase electronic database.
excess reserves earn no interest. When a bank holds excess reserves, it

forgoes the interest it could have earned by making a loan from those
reserves. The higher the market interest rates on loans (or other securities a
bank can purchase), the higher the opportunity cost of holding excess re¬
serves. Thus, the desired ratio of excess reserves will decline with increases
in the market interest rates on loans and other securities. In fact, the increase
in the excess reserve ratio during the 1980s was due partly to the lower
interest rates that prevailed during that period relative to the 1970s, which
reduced the opportunity cost to banks of holding excess reserves.
Risk of Deposit Withdrawals. Banks hold excess reserves mostly

to help them deal with unexpected withdrawals. Having excess reserves
available allows a bank to avoid calling in loans, selling securities, or bor¬
rowing reserves from either other banks or the Federal Reserve when it faces
a withdrawal of deposits. For instance, an earthquake in California or a
hurricane in Florida will generally lead banks in those areas to hold additional
excess reserves in anticipation of people withdrawing funds to repair homes
and the like. The higher the risk of such withdrawals, the greater the excess
reserve ratio.
Interest Rate on Borrowed Reserves. As we mentioned earlier,

a bank facing a sudden need for reserves can avoid calling in loans, at least
for awhile, by borrowing reserves from other banks (in the federal funds
market) or from the Fed (through the discount window). Banks view the
ability to borrow reserves in times of need as a substitute for holding excess
reserves. Instead of holding excess reserves and giving up the market rate
of interest on loans or securities, a bank can choose to be nearly “loaned
up,” with the intention of meeting any sudden need for reserves by borrow¬
ing. When the interest rates paid on borrowed reserves rise, as is the case
when the federal funds rate or the discount rate increases, this option be¬
comes relatively less attractive, and the excess reserve ratio increases. This
tends to reduce the complete money multiplier and thus the money supply.
The Public's Determinants of the Money Supply

The public chooses its desired currency to deposit ratio, which also helps
determine the money supply. If people want to make more cash transactions,
the desired currency to deposit ratio will increase. This will reduce the
complete money multiplier and thus lead to a reduction in the money supply.
A decrease in cd has the opposite effect.
To see how the currency to deposit ratio affects the money supply,
suppose people initially hold 25 percent of their deposits as currency. As¬
suming rr = .2, ed = .05, cd = .25, and MB = $350 billion, the total
money supply is
1 + .25
Ms = X $350,000,000,000
.2 + .05 + .25
= 2.5 X $350,000,000,000
= $875,000,000,000.
If people increase their desired currency holdings to 50 percent of deposits,

but the other parameters remain the same, the money supply decreases:
S _
1 + .5
M X $350,000,000,000
.2 + .05 + .5
= 2 X $350,000,000,000
= $700,000,000,000.
This example illustrates that an increase in the currency to deposit ratio leads
to a reduction in the money supply, because it drains the banking system of
some of its reserves. In this example, the rise in cd reduces the multiplier
from 2.5 to 2.
Figure 14.3 graphs the currency to deposit ratio for the 1972-1993
period. This ratio had an upward trend from about .3 in the early 1970s to
around .4 in the early 1980s. A substantial decline occurred from about 1985
until 1987, followed by an even greater rise until about 1990. Since then
there has been a further decline, back to a level of just over .4. This means
the average depositor holds currency that is about 40 percent of his or her
checkable deposits. Notice the substantial rise in the currency to deposit
ratio from 1987 until 1990 translated into a decrease in the money multiplier
during that period, as we saw in part a of Figure 14.1.
What factors might lead to a change in the public’s desired currency to
deposit ratio? This ratio depends on a number of factors, including the
interest rate on checkable deposits, the fees on these deposits, income, the
probability of bank failure, and the extent of illegal activity. We will discuss
each of these factors in turn.
Interest Rates on Checkable Deposits. Moneyholders face the

choice of holding their money in the form of currency or checkable deposits.
One cost of holding money in the form of currency is the forgone interest
that might be earned on deposits. When the interest rate on checkable de¬
posits rises, the attractiveness of holding currency declines, which reduces
the currency to deposit ratio.
Fees on Checkable Deposits. Moneyholders looking to choose be¬

tween currency and checkable deposits will also be influenced by the fees
charged on checking accounts, both monthly maintenance charges and fees
assessed per check or ATM withdrawal. These fees are a cost of holding
money in the form of deposits, as we learned in Chapter 4. When these fees
Sources: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, various
increase, the attractiveness of holding money in the form of currency in¬

creases, which raises the currency to deposit ratio.
Income. The income of moneyholders also has some influence on the

currency to deposit ratio. In general, the currency to deposit ratio declines
with income. Those with higher incomes tend to be more sophisticated users
of the financial system and rely more on the financial system than on cur¬
rency. Those with lower incomes are more likely to use currency in their
transactions. Thus, the higher the income in an economy, the lower the
currency to deposit ratio and the larger the money supply.
The Probability of Bank Failure. When the probability of a bank

failure rises, moneyholders tend to abandon deposits in favor of currency,
even though deposit insurance in the United States makes bank failures and
failures of other depository institutions such as savings and loans much less
costly than during earlier times. In recent years, in fact, failures of depository
institutions have caused little or no discomfort to most depositors and have

frequently gone almost unnoticed. However, any widespread incidence of
bank failures affecting numerous institutions, called a bank panic, would
likely trigger a movement toward cash, increasing the currency to deposit
ratio and reducing the money supply. Indeed, this is precisely what happened
in the United States during the Great Depression.
Illegal Activity. Since cash can be used to make unrecorded transac¬

tions, it is the medium of choice for illegal activities such as drug dealing,
illegal gambling, and tax evasion. Economists and policymakers often speak
of the underground economy, meaning that part of the economy that is
unrecorded in official measurements of economic activity because it consists
of illegal activities and unrecorded cash transactions. Any increase in the
dollar amount of transactions that take place in the underground economy
will tend to increase the currency to deposit ratio and thus decrease the
money supply.
Summary Assuming other things remain constant, determine the impact of each of the
Exercise 14.4 following events on the money supply: (a) The Fed lowers the required
reserve ratio; (b) the interest rate paid on deposits rises; (c) the Fed raises
the discount rate.
Answer: (a) A reduction in the required reserve ratio increases the com¬
plete money multiplier and thus leads to an increase in the money supply,
(b) An increase in the deposit interest rate will lead to a lower currency to
deposit ratio. This increases the complete money multiplier and thus in¬
creases the money supply, (b) An increase in the discount rate leads banks
to increase their desired excess reserve ratios. This reduces the complete
money multiplier and therefore decreases the money supply.
The Money Supply Curve
Our analysis of the money supply in the previous section showed the amount
of money in the economy ultimately depends on the monetary base, the
excess reserve ratio, the required reserve ratio, and the currency to deposit
ratio. The monetary base and the required reserve ratio are determined by
the Fed, while banks and depositors determine the excess reserve and cur¬
rency to deposit ratios. We conclude our analysis of the money supply
process by graphically depicting the money supply curve, the amount of
money suppliers are willing and able to supply at various interest rates.
Because economists disagree about the impact of interest rates on the
determinants of the money supply, we will distinguish between two views.
One is that the determinants of the money supply in equation 14.23 are
exogenous, that is, determined by outside forces and do not depend on such
variables as interest rates. According to this view, there is an exogenous
money supply: Changes in interest rates do not alter the currency to deposit
ratio or the desired excess reserve ratios of banks. The second view is that
an endogenous money supply exists: The determinants of the money sup¬
ply—most notably the currency to deposit ratio and the desired excess re¬
serve ratio—are endogenous and depend on economic variables like interest
rates. We will now see how these two views affect the picture of the money
supply curve.
Exogenous Money Supply Curve

Suppose banks wish to hold a fixed fraction of deposits as excess reserves
and depositors hold a fixed fraction of deposits as cash. In this case, the
excess reserve and deposit ratios are exogenous, that is, independent of
interest rates and other economic variables. Since these ratios are constants,
the money supply curve is vertical as in Figure 14.4 and called an exogenous
money supply curve.
The term exogenous money supply curve refers to the situation where
the supply of money in the economy is determined by banks’ preferences
for excess reserves and depositors’ preferences for holding cash, and these
preferences are not affected by changes in economic variables like interest
rates. In this case, the complete money multiplier is constant. Consequently
the amount of money supplied along the money supply curve does not vary
with interest rates; it is vertical, as is the case for the curve labeled Ms0 in
Figure 14.4. Our analysis of the simple money multiplier, which assumed
depositors never hold cash (a constant currency to deposit ratio of zero) and
banks never hold excess reserves (a constant excess reserve ratio of zero),
is an example of a situation where the money supply curve is vertical.
How do changes in cd, ed, rr, and MB affect the position of an exogenous
money supply curve? Increases in cd, ed, and rr shift the money supply curve
to the left (from Ms0 to M\ in Figure 14.4), since a rise in any of these
variables leads to a lower complete money multiplier and thus a lower money
supply. A decrease in any of these variables has the opposite effect, shifting
the money supply curve from Ms0 to Ms2. The monetary base has a direct
effect on the money supply: An increase in MB shifts the money supply
curve to the right, whereas a decrease shifts it to the left.
Endogenous Money Supply Curve

As noted earlier, some economists believe the excess reserve and currency
to deposit ratios are not constant but vary systematically with economic
conditions. For instance, as interest rates rise, many banks will decrease their
excess reserves to be able to lend out additional funds at the higher rates.
Similarly, many depositors will wish to hold less currency and more interest-
bearing deposits to earn greater interest income. In these instances, the money
multiplier is not a constant but an increasing function of interest rates. This
gives rise to a money supply curve that is endogenous and upward sloping,
such as the curve labeled Mq in Figure 14.5.
Why is the money supply curve upward sloping in the case of an en¬
dogenous money supply? As interest rates rise, excess reserves fall, and the
amount of money in the economy increases due to the complete money
multiplier. (Recall that a decrease in the excess reserve ratio leads to a higher
complete money multiplier.) Similarly, higher interest rates lead to a lower
currency to deposit ratio, which works through the complete money multi¬
plier to further increase the money supply. Thus, when we graph the money
supply as a function of interest rates as in Figure 14.5, it is an upward sloping
curve. Higher interest rates lead to a greater quantity of money supplied
when the money supply curve is endogenous. We will say more about this
in the next chapter, where we look at how to determine the equilibrium stock
of money in the economy.
When the money supply is endogenous, it is not appropriate to consider
the effect of exogenous changes in cd or ed on the money supply, since they
are functions of the interest rate graphed on the vertical axis. In fact, it is
because of this functional relationship that the money supply curve slopes
upward in the first place: Higher interest rates lead to a lower cd and ed and
thus a higher money multiplier and a greater quantity of money supplied.
Changes in the required reserve ratio (rr) or the monetary base (.MB),
however, will shift the money supply curve, and in the same direction as in
the case of an exogenous money supply curve. For instance, in Figure 14.5
we see that an increase in the required reserve ratio shifts the money supply
curve to the left, resulting in a lower money stock at each interest rate. A
decrease in the monetary base also shifts the money supply curve to the left,
resulting in a lower stock of money at each interest rate.
Summary Suppose the Fed engages in an open market purchase (it purchases govern¬
Exercise 14.5 ment securities from a securities dealer). Graphically illustrate the impact of
this action on the money supply when (a) the money supply is exogenous
and (b) the money supply is endogenous.
Answer: Regardless of whether the money supply is endogenous or ex¬

ogenous, the open market purchase leads to an increase in the monetary base
and thus shifts the money supply curve to the right. Part a of the figure
illustrates the result for the case of an exogenous money supply. Part b of
the figure is relevant when the money supply is endogenous.
Interest Interest
Rate Rate
Ml M:
Money Stock (M)
(a) Exogenous Money Supply
Conclusion
In this chapter, we developed algebraic and graphical models of the money

supply. We discussed the determinants of the money supply, from those
controlled by the Federal Reserve System to those elements determined by
the behavior of banks or the public. In the next chapter, we look at the other
side of the money market, money demand, and examine how equilibrium
between the supply of and demand for money is determined.
KEY TERMS
required reserves high-powered money

excess reserves complete deposit multiplier
legal reserves complete currency multiplier
required reserve ratio complete money multiplier
multiple deposit expansion underground economy
simple deposit multiplier money supply curve
desired currency to deposit ratio exogenous money supply
desired excess reserve ratio endogenous money supply
monetary base
1. The required reserve ratio is 12 percent. lars are often accepted in exchange—that
The desired currency to deposit ratio and is, as money—in countries around the
the desired excess reserve ratio are both world. Suppose something occurs that
zero. If there is an open market purchase makes foreign nationals no longer willing
of $100,000, what will happen to the total to hold dollars so those dollars, once out¬
deposits of the banking system? To the side the U.S. banking system, now get de¬
total loans? To the total reserves? posited in U.S. banks. What is the effect
on the U.S. banking system?
.
2 Consider a situation in which the required
reserve ratio is 15 percent and each bank 8. Go back to 1913, when the Fed was just
individually decides to hold 5 percent of created. Could the Fed easily conduct
every deposit against emergency with¬ open market purchases? What about open
drawals, in addition to the legally man¬ market sales?
dated reserves. There is no holding of
cash. What happens to total banking sys¬ .
9 Suppose an eccentric millionaire buries $ 1
tem deposits if an open market purchase million in a time capsule in 1994. This
of $500,000 occurs? time capsule will be opened by her heir,
her great-granddaughter, in the year 2014.
3. The required reserve ratio is 10 percent. What is the effect on the banking system
The excess reserve ratio is zero, but in 1994? What will be the effect when her
the public desires to hold currency equal great-granddaughter opens the capsule in
to 5 percent of deposits. If an open mar¬ 2014?
ket purchase of $250,000 occurs, what
will happen to total banking system .
10 In the early 1980s, the Federal Reserve
deposits? System lowered the required reserve ratio
on deposits. What was the predicted effect
.
4 Obviously deposits are a crucial input on the banking system?
into the production of loans by banks.
What other resources is a bank likely to .
11 Recently the government warned of an in¬
require to be able to produce (or supply) crease in drug trafficking. Assuming drugs
loans? are purchased solely with cash, what is
the predicted effect on the banking
.
5 If the legal reserve ratio is zero and banks
system?
hold no excess reserves, what will happen
if the level of banking system reserves in¬ .
12 Historically an increase in the number of
creases by $100? Does the currency to de¬ bank failures has been linked with an in¬
posit ratio matter for your answer? crease in the currency to deposit ratio.
6. Explain how the Federal Reserve System However, the recent savings and loan cri¬
creates money by buying U.S. government sis, in which thousands of savings and
loans failed, and the more recent trouble
securities in the open market.
in the banking industry have not led to an
.
7 A substantial amount of U.S. dollars is in increase in the currency to deposit ratio.
circulation outside the United States. Dol¬ Why not?
.
13 In the early 1980s, interest rates such as tend to increase or decrease the money
the prime rate and even the rate on U.S. supply? Why?
Treasury bills reached unheard-of heights,
nearing or even exceeding 20 percent.
.
17 If loan rates and deposit rates both in¬
crease, what is the expected effect on the
What do you think happened to the excess
banking system?
reserve ratio? Why? Would this have
tended to increase or decrease the money .
18 Assume the currency to deposit ratio has
supply? increased. What is the effect on banking
system reserves, deposits, loans, and the
.
14 President Clinton has suggested that taxes
money supply? Can the Fed adjust the re¬
are going to increase. What do you predict
quired reserve ratio and/or the monetary
will be the impact on the currency to de¬
base to restore the money supply to its
posit ratio? Why? What impact will this
initial level? Can the Fed also restore de¬
prediction tend to have on the money sup¬
posits, loans, and reserves to their initial
ply? Can the Fed offset this effect? If so,
levels? Explain.
how?
.
15 The Fed receives interest income from its
.
19 Some states have a tax on wealth, in
which they charge a tax on bank deposits.
holdings of U.S. government bonds. When
What effect would this tax have on the
the Fed increases the monetary base, does
banking system if it were adopted nation¬
this action increase or decrease the Fed’s
wide?
income from its securities portfolio? Other
things equal, would you expect the Fed to .
20 If the Fed has decided to try to minimize
favor increases in the monetary base or currency holdings because it has become
decreases in the reserve ratio as a means too costly to maintain the stock of Federal
of increasing the money supply? Reserve notes, what actions might it take?
.
16 If you read in the paper that the discount
rate has increased, do you think this will
Garfinkel, M., and D. L. Thorton. “The Multiplier Ross, M. H., and R. E. Zelder. “The Discount Rate:
Approach to the Money Supply Process: A Precau¬ A Phantom Policy Tool?” Western Economic
tionary Note.” Federal Reserve Bank of St. Louis Journal, 7 (December 1969), 341-348.
Review, (July/August 1991), 47-64. Tiwari, K. N. “The Money Supply Process under De¬
Kopecky, K. J. “Required Reserve Ratios and Mone¬ regulation.” Financial Review, 21 (February
tary Control.” Journal of Economics and Business, 1986), 111-123.
33 (Spring/Summer 1981), 212-217. Trescott, P. B. “The Behavior of the Currency De¬
Meyer, P. A. “Money Multipliers and the Slopes of posit Ratio during the Great Depression: A Com¬
IS-LM.” Southern Economic Journal, 50 (July ment.” Journal of Money, Credit, and Banking, 16
1983), 226-229. (August 1984), 362-365.
CHAPTER
The Demand for Money

and Equilibrium in
the Money Market
n the previous two chapters, we examined the institutional arrange¬

ments within the U.S. banking system, particularly the role of the Federal
Reserve System and the money supply process. We now turn to a discussion
of the other side of the market: the demand for money. We identify factors
that determine how much money you and other actors in the economy—
including businesses—keep in your pockets, checking accounts, and cash
registers. Our analysis will enable you to predict the effect on interest rates
of Fed policies that increase the money supply.
We begin this chapter with a look at three views of money demand: the
classical view, the Keynesian view, and the modern quantity theory of
money. We will see that classical economists believed people hold money
primarily to make transactions (buy goods). This transactions view of money
demand led the classical economists to conclude that income is the primary
determinant of money demand. This view was later modified by John May¬
nard Keynes, who reasoned that people consider the costs of holding money
when they decide how much to hold. You may think the costs of holding
money are zero; after all, you pay no fees to carry currency in your pocket.
Other forms of money, however, do involve direct costs. Traveler’s checks
often contain a fee of 1 percent of their amount, and many checking accounts
charge fees. While the fees on checkable deposits vary, even “free” check¬
ing accounts usually require you to purchase the checks you will use. More
important, there are also indirect costs of holding money in the form of
currency, traveler’s checks, or checkable deposits. These costs arise because
you earn a lower interest rate on these holdings than you can earn by holding
wealth in some other form. Even checking accounts that pay interest do not
pay as much as you could earn by holding another asset such as a government
bond or even a U.S. savings bond. Thus, holding money has an opportunity
cost, since you lose out on potential interest earnings. This fact led Keynes
to reason that the higher the interest rate, the less money people will hold.
Milton Friedman further developed this view of money in what is known as
the modern quantity theory of money. This theory combines features of the
489
490 Chapter 15 The Demand for Money and Equilibrium in the Money Market
classical and Keynesian views of money and includes wealth, interest rates,
and a host of other factors as determinants of money demand.
After we examine the historical origins and current thought on money
demand, we look at how modern theories and the money supply apparatus
developed in Chapter 14 can be used to predict the effects of monetary
policy on interest rates. We conclude with a discussion of differences be¬
tween monetarist and Keynesian views of monetary policy and examine the
relationship between the analysis in this chapter and the loanable funds
approach we developed in Chapter 4.
The Classical View of Money Demand

The classical view of money demand considers income to be the primary
determinant of money demand: The higher your income, the greater the
amount of money you will hold. In this section, we look at two versions of
the classical view of money demand: the simple quantity theory and the
Cambridge theory. The simple quantity theory is based purely on the trans¬
actions motive for holding money, that is, the need for money to buy goods
and services. The Cambridge theory views money as useful for transactions
purposes and as a store of wealth. But as we will see, both theories imply
similar demand functions for money.
The Simple Quantity Theory

In the early 1900s, Irving Fisher developed the simple quantity theory of
money demand, which views transactions as the primary motive for holding
money.1 It has its roots in the equation of exchange. The equation of
exchange, introduced in Chapter 3, links aggregate spending in an economy
and the stock of money.
To illustrate the development of Fisher’s ideas about money demand,
we consider a simple economy in which total spending in a year, TS, is
$1,000. Furthermore, the stock of money, M, is $200. Why can individuals
in the economy buy $1,000 worth of goods and services if the money stock
is only $200? The answer is that on average, each dollar is used more than
once in a year. In fact, in this example each dollar must be used an average
of five times in a year for a money supply of $200 to support $1,000 of
spending.
The number of times an average unit of money changes hands in one
year is called velocity, or V. We can then write the equation of exchange as
M X V = TS.
1 Later in this chapter, we will trace the origins back to the 1600s.
This equation states that the quantity of money in an economy, M, multiplied

by the number of times the average unit of money changes hands in one
year, V, will equal the value of annual total spending, TS.
The equation of exchange provides a useful way to think about money
and the economy, but it is not a theory. In fact, the equation of exchange is
an identity. It must be true, because velocity is defined so as to make it true.
We do not really observe velocity. Instead we simply divide total spending
by the money supply and call the result velocity.
This doesn’t mean the equation of exchange is useless. For example, it
tells us that if the quantity of money increases, either total spending will
increase, velocity will fall, or both. In our example total spending is $1,000,
velocity is 5, and the quantity of money is $200. If the quantity of money
increases to $400 and velocity stays constant, total spending must increase
to $2,000. Why? Because the economy now has $400 of money, and if
everyone spends this money at a rate that causes it to change hands five
times in a year, spending will have to be $400 X 5 = $2,000.
What if the money supply increases to $400 but total spending stays at
$1,000? Then we know velocity has fallen to 2.5. With $400 of money and
spending at $1,000, money must change hands 2.5 times in one year, on
average.
The equation of exchange does not tell us whether an increase in the
quantity of money will increase total spending or reduce velocity. Likewise,
it is not clear whether an increase in income will raise the quantity of money
or increase velocity.
To develop a theory of money demand, Fisher reasoned that velocity
would be constant since it depends on slowly moving variables such as the
frequency with which people are paid and institutional characteristics of the
payments system. During Fisher’s time the financial system was just devel¬
oping, and it was probably reasonable to view velocity as determined pri¬
marily by the banking system of the day rather than varying in response to
changes in the quantity of money or total spending. At any rate, this line of
reasoning implies the equation of exchange can be used to write money
demand as
V’
Since velocity was thought to be constant, any increase in total spending

would lead to an increase in the demand for money. The idea was that with
constant velocity, an increase in spending required greater money holdings
to satisfy the equation of exchange. Thus, money demand was linked to the
desire to make transactions. This is why we say the simple quantity theory
is based on the transactions motive for holding money.
Fisher had a further insight into money demand, which he got from
considering how total spending could change. In our example, we know total
492 Chapter15 The Demand for Money and Equilibrium in the Money Market
spending is $1,000. What we don’t know is whether this amount consists of

the purchase of 1,000 goods costing $1 each, 2,000 goods costing $.50 each,
or 500 goods costing $2 each. Thus, Fisher separated total spending, TS,
into the quantity of goods purchased, Y, and the price of those goods, P, or
TS = P X Y.
Using this formula, we can write the equation of exchange in its most familiar
form as
M X V = P X Y.
Then the simple quantity theory of money demand becomes
p v y
Md = -. (15.1)
V
This version of the quantity theory of money demand stresses that people
will want to hold more money if either the prices of the goods they purchase
increase or the quantities they want to purchase increase. In either case, the
amount of money required to pay for the purchase of these goods will
increase.
Suppose people are purchasing 1,000 goods at a price of $1 each. From
our previous example velocity is 5, so money demand is $200, calculated as
$1,000/5. What happens if the price of the goods doubles to $2 each? If
people still want to purchase 1,000 goods, they will now make purchases in
the amount of $2,000 per year. With velocity constant at 5, this increases
money demand to $400 ($2,000/5).
Alternatively, consider what happens if the number of goods people
want to purchase rises to 2,000, but the price of each stays at $1. In this
case, the dollar amount of purchases is also $2,000, so with velocity at 5,
money demand is again $400. Thus, with velocity constant, increases in the
price level have the same effect on money demand as do increases in the
number of goods and services purchased. In fact, as our examples show,
when the price level doubles, so does money demand. When prices rise from
$1 to $2, money demand increases from $200 to $400. This is a general
property of the simple quantity theory of money demand. As long as the
physical amount of purchases stays constant, any increase in prices will be
matched by a proportional change in money demand. Thus, the simple
quantity theory of money demand is often called a theory of the demand for
real money balances. Real money balances are the nominal money stock
divided by the price level, M/P, and are a measure of the purchasing power
of money. For a given quantity of dollars, M, increases in the price level
mean less can be purchased. Increases in purchasing power come about from
either an increase in the money stock, M, or a reduction in the price level,
P, while decreases in purchasing power result from either a decrease in the
money stock or an increase in the price level. To write the simple quantity
theory of money demand in terms of the demand for real money balances,
we simply divide both sides of equation 15.1 by the price level, P, to obtain
Md Y
Since the simple quantity theory of money views velocity in equation

15.2 as constant, the demand for real money balances depends solely on real
income. If real income doubled, real money balances would double as well;
twice as much real money balances would be needed to make transactions.
Box 15.1 examines the extent to which this hypothesized relationship holds
for Mexico. Later we will look at data for the United States.
The Cambridge Theory

The Cambridge theory of money demand was developed by economists
in Cambridge, England, at roughly the same time Irving Fisher was pio¬
neering his views on the simple quantity theory of money. The Cambridge
theory says people have two motives for holding money: to make transac¬
tions and to store wealth. Like Fisher, the Cambridge economists viewed
the demand for real money balances as being proportional to real income:
(15.3)
where k is called the Cambridge constant. According to the Cambridge

view of money demand, a doubling of real income leads to a doubling of
real money balances. In particular, with a given price level twice the amount
of money would be desired for making transactions, and twice the amount
would be desired as a store of wealth. Even though the Cambridge econo¬
mists considered a second motive for holding money (storing wealth), they
still ended up with a demand for money that was proportional to income,
just as in the simple quantity theory.
In fact, the Cambridge view of money demand may be rewritten in a
form that is identical to the simple quantity theory. If we take the Cambridge
constant to be the reciprocal of velocity, k = MV, we can rewrite the
Cambridge demand function for real money balances in equation 15.3 as
McI
= k Y = — Y, (15.4)
P V
which is identical to the simple quantity theory specification presented in

equation 15.2.
Both the simple quantity and Cambridge theories view real money de¬
mand as depending primarily on real income. This view would be justified
if k (or, equivalently, velocity) were constant. While the Cambridge econo¬
mists usually treated k as a constant, they recognized that it might in fact
Money Demand in Mexico
The simple quantity theory of money says 1948 until about 1979. During the early 1980s,
people hold real money balances because they real GDP in Mexico remained relatively constant,
want to purchase real goods and services (real but real money balances declined substantially.
GDP). The higher real GDP is, the higher will be Thus, the simple quantity theory did a reason¬
the demand for real money balances. The accom¬ ably good job of explaining movements in real
panying figure shows real money balances and money balances in Mexico until the 1980s. After
real GDP for Mexico from 1948 to 1990. Notice 1980, factors other than real GDP led to reduc¬
the close relationship between increases in real tions in Mexican real money balances.
GDP and increases in real money balances from
Sources: International Monetary Fund, International Financial Statistics, various years; Citibase electronic database.
vary as individuals altered the amounts of money they chose to hold as a

store of wealth. This subtle difference ultimately led John Maynard Keynes,
also a Cambridge economist, to consider factors other than income that might
cause individuals to hold more or less money.
Keynes's View of Money Demand

The modern view of money demand owes much to the work of John Maynard
Keynes in the 1930s.2 While classical economists tended to emphasize the
use of money in making transactions, Keynes identified three motives for
holding money: the transactions motive, the precautionary motive, and the
speculative motive. As with the classical economists, Keynes’s transactions
motive suggested that people hold money because it is useful in making
purchases. Naturally the transactions motive for holding money would give
rise to money demand that is positively related to income: People with higher
incomes typically make more transactions and thus will hold more money.
In fact, like the classical economists, Keynes viewed money held for trans¬
action purposes as being proportional to income.
The precautionary motive for holding money is the desire to hold
money to meet unexpected expenditure requirements, such as for emergen¬
cies or the proverbial rainy day. This motive is a refinement of the store of
wealth function emphasized in the Cambridge theory. Indeed, Keynes also
viewed money held for precautionary purposes as being proportional to
income.
The most novel idea in Keynes’s theory of money demand was the
speculative motive, which implied money demand depends on interest rates.
Keynes recognized that many assets other than money, such as bonds, can
serve as a store of wealth. However, the return on a bond includes not only
the interest payments received during the bond’s term but also any capital
gain or loss resulting from a change in the bond’s price. As we learned in
Chapter 8, when interest rates fall, bond prices rise, resulting in capital gains
to bondholders. Keynes reasoned that when interest rates are above their
“normal” levels, people will hold less money and more bonds because they
expect bond prices to rise when interest rates return to lower, “normal”
levels. This reasoning led Keynes to conclude that money demand is in¬
versely related to the interest rate: The higher the interest rate, the less money
(and more bonds) people will want to hold, since they speculate that bond
prices will rise when interest rates return to normal. The focus on interest
rates is the feature of Keynes’s view of money demand that distinguishes it
from the classical view.
2 See John Maynard Keynes, The General Theory of Employment, Interest, and Money (New York:
Macmillan, 1936).
While the view that interest rates inevitably return to “normal” is no

longer generally accepted, we will see next that economists today do view
interest rates on nonmoney assets as an important determinant of money
demand for reasons somewhat different than those Keynes suggested.
The Modified Cambridge Theory

One major criticism of the classical view of money demand is that it assumes
velocity is constant. As Box 15.2 shows, velocity tends to fluctuate over
time, and these changes appear to be related to changes in interest rates.
Keynes’s view that money demand depends on interest rates provides a way
to explain fluctuations in money demand that changes in income cannot
account for.
To see this, let us modify the Cambridge theory to allow the Cambridge
constant to decrease when interest rates rise according to the functional
relation k = k(i). Under this view, k is no longer a constant but varies with
interest rates. In this case, we will call k the Cambridge k instead of the
Cambridge constant to emphasize that it is no longer assumed to be constant.
This modified Cambridge theory gives rise to a demand for real money
balances of the form
k(i) Y.
p
Since k(i) is assumed to be a decreasing function of interest rates, we see

that money demand is inversely related to interest rates: As interest rates
rise, individuals will hold a smaller fraction of their real incomes in real
money balances and more in bonds (k decreases), and the quantity demanded
of real balances will decline. This modification of the Cambridge model is
clearly consistent with Keynes’s view of money demand. It also suggests
that fluctuations in interest rates will give rise to fluctuations in the demand
for real money balances—something the classical view does not imply.
How is the modified Cambridge theory related to the evidence that
velocity is not constant? The Cambridge k may be viewed as the reciprocal
of velocity: k(i) = 1 /V(i). When interest rates rise, the Cambridge k falls,
which means velocity must rise. Intuitively, the higher the interest rate, the
more rapidly individuals will convert idle money balances into goods or
other assets, since the opportunity cost of holding money has increased. This
higher velocity, in turn, reduces the quantity demanded of real money bal¬
ances. We can more clearly see this link by using equation 15.4 and the
relation k(i) = 1 /V(i) to write the modified Cambridge specification of
money demand in terms of velocity:
1
k(i) Y Y.
P no
Velocity and Interest Rates
Is velocity constant as the classical economists 1975 to 1992. Notice velocity tended to rise
argued, or does it rise when interest rates rise as when interest rates rose and decline as interest
Keynes's theory of money demand suggested? rates fell. Velocity peaked in 1981, when inter¬
The accompanying figure shows the relationship est rates were near an all-time high. Since then
between the velocity of Ml and the interest rate interest rates have declined substantially, as has
on six-month commercial paper over the period velocity.
20
15
5
i—H
"
CD
10 Q>i—H
zo
QJ
r~t*
(V
Sources for data: U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, various
issues. Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, various issues; Citibase electronic
database.
The modified Cambridge theory generalizes the earlier views of money

demand. First, it allows real income and interest rates to affect the demand
for real money balances, just as Keynes argued in his writings. Second, it
provides an explanation for why velocity is not constant: Velocity rises when
interest rates rise and declines when interest rates fall, as Box 15.2 shows.
Despite this feature of the modified Cambridge theory, the theory has
been criticized because it is an ad hoc justification for interest rates to affect
money demand. For this reason, economists began searching for microeco¬
nomic justifications for Keynes’s view that interest rates affect money de¬
mand. The two most important such models are the inventory approach and
the portfolio approach, which we discuss next.
The Inventory Approach to Money Demand

The inventory approach to money demand, pioneered by William Baumol
and James Tobin in the 1950s,3 provides a microeconomic justification for
including interest rates as a determinant of money demand. Like the simple
quantity theory and the Cambridge theory, this approach also stresses the
importance of using money for transaction purposes. In fact, it says money
is held only because it is useful for making transactions. But the inventory
approach also considers the direct and indirect costs of holding money for
transaction purposes and explains how changes in those costs will affect
money demand.
The inventory approach is perhaps most relevant for flow-of-funds man¬
agers at multimillion-dollar corporations. But to keep dollar figures manage¬
able, we will use a very simple example to illustrate the basic ideas. Our
example focuses on the demand for money in the form of Ml, but the basic
ideas can be extended to the demand for broader monetary aggregates.
Consider Sue, a worker who is paid monthly. She is paid on the first
day of the month and receives $3,000. Her monthly expenditures are also
$3,000 and are spread evenly throughout the month, so she spends $100 per
day. Sue’s question is how much of her check to keep as money (in the form
of Ml) and how much to keep in less liquid assets. One answer might be to
keep the entire $3,000 in Ml (cash or checkable deposits), since she knows
she will spend it by the end of the month. This has some benefits, as we will
see, but it also has some costs. The costs are that Sue will lose whatever
interest she could have earned by using some of the funds to invest in another
asset (like a bond or a savings account) for part of the month. The benefits
are that she avoids the cost and bother of investing in some other asset and
then disinvesting a few weeks later, when she needs money for purchases.
3 See William J. Baumol, “The Transactions Approach to Demand for Cash: An Inventory Theo¬
retic Approach,” Quarterly Journal of Economics, 66 (November 1952), 545-556, and James
Tobin, “The Interest Elasticity of Transactions Demand for Cash,” Review of Economics and
Statistics, 38 (August 1956), 241-247.
To be more specific, suppose the interest rate on a savings account is 1

percent each month, or .01. This translates into about 12 percent annual
interest; this is a fairly high rate, but it will keep our calculations simple.
Assume money in the form of currency or checkable deposits pays no
interest. Also assume the cost of going to the bank or otherwise transferring
funds from the savings account into money is $15. This can be a fee the
bank charges or simply a dollar value of hassling with the bank. How should
Sue manage her income?
Suppose Sue keeps the entire $3,000 in the form of Ml. If she does so,
she pays only one fee—the $15 for cashing her check—and gives up the
potential to earn interest on her funds. In fact, with $3,000 at the start of the
month, her average money holdings are $1,500, so her lost interest earnings
are $1,500 X .01 = $15. The fact that her average money holdings are one-
half her initial money holdings is evident in part a of Figure 15.1. Here Sue’s
money holdings are plotted on the vertical axis, while the days of the month
are on the horizontal axis. She starts with $3,000, and spends $100 per day
every day of the month. Halfway through the month she has $1,500 left,
which is her average holdings over the month. By the end of the month she
has no money holdings, but she will receive another paycheck and can start
the process anew. Since her average money holdings are $1,500, she gives
up $15 each month in potential interest earnings by choosing to hold all her
income in the form of Ml. Of course, she also pays $15 in fees at the
beginning of the month, so her total cost of holding Ml is $30.
An alternative Sue might want to consider is to take her check for $3,000
and split it into two equal amounts. The first $1,500 is held as Ml, and the
second $1,500 is deposited in her savings account. After 15 days she will
have spent the first $1,500, so she returns to the bank and withdraws the
remaining $1,500 from her savings account, using these funds to pay for her
purchases in the remaining 15 days. What are the costs of this plan? First,
she has to visit the bank twice, and these costs are $15 X 2, or $30. Second,
she gives up interest on $1,500 for an entire month and interest on the other
$1,500 for half a month. We can calculate her average money holdings using
part b of Figure 15.1. When the month begins, she holds $1,500 in money,
which she spends at $100 a day until the money is exhausted on day 15.
Then she goes to the bank and withdraws the remaining $1,500, which she
also spends at $100 per day until the end of the month. Her average money
holdings in this case are $750, or $1,500/2 (or, for later reference, $3,000/
4), so the interest costs are $750 X .01 = $7.50. Her total cost of holding
Ml, including the $30 in bank fees, is $37.50.
Still another alternative is for Sue to split her $3,000 into three equal
amounts. She holds the first $1,000 as Ml and then uses the remaining
$2,000 to deposit in her savings account. After 10 days she will have spent
the first $1,000, so she returns to the bank and withdraws $1,000 from her
savings account. This lasts her until day 20, at which time she returns to the
bank and withdraws the remaining $1,000 from her savings account. For
Figure 15.1
Money Holdings and the Inventory Approach
In part a, Sue holds all $3,000 of income as money at the beginning of the month. Every day she spends
$100, until by the end of the month she has spent the entire $3,000 and the money is replenished by
another payday. In this case, average money holdings during the month are $1,500.
In part b, Sue holds only $1,500 of income in the form of money at the beginning of the month
and holds the remaining $1,500 as an interest-bearing asset. Every day she spends $100, until by the
end of the 15th day she has spent the entire $1,500. On this day the $1,500 nonmoney asset is con¬
verted into money, and Sue spends $100 per day until the 30th day, when this sum too is exhausted and
the funds are replenished by another payday. The average money holdings during the month are $750.
In part c, Sue holds only $1,000 of income as money at the beginning of the month and holds the
remaining $2,000 as an interest-bearing asset. Sue spends $100 every day, so by the end of the 10th day
the initial $1,000 is exhausted. On this day, Sue withdraws $1,000 of the interest-bearing asset to hold
as money. This lasts until the 20th day, at which time she again withdraws $1,000 to hold as money.
Average money holdings during the month are $500.
this plan, her costs include three trips to the bank, or $45 ($15 X 3). Also,
her lost interest earnings are calculated as follows. First, as part c of Figure
15.1 shows, her money holdings over any 10-day period start at $1,000 and
decline to zero, so her average holdings are $1,000/2, or $500. (For later
reference, note that we can calculate this as $3,000/6.) The interest cost of
holding this money is $500 X .01, or $5. Her costs also include $45 in bank
fees, for a total cost of $50.
By now a pattern has emerged. The more trips Sue makes to the bank,
the higher is her cost in terms of bank fees. But the more trips she makes,
the lower are her holdings of Ml and hence the lower is her opportunity
cost of lost interest. We can formalize our example as follows. If we let the
number of trips to the bank be N, the cost of bank fees is $15 X N. Interest
costs are the interest rate times her average money holdings. Average money
holdings can be determined by the formula $3,000/2/V, so the costs of lost
interest are .01 X $3,000/2/7. (You can verify the formula for average money
holdings by letting N equal 1, 2, or 3 and comparing the results with our
earlier calculations.) The total costs of one trip are $30, which is lower than
that for two trips ($37.50) or for three trips ($50). Thus, given the bank’s
fees, the interest rate, and her income, Sue minimizes the cost of holding
money in the form of Ml by making one trip to the bank each month.
Now let’s see what happens if Sue’s monthly income, and thus her
spending, increases. Suppose Sue gets a hefty raise and her income rises
from $3,000 to $12,000 per month. Thus, Sue now earns $12,000 per month
and spends $12,000 per month. We can use our formulas to find the interest
costs of holding money in the form of M1. The interest cost of making one
trip to the bank is .01 X $12,000/2 = $60, that of two trips is .01 X
$12,000/4 = $30, and that of three trips is .01 X $12,000/6 = $20. Adding
these interest costs to the bank fees per trip reveals that the total cost of one
trip to the bank is $60 + $15 = $75, that of two trips is $30 + $30 =
$60, and that of three trips is $20 + $45 = $65. (Four trips would cost $15
+ $60 = $75.) Obviously two trips minimizes Sue’s total costs of holding
Ml, and with two trips her average holdings of Ml rise to $12,000/4 =
$3,000.
What do we conclude about the effect of the increase in income on Sue’s
optimal money holdings? The increase in income (and in spending) led to
an increase in her average holdings of M1. This is in accord with the basic
idea of the simple quantity theory of money demand, and more generally,
the transactions motive for holding money: Increases in spending will require
larger amounts of money. However, the view of Keynes and the classical
economists was that an increase in spending leads to a proportional increase
in money demand. Thus, a doubling of spending would lead to a doubling
of money demand. This is not the case in the inventory approach. In our
example, for instance, we had a fourfold increase in income and spending,
from $3,000 to $12,000. In this inventory approach, average money holdings
increased only from $1,500 to $3,000, or by a factor of 2. This is but an
example of a general feature of the inventory approach: Money demand

increases only with the square root of the increase in income. Thus, if income
doubles, money demand increases by V2, or by a factor of about 1.4. If
income increases by a factor of 4, money demand increases by a factor of
V4, or 2.
How does a rise in interest rates affect money demand under the inven¬
tory approach? It increases the opportunity cost of holding money and there¬
fore leads to lower real money balances. When the interest rate increases,
the cost of holding money rises, and we can calculate this cost for each
number of trips to the bank from our previous formula for the interest costs.
That formula was .01 X $3,000/2/V, and we merely replace the interest rate
with the higher rate of .04. Thus, interest costs rise to .04 X $3,000/2/V.
With one trip to the bank, interest costs are $60. Two trips lower interest
costs to $30, and three trips lower them still further, to $20. With the higher
interest rate, total costs of one trip to the bank are $60 + $15 = $75, while
two trips cost $30 + $30 = $60 and three trips cost $20 + $45 = $65.
Due to the higher interest rate, it pays Sue to make two trips to the bank
each month instead of one to minimize her total cost of holding money.
What does the rise in the interest rate imply about Sue’s holdings of
Ml? Average money holdings with two trips to the bank fall from $1,500
to $3,000/4 = $750. Thus, the increase in the interest rate caused Sue to
hold less Ml and to hold more funds in her savings account. This is a general
implication of the inventory approach: Increases in the interest rate result in
a reduction in money demand.
The inventory approach we outlined here thus provides a microeconomic
justification for Keynes’s view of money demand. First, money demand is
inversely related to the interest rate. Second, when income increases, real
money demand rises as well. But in contrast to the classical view (and even
the view of Keynes himself), when income rises by some percentage—say,
100 percent—money demand rises by less than 100 percent; as income
increases, money holdings rise, but not as quickly. This finding is in part
due to the assumption that the fee for converting the nonmoney asset into
money does not change with income, nor does it change with the size of the
transfer from the nonmoney asset into money. Changing either assumption
would alter somewhat the prediction that money demand rises with the
square root of the rise in income, but it would not alter the main qualitative
result that money demand both increases with rises in income and decreases
with rises in the interest rate.
The Portfolio Approach to Money Demand

The portfolio approach to money demand (also called the asset approach)
was developed in the 1950s by James Tobin4 and provides an alternative
4 See James Tobin, “Liquidity Preference as Behavior Towards Risk,” Review of Economics
Studies, 25 (February 1958), 65-86.
iCEYNES'S View of Money Demand 503
justification for Keynes’s view that interest rates affect money demand. The
portfolio approach analyzes the decision by moneyholders to allocate their
wealth between other assets and money. Other assets like bonds pay interest
but are subject to the risk of price fluctuations and/or default. In contrast,
money bears little or no interest but is not subject to these types of risk.
The Cambridge economists and Keynes argued that money is also held
because it is a store of wealth, but they did not formally analyze how an
individual would go about determining how much money to hold. Moreover,
they did not consider the effects of risk on the desire to hold bonds or money.
Tobin formally modeled the decision to hold money in an environment of
risk and found that wealth and interest rates are key determinants of money
demand.
The portfolio approach assumes the only reason money is held is that it
is a safe asset. Stocks or bonds may go up or down in value, and investors
will experience capital gains or losses when this happens, but this does not
occur when they hold money. The portfolio approach stresses that unlike
risky assets like bonds, money is not subject to such risk.
How does the presence of an asset that is not subject to these types of
risk, like money, affect how individuals allocate their wealth between money
and other assets? Most people like the return they could get from holding
stocks or bonds but dislike risk. As we learned in Chapter 9, such people
are called risk averse. They may choose to invest a portion of their portfolios
in a safe asset like money. The key idea is that the proportion of money and
bonds in a person’s portfolio determines both the average return and the risk
of that portfolio. The more money, the lower the average return, but also the
lower the risk. The more bonds or stocks, the greater the average return, but
also the greater the risk. Individuals with differing degrees of risk aversion
will choose different mixes of money and risky assets to hold in their
portfolios. Those who are very risk averse may choose to hold only money,
while those who are risk neutral may prefer to hold only stocks.
Tobin used this reasoning to examine what happens to money holdings
when the average rates of return on bonds and stocks changes. For example,
if the average returns on bonds and stocks increases, what will happen to
the proportion of money holdings in a portfolio? In this case, two effects
may counteract. One effect is the substitution effect, in which the higher
return on bonds and stocks means that now a higher reward—that is, a higher
return—is available to investors for accepting any given level of risk. This
higher return means the rewards for accepting risk have increased, and
investors are induced to hold more stocks and bonds. Thus, the substitution
effect unambiguously predicts that an increase in returns on stocks and bonds
will tend to reduce money holdings.
The second effect is the income effect (sometimes called a wealth effect
in this context), in which the higher returns on stocks and bonds leave a
person with more income (or more wealth). When wealth increases, people
may decide to accept more or less risk, depending on their preferences.
Intuition suggests what is regarded as the usual case: The increase in wealth
will lead individuals to accept more risk. But this need not be the case. If
higher wealth leads an individual to accept more risk, the wealth effect
occurs in the same direction as the substitution effect, indicating a higher
return on stocks and bonds will lead to reduced money holdings. However,
if higher wealth leads an individual to desire less risk, the wealth effect
occurs in the direction opposite that of the substitution effect, and the net
effect is ambiguous.
The portfolio approach thus implies that money demand depends on
interest rates and that money is an important component of a portfolio only
because it is a safe asset, not because it is useful as a medium of exchange.
Critics naturally argue that many assets dominate the return to holding
money, especially assets that dominate the return on money in the form of
currency or demand deposits (Ml). For example, savings deposits at insured
banks are as risk free as checking accounts and currency, but they earn
higher interest. The logical conclusion, based on the portfolio approach, is
that individuals would never choose to hold assets in the form of M1, because
they could earn a higher interest rate on other risk-free assets like those
included in M2. Doing so would increase the overall return on the portfolio,
with no increase in risk.
The punch is that Tobin’s portfolio approach does not explain why
people hold currency and demand deposits that are dominated in return by
other risk-free assets. The only explanation for holding these forms of money
lies in the transactions motive, which is absent in the portfolio approach to
money demand. The portfolio approach can, however, explain the demand
for monetary assets like savings deposits or Treasury bills, which are in¬
cluded in broader measures of the money stock like M2 or L. For these
reasons, the portfolio approach provides an incomplete explanation of why
people hold money and does not explain why the demand for narrow mea¬
sures of the money stock (like Ml) vary with interest rates.5 Box 15.3
discusses an extension of the portfolio approach.
Summary In our example of the inventory approach, we started with income of $3,000,
Exercise 15.1 an interest rate of .01, and fees of $15 per transaction. We asked what
happens if income increases to $12,000 while fees and interest rates stay
constant. However, in one interpretation the fees on transactions are the cost
to the individual of the time and effort expended in going to the bank and
switching funds from savings into money in the form of currency or check¬
able deposits. This opportunity cost of time may be proportional to income.
Thus, if the fee is $15 when income is $3,000, it should be $60 when income
5 Recently economists have developed cash-in-advance models that extend Tobin’s model along
these lines. These models are based on the notion that some purchases (e.g., hamburgers from
McDonald’s or Wendy’s) require cash payment, thus providing a reason people might hold forms
of Ml even though they are dominated in return by assets contained in M2.
Currency Substitution
Currency substitution is the idea that domestic empirical evidence suggests that countries expe¬
residents consider domestic and foreign curren¬ riencing high rates of inflation—high rates of
cies to be relatively close substitutes. Because of depreciation in the values of their currencies—
this, demand for the domestic money depends often see their citizens begin holding and using
not only on domestic spending and opportunity a foreign currency for portfolio reasons and
cost variables but also on the rate of return on even for transactions. Thus, several South Amer¬
the foreign currency. The idea of currency sub¬ ican countries experiencing high inflation in the
stitution is a bold extension of the portfolio view 1980s saw an increased use of the U.S. dollar as
of money demand. In this case, the demand for citizens substituted away from the domestic
the domestic currency depends on factors exter¬ currency.
nal to the domestic economy, including influ¬ Formal studies of currency substitution have
ences from foreign policy decisions. yielded mixed results. One study by Bana and
The currency substitution hypothesis has Handa reports finding currency substitution in
been extensively investigated. This is due partly Canada under the flexible exchange rate re¬
to its policy implications. In a world of flexible gime, in which Canadian money demand re¬
exchange rates, the domestic central bank is sponded to the rate of return on the U.S. dollar.
supposed to be able to pursue monetary policy Bergstrand and Bundt also report evidence of
independently of the policy pursued by foreign currency substitution between a number of cur¬
central banks. The idea is that differential mone¬ rencies and the U.S. dollar. Their study included
tary policies will cause exchange rate move¬ the currencies of Canada, Italy, Switzerland, the
ments but will not affect real output or employ¬ United Kingdom, and West Germany and em¬
ment. However, the currency substitution phasizes that evidence of currency substitution is
hypothesis restores the ability of foreign policy most likely to be found when looking at rela¬
actions to affect the domestic economy. Under tionships over long time spans. Over short
this hypothesis, changes in foreign monetary spans, the evidence is less supportive of the cur¬
policy that alter the relative rate of return on the rency substitution hypothesis.
foreign currency will lead to changes in demand
for the domestic currency and hence changes in
Sources: Ismail Mahomed Bana and Jagdish Handa,
the domestic interest rate or the level of real
"Currency Substitution: A Multicountry Study for Can¬
money balances. ada," International Economic Journal, 1 (Autumn 1987),
Does currency substitution really exist? It 71-85; Dallas S. Batten and R. W. Hater, "Money, In¬
does not seem reasonable to presume the aver¬ come, and Currency Substitution: Evidence from Three
Countries," Federal Reserve Bank of St. Louis Review, 67
age U.S. citizen holds much foreign currency or
(May 1985), 27-35; Jeffrey H. Bergstrand and Thomas P.
responds to changes in the rate of return on for¬
Bundt, "Currency Substitution and Monetary Autonomy:
eign currency. At the same time, other countries The Foreign Demand for U.S. Demand Deposits," Journal
may have a different experience. Indeed, casual of International Money and Finance, 9 (1990), 325-334.
is $12,000. (a) If fees and income rise by the same factor of 4 (to $60 and
$12,000 respectively) what happens to average money holdings and to the
number of trips to the bank under the inventory approach? (b) Is your finding
in part a consistent with what the modified Cambridge theory would predict?
Answer: (a) The increase in income raises interest costs to .01 X $12,000/
2N. One trip to the bank costs $60 in lost interest, two trips cost $30, three
trips cost $20, and four trips cost $15. Fees for trips to the bank also rise to
$60 X N, or $60, for one trip, $120 for two trips, $180 for three trips, and
so on. The total cost of one trip is $60 + $60 = $120, that of two trips is
$30 + $120 = $150, and that of three trips is $20 + $180 = $200. Clearly
one trip to the bank is the best choice. Average money holdings increase to
$12,000/2 = $6,000. (b) Notice that income has increased by a factor of 4,
and so have average money holdings, from $1,500 to $6,000. Thus, in this
case the inventory model predicts that money demand will increase in the
same proportion as income. That is, with fees proportional to income, money
demand will be proportional to income just as in the modified Cambridge
theory (and the classical theories, for that matter).
The Modern Quantity Theory of Money Demand
Our discussion of the classical and Keynesian views of money demand has
demonstrated some of the properties we would expect money demand to
satisfy. In particular, money demand should increase when income rises,
decline when interest rates increase, and increase when the risk of holding
other assets increases. The modern quantity theory of money demand
builds on the simple quantity theory by emphasizing the unique property of
money as the medium of exchange. However, the modern quantity theory
combines this emphasis on transactions motives for holding money with a
consideration of the other determinants of money demand that are present
in Keynes’s view and in Tobin’s portfolio approach to money demand. Thus,
the modern quantity theory incorporates the insights gained from these al¬
ternative approaches to money demand while maintaining the concept of
money as being uniquely valuable as the medium of exchange.
The modern quantity theory was pioneered by Milton Friedman in the
mid-1950s.6 It describes the determinants of money demand, the factors that
indicate how much money households and businesses will demand. Friedman
viewed money demand as being similar to the demand for any durable good.
For example, a car is valued because it provides a stream of transportation
services that lasts many years. A house is valued because it provides a stream
6 Milton Friedman. “The Quantity Theory of Money: A Restatement,” in M. Friedman, ed., Studies
in the Quantity Theory of Money (Chicago: University of Chicago Press, 1956).
The Modern Quantity Theory of Money Demand 507
of housing services for many years. Money is valued because it provides a

stream of purchasing power services; that is, it can be used to purchase goods
and services.
Friedman emphasized that both individuals and businesses demand
money. Individuals hold money to buy goods, but it is just one of a number
of assets in their portfolios. Businesses hold money because it serves as a
factor of production, making it easier for them to pay for inputs needed in
production. To buy inputs, it is easier for firms to send a check or electron¬
ically wire funds than to engage in nonmonetary exchanges like barter. Firms
also need money to facilitate sales. Imagine how few Big Macs your local
McDonald’s would sell if it didn’t have cash on hand to make change!
Friedman organized the factors that influence individual and business
demand for real money balances into three broad categories: (1) individual
wealth and the scale of business activities, (2) factors that influence the
opportunity cost of holding money, and (3) tastes and preferences.
Individual Wealth and the Seale of Business Activity

In the tradition of the quantity theory, Friedman viewed the primary motive
for holding money to be its use in facilitating transactions. However, Fried¬
man broadened the classical economists’ view of variables that are related
to this transactions demand for money. He reasoned that the real wealth of
individuals determines the size of their portfolios, which are allocated among
money and other assets. It also determines how many transactions can be
made. In fact, wealth plays the same role here as income does in the classical
theory of money demand. But the role of wealth is broader in that it allows
individuals to spend more than their current incomes on goods. The reason
is that wealth is a stock—the accumulation of assets over time. You can
spend more than your current income if you draw on your wealth (by selling
your car or using past savings). Like increases in income in the classical
view, increases in wealth in the modern quantity theory lead to increases in
purchasing power and spending and thus to a rise in the demand for real
money balances.
Friedman also discussed the importance of the liquidity of wealth, that
is, the ease with which wealth can be transformed into money. The easier it
is to convert wealth held as stocks, bonds, houses, or cars into money, the
less attractive it will be to hold money, and vice versa. Friedman acknowl¬
edged that much of an individual’s wealth is in the form of human capital,
the skills and training an individual acquires that makes her or him a pro¬
ductive employee or entrepreneur. Such wealth is not liquid. It can be rented
at a wage rate to an employer, but it cannot be sold outright because human
capital cannot be separated from the individual, and slavery is illegal. Be¬
cause human capital is illiquid, it may have less of an effect on money
demand than do other, more liquid forms of wealth. If an individual decides
to switch between bonds and money, this is relatively easy to do, but it is
much less easy to switch between the person’s human capital and money.
Friedman further broadened the classical view by recognizing the utility

of real money balances for business operations. Unlike the individual, the
business firm is not constrained by a wealth variable. However, businesses’
demand for real money balances depends on a variable that measures the
scale of the business, such as the total value of the firm, the firm’s total
revenue, or some similar measure that indicates the firm’s size. Increases in
the size of the firm lead to increased numbers of transactions and hence to
increased real money balances.
Factors That Influence the

Opportunity Cost of Holding Money
While Friedman viewed real wealth as the primary determinant of money
demand, he admitted that other factors also influence money demand. He
noted, for instance, that individuals and businesses give up the interest they
could earn on other assets when they hold money. The interest forgone is
the opportunity cost of holding money. Increases in the interest rates on
these alternative assets will reduce the amount of money people and busi¬
nesses hold as they substitute away from money and toward these other
assets.
Like Tobin, Friedman also realized that when the risk of holding other
assets such as stocks and bonds increases, people will generally want to
concentrate more of their portfolios in safe assets such as money. Thus,
Friedman reasoned that in times of widespread uncertainty and volatility in
financial markets, people will increase their holdings of money.
Furthermore, Friedman pointed out that inflationary expectations might
influence the demand for money. Inflation is an ongoing increase in the price
level and as such represents an ongoing increase in the prices of goods.
When the prices of goods increase at a substantially higher rate than does
interest on money holdings, people will tend to hold goods instead of money
to earn a higher return. In periods of hyperinflation, such as existed in
Germany in the 1920s, when the inflation rate exceeded 1 million percent,
people may actually resort to barter rather than hold money. Carrying around
loaves of bread to trade for steak is certainly an inconvenience. But during
periods of hyperinflation, it is much more costly to leave your house to buy
a $100 steak, only to find that it costs $200 by the time you arrive at the
butcher. If you hold bread instead of money, the price of bread will grow at
the rate of inflation while you are on your way to the butcher. For these
reasons, Friedman believed a rise in the expected inflation rate increases the
expected opportunity cost of holding money and results in a reduction in the
amount of real money balances people will hold.
Tastes and Preferences

Finally, Friedman included a catch-all category called tastes that influences
the demand for real money balances. For instance, if people’s tastes or
Money Demand: An Historical Note and Current Thought 509
preferences for risk change, this will change how they split their portfolios
of assets among safe assets like money and risky assets like stocks or bonds.
If people become more willing to accept risk, they might hold more bonds
and less money.
Money Demand: An Historical Note and Current Thought
We conclude our survey of money demand by noting the considerable over¬

lap in the theories of Fisher, Keynes, and Friedman, and by pointing out that
their views were shaped by the work of economists over several hundred
years. We also present the main model of money demand that is widely
accepted and used by economists today to analyze the effects of monetary
policy.
An Historical Note1
While Fisher, Keynes, and Friedman were the first economists to formalize
money demand in the modern era of checkable deposits, we need to under¬
stand two points to put their views in context. First, their theories are in
many respects only refinements of the verbal arguments of economists in
the seventeenth, eighteenth, and nineteenth centuries, who discussed money
demand at a time when money consisted of gold and silver coins. The first
discussion of velocity and its determinants appeared in Sir William Petty’s
writings in the 1660s. He identified income, among other things, as a key
determinant of velocity. In the 1690s, John Locke made the link between
velocity and money demand more explicit and introduced the interest rate
as an important determinant since it represents the opportunity cost of hold¬
ing money. Richard Cantillon extended the work of Petty and Locke in 1775
by introducing expectations into velocity, and Henry Thorton, J. B. Say, and
Nassau Senior incorporated inflationary expectations in the early 1800s.
Thus, by the 1800s most of the variables that might influence the demand
for money were already known. In a sense, all Fisher, Keynes, and Friedman
did was popularize and refine the ideas of their predecessors.
The second point is that our earlier presentations of Fisher’s, Keynes’s,
and Friedman’s views on money demand are very simplified in that we
merely highlighted what each viewed as the primary determinants of money
demand. Fisher tended to emphasize the impact of real income on money
demand, Keynes stressed the interest rate, and Friedman emphasized real
wealth (in later writings, Friedman referred to real wealth as permanent
income). A closer look at their writings, however, reveals they all recognized
that other variables could influence money demand. For instance, Fisher’s
7 This section is based on the excellent survey by Thomas M. Humphrey, “The Origins of Velocity
Functions,” Economic Quarterly, 79 (Fall 1993), 1-17.
specification of velocity actually included most of the variables contained in

Friedman’s modern quantity theory, including interest rates, expected infla¬
tion, real income, and tastes. Indeed, there is considerable overlap among
each of these views of money demand; the primary differences are in their
emphasis.
Current Thought
After centuries of reflection, economists today recognize that many variables
might influence money demand. However, the evidence suggests that two
main variables influence the demand for real money balances, be it in the
form of Ml or M2 (see Box 15.4). The first is a scale variable, such as real
wealth or income. This affects money demand because greater real wealth
or income leads to more purchases, which in turn require additional real
money balances. The second variable reflects the opportunity cost of holding
real money balances, such as the interest rate. Thus, the conventional demand
function for real money balances used today can be summarized by
(.M/P)d = Ld(Y, 0,
where Y is real income and i is the interest rate.
Notice that the demand for money can be expressed in nominal or in
real terms. Money is demanded in part to purchase goods and services, so
increases in the price level raise the amount of money required to purchase
a given amount of goods. If the price level increases, money demand rises
proportionally. But if we write money demand in real terms as above, then,
if everything else remains the same, an increase in the price level, P, leads
to a proportional increase in the quantity of money demanded, M, so that
real money demand—the ratio of M to P—stays the same.
Figure 15.2 illustrates real money demand. On the vertical axis is the
interest rate on nonmonetary assets (like bonds), while on the horizontal axis
we graph the real quantity of money demanded. For reasons we explained
throughout the chapter, a rise in the interest rate increases the opportunity
cost of holding money and therefore leads to a reduction in the quantity of
real money balances demanded. This gives the demand for real money
balances its downward slope. Notice that a rise in real income increases the
demand for real money balances, as represented by a rightward shift in the
demand curve. The reason is that greater real income generally implies more
purchases, thus requiring more real money balances at each interest rate.
Equilibrium in the Money Market
Our analysis of money demand in this chapter, along with the model of
money supply we developed in Chapter 14, now allows us to determine
equilibrium between the demand for and supply of money. Recall there are
Equilibrium in the Money Market 511
Money Demand in the United States
The theories of money demand examined in the plots real money balances in the United States
text suggest that many variables might affect (defined as the money aggregate M2 divided by
money demand. Empirical evidence suggests, the price level) along with real GDP (part a). Part
however, that real income (GDP) and interest b shows the relationship between real money
rates are the primary factors influencing the de¬ balances and the six-month commercial paper
mand for real money balances. Since we ex¬ interest rate.
plored Ml in Box 15.2, we now look at how In part a, we see that the hypothesized rela¬
well these theories stack up when we look at a tionship between real M2 demand and real in¬
broader measure of the money supply, that come holds fairly closely: Decreases in real M2
measured by M2. The accompanying figure Continued on p. 512
(a) U.S. Recessions and the Behavior of

Real GDP and Real M2, 1972-1992
26
25
24
23
PO
22 CD
QJ
21
NJ
20
19
18
17
(b) Real M2 Money Balances and Six-Month

Commercial Paper Rate, 1972-1992
30 r 30
0 I I I I I i—r n—|~I—i—i—r 0
'72 77 ‘82 87 92
Continued from p. 511 many assets in M2 (like savings and time depos¬
accompany periods of slow growth or declines its) pay interest, the effect of changes in interest
in real GDP. The shaded regions correspond rates on M2 was less pronounced than that we
with periods of recession. Notice that with the saw when we looked at Ml in Box 15.2. In ef¬
exception of the recession in the early 1980s, fect, higher interest rates affected the opportu¬
the decline in real GDP during those recessions nity cost of holding Ml more than it did that of
was accompanied by a fall in real money bal¬ holding M2, since the rates paid on M2 bal¬
ances. ances tended to rise when rates paid on com¬
Part b shows the extent to which changes mercial paper and bonds increased.
in interest rates affect real M2 balances. The
graph indicates that increases in the interest rate
Sources: Board of Governors of the Federal Reserve Sys¬
such as that in the mid-1970s or the late 1970s
tem, Federal Reserve Bulletin, various issues; U.S. Depart¬
and early 1980s were accompanied by reduc¬ ment of Commerce, Bureau of Economic Analysis, Survey
tions in real M2, just as we would predict from of Current Business, various issues; Citibase electronic da¬
our theories of money demand. However, since tabase.
two views on money supply. The first is that the money supply is exogenous;
the quantity of money supplied does not change with changes in the interest
rate. The second is that the money supply is endogenous; the quantity of
money supplied fluctuates with changes in the interest rate.
Equilibrium with an Exogenous Money Supply

In Chapter 14, we saw that the money supply equals the money multiplier
times the monetary base, or
1 + cd
Ms = MB,
rr + ed + cd
where cd is the desired currency to deposit ratio, rr is the required reserve

ratio, and ed is the desired excess reserve ratio. If the money multiplier is
exogenous, and in particular does not respond to changes in the interest rate,
the money supply does not vary with changes in the interest rate.
The real money supply curve is the exogenous money supply, Ms, di¬
vided by the price level, P, or MSIP. In this case, the real money supply is
perfectly inelastic; the money supply curve is vertical as shown in Figure
15.3. In this graph, the vertical axis measures the interest rate on non¬
monetary assets like bonds, while the quantity of real money balances, M/P,
is measured on the horizontal axis. The money demand curve is downward
sloping as it is in Figure 15.2. The intersection of the real money supply and
demand curves at point A determines the equilibrium interest rate, /0. With
an exogenous money supply, the equilibrium quantity of real money balances
is determined solely by the position of the real money supply curve.
Notice that an interest rate below i0 would result in a shortage of real

money balances, since individuals and businesses would want to hold more
real money balances than the amount supplied at that interest rate. This
would put pressure on the interest rate to rise. As the interest rate increased,
the quantity demanded of real money balances would fall, because individ¬
uals and firms would find it more attractive to hold less money and more
assets earning the higher interest rate, like bonds. Ultimately the interest rate
would rise until this imbalance was corrected. Similarly, if the interest rate
were above /0, a surplus of real money balances would occur, putting pressure
on the interest rate to fall. Only at point A would there be neither a shortage
nor a surplus of real money balances.
jfi mm.
The money supply is ,
■■■ ■ ■ .
Figure 15.3
exogenous in this Equilibrium Real Money Balances with an Exogenous Money Supply
graph; that is, it is not .
affected by changes in
the interest rate. Thus,
the equilibrium quan¬
tity of real money bal¬
ances is determined
solely by the position
of the money supply
curve. The equilibrium
interest rate is deter¬
mined by the intersec¬
tion of real money de¬
mand and real money
supply at point A
The real power of our money demand and supply framework is that it
provides an easy way to show how monetary policy affects interest rates.
As we learned in the previous chapter, the Fed can change the money supply
by changing the monetary base (through open market operations) or by
taking other actions that affect the money multiplier (like changing reserve
requirements or the discount rate). Increases in the monetary base or the
money multiplier will shift the real money supply to the right, as will
decreases in the price level. Decreases in the monetary base or multiplier,
or increases in the price level, will reduce the real money supply, shifting it
to the left. Table 15.1 summarizes these variables.
Figure 15.4 illustrates how changes in the money supply affect interest
rates. Starting at the initial equilibrium at point A, suppose the Fed increases
the money supply through an open market purchase. This shifts the real
money supply curve to the right as banks ultimately convert the increase in
reserves created by the open market purchase into loans that wind up as new
deposits at banks. The result is a new equilibrium at point B, where the
equilibrium interest rate falls to /, as a result of the Fed’s easing of credit.
Notice that the quantity of real money balances rises to M\/P.
If the Fed tightens credit by reducing the money supply, the result is
just the opposite. The equilibrium changes from point A to point C in Figure
Table 15.1
Determinants of the Real Supply of Money
The position of the supply curve for real money balances depends on the monetary base,
the required reserve ratio, the excess reserve ratio, the currency to deposit ratio, and the
price level. Increases in the monetary base raise the real money supply, while an increase in
the required reserve, excess reserve, or currency to deposit ratio or in the price level leads
to a decrease in the real money supply.
Effect of an Increase in the Variable

Variable on the Real Money Supply
Monetary base +
Required reserve ratio
Excess reserve ratio
Currency to deposit ratio
Price level
15.4, meaning the interest rate rises to i2 and the equilibrium quantity of real
money balances falls to M%JP. The interest rate rises because the Fed’s action
reduces banking system reserves, which decreases the amount of loans banks
make. Thus, our money demand and supply framework shows how actions
by the Fed influence interest rates in the economy.
Notice that changes in the price level also change the real supply of
money, even if the nominal money supply remains unchanged. For instance,
holding the nominal money supply constant, a decrease in the price level
increases the ratio of Ms to P, thus shifting the real money supply curve to
the right. This results in lower interest rates. In particular, when the price
level falls goods become cheaper, and if the interest rate did not fall the
economy would have more real money balances than needed to make trans¬
actions. As idle real money balances are converted into other assets like
bonds, downward pressure is exerted on the interest rate until a new equilib¬
rium occurs at a lower interest rate.
Equilibrium with an Endogenous Money Supply

In Chapter 14 we developed our model of the endogenous money supply
curve, which views the currency to deposit ratio, cd, and the excess reserve
ratio, ed, as decreasing functions of interest rates. This gives rise to an
upward-sloping money supply curve. Figure 15.5 graphs the endogenous
real money supply and real money demand. Again the interest rate on loans
is on the vertical axis. The real money supply curve is upward sloping,
because increases in the interest rate lead to a higher money multiplier and
thus to an increase in the quantity of money supplied. The equilibrium in

Figure 15.5 is at the intersection of money supply and money demand at
point A, where the interest rate is i0 and M0/P is the equilibrium quantity of
real money balances.
Our analysis of changes in an exogenous money supply is similar to
what we learned for the case of an exogenous money supply. A Fed open
market purchases increases the monetary base, which shifts the real money
supply curve to the right as illustrated in Figure 15.6. The initial equilibrium
is at point A, with a real money supply of MVP. The increase in the monetary
base shifts the real money supply to M\/P, moving equilibrium to point B,
where the equilibrium interest rate is lower and the equilibrium quantity of
real money balances higher than at point A. Likewise, we can also see the
impact of any reduction in the money supply in Figure 15.6. Starting at the
initial equilibrium at point A, a reduction in the real money supply causes a

leftward shift in the real money supply from Ms/P to Ms2/P. The new equi¬
librium that occurs at point C has a higher interest rate and a lower equilib¬
rium quantity of real money balances. Thus, we see that the effects of
changes in an endogenous money supply are qualitatively the same as those
for an endogenous money supply: Increases in the real money supply lead
to lower interest rates.8 For simplicity, we will assume an exogenous money
supply in the remainder of our analysis.
Summary If tax rates increase and lead to a rise in tax avoidance activity, how might
Exercise 15.2 this affect money supply? How might it affect the interest rate and real
money balances? Assume the money supply is exogenous.
Answer: An increase in tax avoidance activity usually means, in part, an

increase in cash transactions. This is an increase in the currency to deposit
ratio, which lowers the money multiplier and hence the money supply. This
8 However, when the money supply is endogenous, a given change in the money supply results in
a smaller change in the interest rate than would occur if the money supply were exogenous.
reduction in the money supply causes a leftward shift in the real money
supply, as illustrated in Figure 15.4 by the shift from Ms/P to Ms2/P. The
interest rate would increase, and the quantity of real money balances would
decline.
Keynesian and Monetarist Views on Money Demand_
We have seen that both Keynes and Friedman viewed the demand for real
money balances as a decreasing function of interest rates. Despite this sim¬
ilarity, there is considerable disagreement about exactly how much the quan¬
tity demanded of money rises when interest rates fall. Followers of John
Maynard Keynes, called Keynesians, view money demand as relatively
sensitive to the interest rate. In contrast, Milton Friedman and other mone¬
tarists admit that interest rates affect money demand, but view the effect as
relatively small. Expressed differently, Keynesians view interest rates as a
very important determinant of money demand, whereas monetarists consider
it a relatively minor determinant.
We can use our demand and supply framework to shed light on how this
seemingly minor difference between Keynesians and monetarists actually
Keynesian and Monetarist Views on Money Demand 519
leads to a major difference in opinion about the impact of monetary policy.

Suppose the Fed decides to increase the money supply in an attempt to lower
interest rates and stimulate the economy. Would Keynesians and monetarists
agree on the quantitative impact of this monetary policy?
Part a of Figure 15.7 illustrates the monetarist view of equilibrium in
the money market, and part b shows the Keynesian view. The money demand
curve in part a is much steeper than that in part b, reflecting that monetarists
believe the quantity demanded of money is not very sensitive to changes in
the interest rate. In both cases the initial equilibrium is at point 1 in each
part, where the equilibrium interest rate is 5 percent.
Now suppose the monetary base increases due to an open market pur¬
chase by the Fed and everything else remains constant (including the price
level). Both Keynesians and monetarists would expect this activity to shift
Part a shows the mon¬ Figure 15.7

etarist view of money Keynesian and Monetarist Views of Money Demand
demand, while part b
shows the Keynesian
view. Under the mon¬
etarist view, an in¬
crease in the money
supply leads to a large
decline in the interest
rate, here from 5 to 2
percent. Under the
Keynesian view, an in¬
crease in the money
supply has little effect
on interest rates; here
it declines only from 5
to 4.5 percent.
the money supply curve to the right, resulting in a new equilibrium at point
2 in each part of Figure 15.7. But now we see a major difference caused by
the seemingly minor differences in the Keynesian and monetarist views of
money demand. In part a the interest rate falls to 2 percent, but in part b it
falls only to 4.5 percent. Since Keynesians believe the quantity demanded
of money is less sensitive to interest rates than monetarists do, Keynesians
expect monetary policy to have a relatively minor quantitative impact on
interest rates. This illustrates what many view to be an important difference
between Keynesians and monetarists. Monetarists view monetary policy as
having pronounced effects on the economy (in this case, on interest rates),
whereas Keynesians consider the effects of monetary policy to be relatively
minor.
We emphasize that the analysis of a one-shot increase in the money
supply illustrated in Figure 15.7 assumes everything else remains constant,
including the price level. But in Chapter 3, we learned that increases in the
money supply eventually lead to rises in the price level. When the price level
rises, the real money supply shifts back to the left (since as P increases, M7
P declines). This ultimately raises interest rates back toward their previous
level of 5 percent. Thus, while monetary policy can have a short-run impact
on interest rates, in the long run a one-shot increase in the money supply
leads to a higher price level, which may ultimately return interest rates to
their previous level. Monetarists and many Keynesians believe monetary
policy has no long-term effects on interest rates. The disagreement concerns
primarily the short-run effects of monetary policy on macroeconomic vari¬
ables like interest rates.
Summary Would the Keynesian and monetarist views of money demand lead to dif¬
Exercise 15.3 ferent conclusions regarding the impact of a reduction in the money supply
on interest rates? Explain.
Answer: While both Keynesians and monetarists believe a reduction in

the money supply would lead to higher interest rates in the short run, mon¬
etarists would expect rates to increase more. In the long run, however,
monetarists and Keynesians would agree that the price level would fall and
therefore interest rates would tend to rise toward their previous level.
Using the Loanable Funds Model

to Analyze Monetary Policy
There is another, equivalent way to look at the impact of monetary policy
on interest rates: the loanable funds model we developed in Chapter 4. The
money demand/supply and loanable funds frameworks are related because
when people demand more money, they in effect supply fewer loanable
Using the Loanable Funds Model to Analyze Monetary Policy 521
funds. The more money you hold, the less you have to lend in the bond
market. Let us briefly see how the loanable funds approach is an alternative
to the money demand/money supply approach we developed earlier.
Suppose the Fed increases the money supply. As we learned in Chapter
14, the primary way the Fed accomplishes this goal is through an open
market purchase, which increases the monetary base. The loanable funds
approach focuses on the impact of this Fed action in the loanable funds
market. In particular, in an open market purchase the Fed purchases existing
government securities, effectively financing government debt by “printing
money.” This reduces the amount of government debt instruments in the
loanable funds market, effectively reducing the demand for loanable funds.
(Wouldn’t your demand for loanable funds decline too if you had the ability
to print money?) This is represented by a decrease in the demand for loanable
funds in Figure 15.8, which leads to a new equilibrium in the loanable funds
market at point B and a lower interest rate. Thus, we reach the same conclu¬
sion using the loanable funds approach that we reached earlier with our
money demand/money supply framework: An increase in the money supply
(caused by an open market purchase) leads to a decline in the interest rate.
Even though both approaches are equivalent, there are advantages to
using money demand and supply analysis to evaluate the effects of monetary
policy.9 First, this approach explicitly incorporates factors that cause changes
in the money supply and explicitly shows that changes in the money supply
affect interest rates through its effect on the amount of money supplied.
Second, the loanable funds approach focuses on how changes in monetary
policy affect interest rates through its effect on the demand for loanable
funds, but says nothing about what happens to the equilibrium stock of
money in the economy as a result of an open market purchase. Since the
amount of money in the economy has far-reaching effects on such variables
as the price level, it is more informative to use money demand and supply
analysis to see the effects of monetary policy on both the interest rate and
the equilibrium stock of money.
While the money demand and supply framework is the easiest way to
see how monetary policy or changes in the price level affect interest rates,
the loanable funds framework developed in Chapter 4 is the simplest way
to see how other changes affect interest rates. For instance, it is generally
easier to use the loanable funds framework to see how changes in inflationary
expectations or risk affect interest rates. For this reason, the loanable funds
and money demand/supply frameworks are complementary tools in analyz¬
ing the banking and financial system.
Conclusion
The classical economists viewed income as the primary determinant of

money demand. Keynes reasoned that money demand depends on interest
rates as well as on income, and over the years several models have been
developed that provide microeconomic justifications for Keynes’s view. The
two major models are the inventory approach and the portfolio approach.
The modern quantity theory has broadened the view of money demand to
include not only income and interest rates but other determinants as well.
Our analysis in this chapter showed how the money demand and money
supply framework can be used to analyze monetary policy. The major con¬
clusion from using this approach is that an increase in the money supply
leads to a reduction in interest rates, at least in the short run. We also pointed
out some disagreements about the magnitude of such effects. Monetarists
view monetary policy as having pronounced effects on interest rates in the
short run, whereas Keynesians take the view that the short-run effects are
small. In Chapters 16, 19, and 20, we discuss monetary policy in more detail
and we examine the extent to which Fed policies affect real GDP, the level
of employment, and the price level.
9 If you are puzzled by the fact that these seemingly different approaches give identical results, an
analogy from arithmetic might help. There are two ways to calculate 1/2 X 100. One way is to
divide 100 by 2; the other is to multiply 100 by .5. Both methods are equivalent, but sometimes it
is easier to divide than to multiply. Similarly, sometimes it is easier to use the money supply and
demand framework to look at the determination of interest rates.
KEY TERMS
classical view of money demand modified Cambridge theory

simple quantity theory of money demand inventory approach to money demand
real money balances portfolio approach to money demand
Cambridge theory of money demand modern quantity theory of money demand
Cambridge constant human capital
transactions motive hyperinflation
precautionary motive Keynesians
speculative motive monetarists
Cambridge k
1. Describe the simple quantity theory of do they affect the demand for real money
money demand. Explain how this theory balances? Why?
is based on the transactions motive for
holding money and how changes in in¬ 7. What are the determinants of money sup¬
come and prices affect money demand ply in the model used in this chapter and
under this view. adapted from the one in Chapter 14? How
and why does each determinant shift the
2. Describe the inventory approach to money real money supply curve?
demand. Explain how this theory is based
on the transactions motive for holding 8. What is the effect of an increase in the
money and how interest rates and bank price level on real money demand? On the
fees affect money demand under this real money supply? On equilibrium in the
approach. money market?
3. Describe the portfolio approach to money 9. What are the effects of an increase in fees
demand. Does the transactions motive on deposits in the inventory model of
play any role in money demand under this money demand?
approach? Explain.
.
10 What is the effect of an increase in wealth
.
4 What are the determinants of money de¬ on money demand in the portfolio
mand under the inventory approach? Un¬ approach?
der the portfolio approach?
5. How does the modern quantity theory ex¬

.
11 There has been some discussion about
adding a transactions tax to all securities
pand on the simple quantity theory, on the
market purchases or sales. This would be
inventory approach, and on the portfolio
much like an additional broker’s fee, but
approach to money demand?
would be paid to the federal government.
6. What are the determinants of money de¬ What effects might this tax have on
mand in the modern quantity theory? How money demand?
12. Suppose the price level increases. Assum¬ 14. The Fed wants to achieve an interest rate
ing other things remain unchanged, what of 3 percent and a money stock of $700
impact will this change have on the equi¬ billion. The economy is currently in equi¬
librium interest rate? Could the Fed librium, with an interest rate and a money
change the monetary base in such a way stock that are both below the Fed’s tar¬
as to offset the higher price level’s effect gets. Can the Fed use its policy instru¬
on the interest rate? Explain. ments (like open market operations or re¬
serve requirements) to raise interest rates
13. Suppose real income increases. Assuming
and the equilibrium stock of money to
other things remain unchanged, what im¬
their desired levels? Explain.
pact will this change have on the equilib¬
rium interest rate? Could the Fed change 15. Why do Keynesians and monetarists
the monetary base in such a way as to off¬ sometimes disagree about the effects of
set the effect of higher real income on the monetary policy on interest rates?
interest rate? Explain.
Ahking, F. W. “International Currency Substitution: Velocity.” Economic Inquiry, 15 (January 1977),

A Reexamination of Britain’s Econometric Evi¬ 26-32.
dence: A Comment.’’ Journal of Money, Credit, Goldstein, H. N., and S. E. Haynes. “A Critical Ap¬
and Banking, 16 (November 1984), 546-556. praisal of McKinnon’s World Money Supply Hy¬
Amsler, C. “A ‘Pure’ Long-Term Interest Rate and pothesis [Currency Substitution and Instability in
the Demand for Money.’’ Journal of Economics the World Dollar Standard].” American Economic
and Business, 36 (August 1984), 359-370. Review, 74 (March 1984), 217-224.
Douglas, R. W., Jr. “A Three-Asset Determination of Graham, F. C. “A Note on the Vanishing Liquidity
the Transactions Demand for Money.” Journal of Effect of Money on Interest.” Economic Inquiry,
Macroeconomics, 11 (Winter 1989), 95-108. 24 (July 1986), 497-503.
Dutton, D. S., and W. P. Gramm. “Transactions Hafer, R. W., and D. W. Jansen. “The Demand for
Costs, the Wage Rate, and the Demand for Money in the United States: Evidence from Co¬
Money.” American Economic Review, 63 (Sep¬ integration Tests.” Journal of Money, Credit, and
tember 1973), 652-665. Banking, 23 (May 1991), 155-168.
Falls, G. A., and H. Zangeneh. “The Interest Rate Hoffman, D., and R. H. Rasche. “Long-Run Income
Volatility and the Demand for Money: The Empir¬ and Interest Elasticities of Money Demand in the
ical Evidence.” Quarterly Journal of Business and United States.” Review> of Economics and Statis¬
Economics, 28 (Winter 1989), 26-42. tics, 78 (1991), 665-674.
Gerdes, W. D. “The Demand for Money in Socialist Marquis, M. H., and W. E. Witte. “Cash Management
Tanzania.” Atlantic Economic Journal, 18 (Sep¬ and the Demand for Money by Firms.” Journal of
tember 1990), 68-73. Macroeconomics, 11 (Summer 1989), 333-350.
Girton, L., and D. Roper. “Theory and Implications Meyer, P. A., and J. A. Neri. “A Keynes-Friedman
of Currency Substitution.” The Monetary Ap¬ Money Demand Function.” American Economic
proach to International Adjustment. New York and Review, 65 (September 1975), 610-623.
London: Greenwood Press, Praeger 1986, 212- Swofford, J. L., and G. A. Whitney. “Nominal Costs
235. and the Demand for Real Transactions Balances.”
Goldberg, M., and T. B. Thurston. “Monetarism, Economic Inquiry, 23 (October 1985), 725-740.
Overshooting, and the Procyclical Movement of
PART FIVE
Money and the

Macro Economy
CHAPTERS
16
A Simple Macroeconomic Model
17
Open Economy Macroeconomics
18
Money and Economic Activity:
A Look at the Evidence
A Simple Macroeconomic
Model
he United States is one of the most productive and prosperous nations

in the world, producing more than one-fourth of the world’s total output
with a mere 5 percent of the world’s population. Despite this prosperity,
periods of recession occur during which the real output produced in the
United States declines. During these hard times, politicians and economists
frequently propose policies designed to stimulate the economy. Undoubtedly
you have heard commentators debate the pros and cons of boosting the
economy by changing government spending or taxes, lowering interest rates,
or expanding the money supply. Have you ever wondered precisely how and
why these policies affect prices and output in the economy, or why so much
disagreement exists about the “best” policy? In this chapter, we synthesize
much of what we learned in previous chapters into a single, simple frame¬
work that allows us to answer this question.
We begin by developing a model of the aggregate demand and supply
of goods and services in the economy that explains how variables such as
the money supply, government borrowing, and taxes affect interest rates, the
price level, and real output (real GDP). We then use our model of the
macroeconomy to examine the impact of government policies designed to
stimulate the economy.
When you study the macroeconomic model we develop, you should be
aware that there are competing approaches to developing both aggregate
demand and aggregate supply frameworks. For example, we looked at the
equation of exchange and the quantity theory of money in Chapter 3. We
can use these concepts to develop an aggregate demand curve that empha¬
sizes the importance of the nominal money supply and velocity. We will use
an alternative approach that is fairly simple yet flexible and can be adapted
in various ways to address important economic issues. There are still other
approaches, however, including one that develops aggregate demand from
what is known as the IS-LM model. In the appendix to this chapter, we see
how the approach we use in the text can be used to construct an IS-LM
model and then to derive aggregate demand from that model.
Finally, we concentrate on a model in which net exports are exogenous.
If net exports are zero, this is called a closed economy model, a model that
does not analyze the impact of international trade on the aggregate
526
economy. We consider this model first to avoid the complications introduced

by endogenous net exports when first presenting our macroeconomic model.
After developing this essentially closed economy model, we expand it to an
open economy model that does consider the impact of international trade on
the aggregate economy in Chapter 17.
Aggregate Demand
The first step in building a macroeconomic model of the economy is to
describe aggregate demand, the total demand for all goods and services in
the economy. Aggregate demand is, in essence, the demand for real gross
domestic product (the sum of consumption, investment, government pur¬
chases, and net exports). Consumption goods are the goods and services
demanded by households—goods such as food and medicine, rental pay¬
ments on housing, and purchases of leisure items. Investment goods rep¬
resent outlays for the purchase of new machines, buildings, and structures,
things economists call physical capital. Goods and services demanded by
government make up the third category, called government purchases.
Government purchases include spending on national defense, highways, and
other goods and services. However, not all government spending is govern¬
ment purchases. A government transfer payment is government spending,
but it is not the purchase of goods and services. Instead, transfer payments
are funds dispensed to individuals with no required exchange of goods,
services, or labor. Transfer payments include such things as unemployment
insurance payments and welfare payments. The final category of aggregate
demand is net exports, the difference between exports and imports. For the
United States, exports are sales of U.S. goods and services overseas, while
imports are purchases of foreign-made goods and services by U.S. residents.
The difference, net exports, represents the net demand for goods and services
produced by the United States from international trade and may be positive
or negative. We say more about net exports in Chapter 17.
Aggregate demand (AD) in the economy is defined as the total demand
for all four categories of goods and services:
AD = Cd + Id + Gd + NXd,
where Cd is the total demand for consumption goods, Id is total investment

demand, Gd is government demand for goods and services, and NXd is net
demand for exports.
In microeconomics, we define a demand curve as the graph of the
relationship between the price of a good and the quantity demanded of the
good, holding all other variables constant. Similarly, we define an aggregate
demand curve as the relationship between the aggregate price level (P) and
the aggregate quantity of goods and services demanded (F), holding other
variables constant.
528 Chapter 16 A Simple Macroeconomic Model
Figure 16.1 shows an aggregate demand curve. Note that an increase in

the price level from P0 to Px leads to a decrease in the aggregate quantity
demanded of all goods and services from F0 to Yx.
Why the Aggregate Demand Curve Slopes Downward

One of the most important ideas in economics is the law of demand, which
states that as the price of a good rises and other prices remain the same,
consumers will purchase less of that good. There are two reasons for this.
First, the good has become more expensive relative to other goods, making
it attractive for the consumer to substitute toward other goods. This is called
the substitution effect. Second, the increase in price reduces the consumer’s
purchasing power, since any given income level purchases a smaller quantity
of the good whose price has increased. This is called the income effect.
While it is tempting to apply the law of demand to the aggregate demand
curve to explain its downward slope, a word of caution is in order. The
aggregate demand curve is a relationship between the aggregate price level,
P, and the quantity demanded of all goods and services in the economy, Y.
Since the price level is a measure of the prices of all goods and services in
This figure illustrates Figure 16.1

the aggregate de¬ Aggregate Demand
mand curve. Notice
that as the price level
rises, the quantity de¬
manded of goods and
services falls.
the economy, an increase in the price level is an increase in all prices; this
increase does not lead to substitution toward other goods.
Why, then, is the aggregate demand curve downward sloping? There are
three reasons, two of which we will discuss here and one that we reserve
until the next chapter. The first is a wealth effect, similar in some ways to
the income effect just mentioned. This effect arises because an increase in
the overall price level reduces the purchasing power of a consumer’s money.
For example, suppose you have $1,000 in your checking account. Your
$1,000 gives you more purchasing power when the price level is 100 than
when it is 200 and all goods are twice as expensive. Keeping other things
the same, an increase in the price level reduces the purchasing power of
your money holdings and leads to a decrease in the quantity of goods and
services you demand. Moreover, an increase in the price level reduces the
purchasing power of everyone’s money holdings, so the aggregate quantity
demanded of goods and services is decreased. Thus, one reason the aggregate
demand curve is downward sloping is due to this wealth effect: A higher
price level reduces the purchasing power of money holdings, leading to a
lower aggregate quantity demanded.1
A second reason the aggregate demand curve slopes downward is due
to a real interest rate effect. Recall from Chapter 15 that we can look at
equilibrium between money demand and money supply as determining the
interest rate. In that chapter, we presented money demand and money supply
as demand for and supply of real money balances. An increase in the price
level lowers the real money supply, shifting the money supply curve to the
left and creating a higher equilibrium interest rate. Holding other variables
constant (including the expected inflation rate), this higher interest rate is a
higher real interest rate and causes a reduction in aggregate demand. Both
investment demand and demand for consumption spending on durable goods
will decline. While low interest rates encourage borrowing and spending on
capital goods such as factories, houses, and consumer durables, high interest
rates discourage this borrowing and spending and thus reduce the aggregate
quantity demanded.2 Hence, a higher price level results in a lower real money
supply, which directly reduces aggregate quantity demanded via a wealth
effect and indirectly reduces aggregate quantity demanded through an in¬
crease in the interest rate, which discourages borrowing and spending.
There is a final reason for downward-sloping aggregate demand, which
arises from the effect of the price level on net export demand. We discuss
1 Because this wealth effect works through changes in the purchasing power of money holdings, it
is often called a real balance effect. Real money balances, M/P, are reduced when the price level
increases, and hence the aggregate quantity of goods and services demanded declines.
2 This is sometimes called the intertemporal substitution effect to stress the fact that a higher
interest rate encourages a substitution of fewer goods purchased today in exchange for more goods
purchased in the future.
this effect in the next chapter, where we look at net export demand and the
open economy in detail.
The Components of Aggregate

Demand and Their Determinants
Just as the demand by a consumer for an automobile depends on factors
other than the price of the car (such as the consumer’s income), so does
aggregate demand depend on factors other than the price level. These factors,
which ultimately determine the position of the aggregate demand curve, are
known as the determinants of aggregate demand. The easiest way to under¬
stand the determinants of aggregate demand is to consider the determinants
of each component of aggregate demand: consumption, investment, and
government purchases, which we turn to next. We will discuss determinants
of the final component of aggregate demand, net exports, in Chapter 17. Box
16.1 shows the relative sizes of the four components of aggregate demand
for the U.S. economy.
Determinants of Consumption Demand. The consumption

component of aggregate demand consists of spending by households on
goods and services that directly satisfy their wants, such as food, medicine,
rental payments on housing, and leisure items. We will look at consumption
demand, and indeed at all components of aggregate demand, in real terms.
Thus, consumption demand is the real quantity demanded of consumption
goods and services.
Consumption demand depends on many variables, including the price
level, wealth, the real interest rate, and taxes, as well as household tastes
and expectations. It is convenient to summarize all the factors that influence
consumption demand with the consumption demand function
C = C(P, W, r, 7Te, Th\
where C* is the total quantity demanded of all consumption goods, P is the

price level, W is household real wealth (including consumer holdings of
money, bonds, stocks, and other assets like real estate), r is the ex ante (or
expected) real interest rate calculated as r = i — ire, tY is the expected
inflation rate, and TH is taxes paid by households. Consumer tastes or pref¬
erences also affect consumption demand but are not included as a separate
variable.
You may notice that one important variable is missing from this list:
income. We have all seen simple consumption functions specifying that
consumption is a function of income. Why is income not listed in the above
specification? The answer is that consumption does depend on income, but
our specification of aggregate demand already takes into account the effect
of income on consumption demand. We will demonstrate how this works
when we discuss aggregate demand as the sum of these four components;
for now we simply outline the argument. Aggregate consumption does de-
Components of Aggregate Demand
We divide up total spending in the economy, of goods and services (not transfer payments)
GDP, into four components: consumption, in¬ were $939 billion, while gross investment in
vestment, government purchases, and net ex¬ new capital (not subtracting depreciation) was
ports. The accompanying figure shows values in $742 billion. Net exports were a negative num¬
1987 dollars for these four components in 1992. ber, -$55 billion, reflecting the U.S. interna¬
Consumption was by far the largest component, tional trade deficit.
equal to $3,354 billion. Government purchases
00 2,000 H
O'!
m 1,500 -
C
o
-500
Consumption Government
Investment Net Exports
Expenditures Purchases
Source: U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, various issues, and
pend on aggregate income, and in equilibrium aggregate income equals

aggregate spending, or GDP. But GDP equals consumption plus investment,
government purchases, and net exports. Thus, we have a somewhat circular
situation in which consumption depends on income and income depends on
consumption plus other components of GDP. We demonstrate later that our

specification of aggregate demand—and our specification of consumption
demand—can be written in a form that already incorporates the effect of
income on consumption. As we will see, this effect makes consumption
demand and aggregate demand less steep than it would appear with income
held fixed, but otherwise does not change the basic downward slope of these
curves or the effects of changes in the determinants of consumption demand
or aggregate demand on these curves.
If we hold everything constant except the price level and consumption,
we obtain the relationship between the price level and real expenditures
graphed as the curve C1 in Figure 16.2. This consumption demand curve
Figure 16.2
Consumption Demand
The demand curve for consumption goods shows the inverse relationship between con¬
sumption demand—consumption expenditures—and the price level. A rise in the price
level will lead to a decrease in quantity demanded, such as a change from point A to point
B. A change in the determinants of consumption demand other than the price level will
change the position of the consumption demand curve. The consumption demand curve
will shift to the right if there is an increase in wealth, a decrease in the real interest rate, a
decrease in household taxes, or an increase in inflationary expectations. The consumption
demand curve will shift to the left if there is a decrease in wealth, an increase in the real
interest rate, an increase in taxes, or a decrease in inflationary expectations.
indicates that, holding other things constant, an increase in the price level
will reduce demand for real expenditures on consumption goods as illustrated
by the movement from A to B in Figure 16.2. This occurs because an increase
in the price level reduces the real value of consumer wealth, especially money
holdings, and thereby reduces the quantity demanded of aggregate con¬
sumption goods and services. Thus, from changes in the price level, a wealth
effect occurs that operates through consumption. In addition, changes in the
price level will change the real money supply and hence change the equilib¬
rium interest rate in the money market, resulting in a further change in the
quantity demanded of aggregate consumption goods and services. Both of
these effects are summarized by the price level, P, in the consumption
demand function.
In contrast, whenever there is a change in wealth or the real interest rate
that is not due to changes in the price level, or whenever changes in the
expected inflation rate, taxes paid by households, or tastes occur, the entire
consumption demand curve will shift. This is illustrated in Figure 16.2 by a
shift from Cd to either Cf or Cd. Now we look at how and why these
determinants shift the consumption demand curve.
Wealth. An increase in household real wealth—be it in the form of

money, bonds, stocks, real estate, or other assets—increases the opportunities
households enjoy for purchasing goods and services. Consequently, an in¬
crease in aggregate household wealth from sources other than a change in
the price level will tend to increase the total demand for consumption goods.
This is represented by a rightward shift in the consumption demand curve
in Figure 16.2 from Cd to Cf.
Wealth can change for any number of reasons. Wealth tomorrow is
wealth today, plus any income generated by that wealth, plus other income
such as wage and salary income, minus consumption spending and taxes.
Wealth may change if there is an increase in the money stock, holding the
price level constant, or if an appreciation in value of real estate or stock
market holdings occurs, both of which will be income in the form of capital
gains. Any of these increases in wealth will make consumers feel wealthier
and lead to increased consumption demand.
The Real Interest Rate. Not all changes in the interest rate are due
to changes in the price level. Changes in the real interest rate may arise for
any number of reasons and will alter households’ incentives to save or spend.
While changes in the real interest rate have both an income and a substitution
effect, we will assume higher real interest rates lead you and other consumers
to save more, since the return to saving is now higher. However, when you
save more at a given level of income, you have less left over for consumption
spending. Similarly, the lower the real interest rate, the greater will be
consumer spending. Graphically, an increase in the real interest rate shifts
the consumption demand curve to the left, whereas a decrease in the real
interest rate shifts the consumption demand curve to the right.
534 Chapter16 A Simple Macroeconomic Model
Expected Inflation. If the expected inflation rate increases, consumers

will tend to purchase more consumption goods today, before prices rise.
Thus, increases in expected inflation will raise consumption demand (shifting
it to the right), whereas a decrease in expected inflation will lower con¬
sumption demand.
Taxes. Taxes on consumption have the obvious effect of decreasing con¬

sumption demand, but other taxes affect consumption demand as well. Taxes
on income, such as federal and state income taxes and the social security
tax, have an indirect effect on household wealth. Income taxes reduce house¬
holds’ ability to save for any given income level, which may reduce their
ability to purchase consumer goods. Thus, increases in household income
taxes may reduce consumption demand, shifting consumption demand to the
left. A decrease in taxes would have the opposite effect, increasing con¬
sumption demand (a shift to the right).
Tastes. Consumer tastes can also affect consumption demand. For ex¬
ample, households may experience an increase in thriftiness. This will de¬
crease consumption demand, which is represented by a leftward shift in the
consumption demand curve.
The second component

of aggregate demand is investment, the purchase of new physical capital.
Firms buy physical capital today to increase their future profits, by either
expanding their productive capacity or making their current productive ca¬
pacity more efficient. Thus, the gain firms realize from investing in physical
capital is the expectation of higher future profits. The greater the expected
future value from a given expenditure today, the greater will be investment
demand.
There is another side to investment: the cost of paying for physical
capital today. To pay for investment, a firm might borrow loanable funds.
In this case, investment has a very explicit cost: the interest payments on
the borrowed funds. Alternatively, the firm might use internal funds such as
retained earnings to fund investment. Then the cost to the firm of the in¬
vestment is the implicit interest it forgoes by not investing the funds in the
loanable funds market. Thus, regardless of whether a firm uses retained
earnings or the capital market to fund investments, the relevant price of an
investment is the real interest rate. The higher the real interest rate, the less
desirable firms will find it to engage in investment.
To see why the real interest rate plays a central role in determining the
level of investment, consider the following investment opportunity faced by
a firm. The firm can purchase a new machine for $150 million today, year
1, which it expects will generate returns in real terms, adjusted for inflation,
of $100 million in year 2 and another $100 million in year 3. After year 3
the machine will have no value, and the firm plans to give it to the junkyard
in exchange for hauling it away. Will the firm choose to purchase this
machine? The answer depends on the present value of the costs and earnings
of the investment. The present value of the cost is simply the $150 million
spent in year 1. The present value of the earnings depends on the real interest
rate and is calculated as
$100,000,000 $100,000,000
PV
(1 + r) (1 + r)2
For instance, if the real interest rate is zero, the present value of earnings is
just $200 million and the present value of the firm’s profits is $200 million
— $150 million, or $50 million. If the real interest rate is 10 percent, or r
= .10, the present value of earnings is $173.55 million and the present value
of profit is $23.55 million. If the real interest rate is 20 percent, the present
value of earnings is $152.78 million and the firm earns a profit of $2.78
million in present value terms. This example illustrates how a rise in the real
interest rate reduces the profits from an investment.
Another way to describe the firm’s decision is to calculate the rate of
return on the investment project that makes the present value of earnings
equal to the present value of costs; that is, what is the real interest rate that
just makes
$100,000,000 $100,000,000
$150,000,000?
(1 + r) + (1 + r)2
The answer is that a real interest rate of about 21.5 percent will make the
costs equal to the present value of earnings, so profits are zero. At any higher
real interest rate, this firm will not pursue the investment opportunity, be¬
cause the cost of the investment will exceed the present value of the firm’s
earnings. More generally, for an investment yielding a given stream of
expected future earnings and expected costs, the higher the real interest rate,
the less likely a firm is to make the investment. For the entire economy, with
many firms facing different investment opportunities, increases in the real
interest rate will make some investment opportunities unprofitable. The
higher the real interest rate, the fewer will be the investment opportunities
that remain profitable for firms.
Taxes on the earnings of firms—either corporate profits taxes or indi¬
vidual income taxes on business owners—will also reduce investment. Con¬
tinuing with the preceding example, suppose the government imposes a 10
percent income tax on the earnings of the firm over and above the initial
cost of the investment. In year 1 no tax is due, since the $100 million in
earnings has not yet covered the initial cost of $150 million. In year 2 the
first $50 million in earnings is also exempt from taxation, but the next $50
million is subject to the 10 percent tax, since earnings will exceed the cost
of the machine. Thus, after taxes the firm will earn $95 million in year 2—
the pretax earnings of $100 million minus the 10 percent tax on the $50
million of net earnings, or $5 million.
What is the present value of the aftertax profits in this example? With a
real interest rate of zero, the present value of aftertax earnings is $195 million
and the present value of costs is $150 million, so the present value of aftertax
profits is $45 million. If the real interest rate is 10 percent, the present value
of aftertax earnings is $100 million/1.10 + $95 million/1.102, or $169.42
million, so the present value of aftertax profits is $19.42 million. Finally, if
the real interest rate is 20 percent, the present value of aftertax earnings is
$149.31 million, so the present value of aftertax profits is - $0.69 million—
a loss. The income tax has transformed an investment that was profitable at
20 percent interest into one that is unprofitable at that real interest rate. This
is generally the case: The higher the tax rate on income or earnings, the less
likely the firm will be to make an investment at any given real interest rate.
Since investment demand depends on expected future profits, the real
interest rate, and income taxes, we can summarize the relationship between
investment demand and its determinants with the following investment de¬
mand function:
I'1 = I(P, EE, r, 7>,).

Here Id is the real quantity demanded of total investment goods, r is the ex
ante real interest rate, EE is expected earnings, and 7> is income taxes paid
by owners of firms.
Figure 16.3 illustrates the inverse relationship between the real interest
rate and investment by firms. When the real interest rate is 5 percent, 200
billion real dollars of investment spending will occur. If the real interest rate
increases to 10 percent, only 150 billion real dollars of investment spending
will occur. Based on Figure 16.3, we see that the higher the real interest
rate, the lower the real quantity demanded of investment goods.
Why does investment demand depend on the price level? Our investment
demand function contains no wealth effect. However, we have already seen
how changes in the price level change the real money supply and hence the
equilibrium interest rate in the money market. An increase in the price level,
by lowering real money supply, will raise the interest rate and, for a given
level of expected inflation, increase the real interest rate. This results in a
reduction in the quantity demanded of aggregate investment goods and serv¬
ices.
Of course, it is important to remember that the price level effect via
changes in the interest rate is only one way interest rates can change and
thereby alter investment demand. Changes in the real interest rate due to
factors other than a change in the price level are represented by the real
interest rate variable in the investment demand function.
We can also graph investment demand, Id, as a function of the price
level, as we do in Figure 16.4. Here the price level is measured on the
vertical axis and real expenditures for investment goods and services on the
horizontal axis. The downward slope of the investment demand curve is due
The investment de¬

mand curve shows the Investment Demand and the Real Interest Rate
inverse relationship
between the real in¬
terest rate and invest¬
ment demand, hold¬
ing constant any other
factors that might af¬
fect investment de¬
mand. Here a rise in
the real interest rate
from 5 to 10 percent
causes a decline in the
quantity demanded of
investment goods
from $200 billion to
$150 billion (in real
dollars).
to the effect of changes in the price level on the interest rate and hence on
investment demand. Changes in expected future profits, in the real interest
rate (due to factors other than the price level), or in income taxes will all
shift the investment demand curve in Figure 16.4 to the right (to /f) or to
the left (to li). We summarize the causes of such shifts next.
Expected Earnings. Expected earnings are the reason a firm invests.

The higher the level of future expected earnings, the greater is the firm’s
interest in investment spending. This would be represented by a rightward
shift in investment demand. A decrease in expected future earnings would
shift investment demand leftward.
The Real Interest Rate. The real interest rate is the opportunity cost
of purchasing physical capital, so an increase in the real interest rate results
in a reduction in real investment demand. This is represented by a leftward
shift in investment demand. Similarly, a decrease in the real interest rate
reduces the opportunity cost of investment and leads to a rightward shift in
investment demand.
Taxes on Income from Investment in Physical Capital. If

government tax policy reduces the expected aftertax profitability of invest¬
ing, firms will make fewer investments. This shifts the investment demand
curve to the left. Conversely, tax policies that increase the aftertax profita¬
bility of investments (such as investment tax credits) shift the investment
demand curve to the right.
Government purchases of goods and serv¬

ices are the third component of aggregate demand and result from the polit¬
ical process. For now we take this process to be exogenous, meaning gov¬
ernment purchases of goods and services do not depend on interest rates,
prices, or any other variable. Box 16.2 describes how government spending
has been divided among government purchases, interest payments on the
debt, and government transfer payments in various past years.
When we graph government purchases against the price level, the gov¬
ernment’s demand for goods and services is vertical, or perfectly inelastic,
as illustrated by the curve Gd in Figure 16.5. We will assume the government
demand for goods and services is
Gd = G,
Components of Government Spending;

1962, 1977, and 1992
Total government spending includes govern¬ 1977, government purchases were 34.5 percent
ment demand for goods and services, labeled of government spending, while transfer pay¬
Gd in the text, plus government transfer pay¬ ments had climbed to 40.1 percent and interest
ments and net interest payments on the na¬ payments to 7 percent. In 1992, transfer pay¬
tional debt. In 1962, purchases of goods and ments were 42.9 percent of government spend¬
services made up 58.9 percent of government ing, purchases of goods and services were 30.5
spending, followed by transfer payments (25.1 percent, and interest payments had risen to
percent) and interest payments (6.3 percent). By Continued on p. 540
1962 1977
Other
Purchases of Goods Purchases of Goods
(18.4%)
and Services
and Services
(58.9%) Interest
Interest (34.5%)
Payments
Payments (7.0%)
(6.3%)
Transfer Payments
(40.1%)
Transfer Payments
(25.1%)
1992
Other
(14.2%) Purchases of Goods
Interest and Services
Payments (30.5%)
(12.4%)
Transfer Payments
(42.9%)
Continued from p. 539 ernment spending devoted to paying interest on

12.4 percent. Thus, there has been a large in¬ the national debt.
crease in government transfer payments relative
to government purchases over the last three Source: U.S. Department of Commerce, Bureau of Eco¬
decades, as well as a rising percentage of gov¬ nomic Analysis, Survey of Current Business, various issues,
and Citibase electronic database.
where G is the exogenous level of government purchases. Given these as¬

sumptions, an increase in government purchases shifts the government de¬
mand for goods and services to the right (to Gf), whereas a decrease in
government purchases shifts government demand to the left (to Gd), as in
Figure 16.5.
Net Exports. Net exports depend on several factors that we explore in

Chapter 17. For now we take net exports as given and not a function of
either the interest rate or the price level. When we graph net export demand
against the price level, we get a vertical line, indicating net export expen¬
ditures do not change with changes in the price level. In this way, net exports
are treated in this chapter in the same way government purchases are—as
exogenous. Hence we write
NXd = NX.
Adding Up the Components to Obtain Aggregate Demand

We have outlined the four major components of aggregate demand: con¬
sumption, investment, government purchases, and net exports. We have also
discussed the determinants of each component and examined the effects of
changes in the determinants of each. Now we turn our attention to aggregate
demand, which is the sum of these components:
AD = Cd + Id + Gd + NXd.
Figure 16.6 illustrates that aggregate demand is the sum of consumption,

investment, government, and net export demand by combining them in one
graph. The aggregate demand curve is the horizontal summation of con¬
sumption demand, investment demand, government purchases demand, and
net export demand. For example, when the price level is P0, investment
demand is at point I on the Id curve, government demand is at point G on
the Gd curve, net export demand is at point NX on the NXd curve, and
consumption demand is at point C on the C1 curve. If we add the horizontal
distances together, we get the point labeled / + NX + G + C on the AD
curve, which represents aggregate demand when the price level is P0. Notice
Government pur¬ Figure 16.5

chases of goods and Government Purchases
services are assumed
to be the result of a
political process and Price
independent of any Level
~d c
economic variable. c4 < 4
Therefore, govern¬
ment demand for
goods and services is
represented by a verti¬
cal line. An increase in
government demand
is reflected in a right¬
ward shift in the verti¬
cal line, such as from I G t G
Gd to 6f. A decrease
in government de¬
mand is represented
by a leftward shift in
the vertical line, such
as from Gd to Gd.
Real Expenditures on Government Purchases
that the aggregate demand curve is downward sloping because consumption

and investment demand are downward sloping.
Why Aggregate Demand Does Not Depend

on Income
We use an aggregate demand curve that subsumes the effect of income on
aggregate demand. Figure 16.7 shows how we can justify this approach.
Consider first the curve labeled ad(Y0), which is desired aggregate expen¬
ditures when income is Y0. Suppose we start at point A, where the price level
is P0. What happens when the price level falls to P{!
On ad(Y0) we move to point B, with a price level of Px. However, on
ad(Y0) spending is a function of income, and income is held constant at F0-
But this is not an equilibrium position, since in equilibrium aggregate income
must equal aggregate real expenditures. When real expenditures increase to
Y\, aggregate income in the desired aggregate spending equation also in¬
creases to Yx in equilibrium. This increase in income raises consumption and
shifts aggregate expenditures to the right from ad(Y0) to ad{Y\). Thus,
when real expenditures and income increase to Yx, increasing consumption
Aggregate demand is Figure 16.6

the summation of in¬ Derivation of Aggregate Demand
vestment demand,
government purchases
demand, consumption
demand, and net ex¬
port demand. Notice
the aggregate de¬
mand curve slopes
downward due to the
consumption and in¬
vestment demand
curves.
and hence aggregate spending, we are at point B' on ad(Y{), with a price
level of Pi.
Similarly, starting at point A and with real expenditures equal to real
income of F0, we ask what will happen when real expenditures decline to
Y2. When real expenditures fall to Y2 but aggregate income is held constant
at Y0, we move to point C on ad(Y0), where the price level is P2. As before,
however, this is not an equilibrium, since total real spending must equal total
real income in macroeconomic equilibrium. Thus, when real expenditures
decline to Y2, we want real income to also drop to Y2. This reduces con¬
sumption and hence aggregate spending, shifting us to ad(Y2). On ad(Y2),
real expenditures of Y2 place us at point C', where the price level is P2.
To summarize, we have found three equilibrium points where total real
expenditures equal total real income: C', A, and B'. We draw a curve con¬
necting these points and call it aggregate demand, or AD. However, this AD
curve does not depend on the level of income. Instead it is the AD curve we
derive under the condition that total expenditures and total income must be
equal. In this way, we derive consumption demand C^(r, W) and hence
aggregate demand AD, which do not depend on income.
Changes in Aggregate Demand

Changes in the price level lead to a movement along the aggregate demand
curve (such as the movement from A to B in Figure 16.8), whereas a change
in the determinants of any component of aggregate demand causes a shift in
the aggregate demand curve to either ADX or AD2. Thus, the determinants
of aggregate demand are simply the determinants of the components of
aggregate demand.
We can emphasize these determinants by writing aggregate demand as
AD = AD(P, r, ire, W, TH, 7>, EE, G, NX).

Figure 16.8
Changes in Aggregate Demand
_ _
■tmm
Since aggregate demand is the sum of investment demand, government demand, net ex¬
port demand, and consumption demand, any factors that cause these to shift will also
cause the aggregate demand curve to shift. Therefore, the aggregate demand curve will
shift to the right if there is an increase in wealth, an increase in government purchases, a
rise in inflationary expectations, an increase in expected future profits, a decrease in the
real interest rate, a decrease in taxes, or an increase in net exports. The aggregate demand
curve will shift to the left if there is a decrease in wealth, a decline in government pur¬
chases, a decrease in inflationary expectations, a reduction in expected future profits, an
increase in the real interest rate, an increase in taxes, or a reduction in net exports.
The effect of each variable in parentheses on aggregate demand is determined

by its impact on the three components of aggregate demand discussed above.
Table 16.1 summarizes these effects.
Before concluding this section, we again point out there are alternative
ways to derive aggregate demand. Some might prefer the quantity equation
of Chapter 3 and others what is known as the IS-LM model In the appendix
to this chapter, we outline a version of the IS-LM model that builds on the
four components of aggregate expenditures we introduced in this chapter.
Summary Holding other things constant, determine the impact of each of the following
Exercise 16.1 scenarios on consumption demand, investment demand, and aggregate de¬
mand: (a) an increase in the real interest rate not caused by a change in the
price level, (b) an increase in taxes paid by households, and (c) an investment
tax credit for businesses.
(a) An increase in the real interest rate dereases both consump¬

tion demand and investment demand. Thus, an increase in the real interest
rate decreases aggregate demand (shifts aggregate demand to the left), (b)
Table 16,1
Determinants of Aggregate Demand
Change in Determinant Effect on Effect on

of Consumption Demand Consumption Demand Aggregate Demand
Increase in wealth (W) Shifts right Shifts right

Increase in real interest rate (r) Shifts left Shifts left
Increase in expected inflation Shifts right Shifts right
rate (ire)
Increase in taxes on house- Shifts left Shifts left
holds (Th)

of Investment Demand Investment Demand Aggregate Demand
Increase in real interest rate (r) Shifts left Shifts left

Increase in expected earnings Shifts right Shifts right
(EE)
Increases in taxes on firms Shifts left Shifts left
(tf)

of Government Demand Government Demand Aggregate Demand
Exogenous increase in govern- Shifts right Shifts right

ment purchases (G)

of Net Exports Net Exports Aggregate Demand
Exogenous increase in net Shifts right Shifts right

exports (NX)
An increase in taxes paid by households decreases consumption demand,

but has no effect on investment demand. Thus, an increase in taxes paid by
households decreases aggregate demand (shifts aggregate demand to the
left), (c) An investment tax credit increases the expected aftertax profits of
businesses, thus increasing investment demand. It has no effect on con¬
sumption demand. Thus, an investment tax credit increases aggregate de¬
mand (shifts aggregate demand to the right).
Short-Run^ Aggregate Supply

Short-run aggregate supply (SRAS) characterizes the supply side of the
macroeconomy in the short run. More specifically, short-run aggregate sup¬
ply summarizes the total quantity of goods and services an economy will
produce at various price levels, holding constant such things as the prices of
inputs, taxes, the stock of physical capital available to firms, and the level
of technology. We can summarize the dependence of short-run aggregate
supply on these variables by the relation
SRAS = SRAS(P, wL, Z, TL, K, Tech),
where SRAS represents the total (or aggregate) real quantity of goods and
services supplied, P is the price level, wL represents nominal wages paid to
labor, Z denotes the price of variable inputs other than labor (such as ma¬
terials or energy inputs), TL represents taxes on labor, K stands for the capital
stock (which is fixed in the short run), and Tech represents the available
technology.
The short-run aggregate supply curve holds everything but the price
level constant and is sketched in Figure 16.9. An increase in the price level
leads to a movement along a given short-run aggregate supply curve, such
as from point A to point B along the curve labeled SRAS. A change in one
of the other variables that influence aggregate supply, such as a change in
the price of inputs, shifts the short-run aggregate supply curve to the right
(to SRASi) or to the left (to SRAS2). Next, we explain why the short-run
aggregate supply curve slopes upward and how changes in input prices,
taxes, the capital stock, or technology affect the position of the short-run
aggregate supply curve.
Why the Short-Run Aggregate

Supply Curve Slopes Upward
The reason the short-run aggregate supply curves in Figure 16.9 slope up¬
ward is that we are holding the nominal wage rate and other nominal input
prices constant along each aggregate supply curve. As the price level in¬
creases, firms are able to sell their goods at higher prices, while continuing
to pay the same nominal wage to workers or the same prices for other inputs.
Consider what this means to workers. With their nominal wages held con¬
stant and the price level increasing, their real wages—the ratio of the nominal
Short-Run Aggregate Supply 547
wage to the price level, or wLIP—will decline. Thus, the increase in the price
level results in higher output along the short-run aggregate supply curve
precisely because firms find it less expensive in real terms to hire workers
and buy other inputs. This induces firms to use more of these inputs and
leads to greater aggregate output. As we will see later in this chapter, if
nominal input prices (such as nominal wages paid to workers) increase in
the same proportion as the rise in the price level, firms will not use more of
these inputs, output will not increase, and the aggregate supply curve will
be vertical.
You may be thinking that if the price level rises by 10 percent, workers
will receive a 10 percent cost-of-living adjustment to compensate for higher
prices, and this will keep real wages constant. However, many economists
stress that in the short run, wages adjust more slowly than does the overall
price level. This view has a long history in economics and has come to be
considered a Keynesian view, since Keynes argued that nominal wages are
rigid—unresponsive to changes in the price level. However, not only
Keynesians suggest nominal wages are not perfectly flexible in response to
price level changes. Friedman and other monetarists have also made such
claims, and both new Keynesians and new classical economists have made
relatively sophisticated arguments to this effect.
One reason wages do not adjust instantaneously to changes in the price
level is that many wages are set by formal contract between workers and
firms and adjusted only at set intervals, say, every six months or every year.
Union workers have contracts that often set wages for a period of three years.
Even workers without formal contracts often receive wages that are adjusted
only at intervals. If you have ever had a summer or part-time job, you were
probably offered a certain nominal wage that was adjusted only at regular
intervals, if at all, and did not adjust each month when the government
announced new values for the consumer price index. Thus, we see that both
formal and informal contracts mean nominal wages adjust only periodically,
and we say there is nominal wage rigidity (or nominal wage stickiness).
Due to this nominal wage rigidity, the price level can increase and at least
temporarily depress the real wage, giving rise to an upward-sloping aggregate
supply curve in the short run. A similar argument may be made about other
input prices, since firms often have contracts with suppliers that specify
prices over months or even years.
An alternative argument about nominal wage rigidity can be made based
on price expectations. According to this argument, nominal wages are flex¬
ible, but firms and workers are not equally informed about the price level.
In particular, workers are seen as having imperfect information about the
current price level. They form an expectation of the price level and use this
expectation, along with the known nominal wage, to form an expectation of
the real wage. If the actual price level changes but workers’ expectations of
the price level do not adjust fully, workers are slow to realize the full extent
of the change in their real wages. In this situation, an increase in the actual
price level may lower real wages more than workers realize, allowing firms
to temporarily hire workers at a lower real wage, thus inducing an increase
in employment and output.
In this explanation for the slope of short-run aggregate supply, the extent
of the nominal wage rigidity depends heavily on how expectations are
formed. Friedman presented this argument assuming price expectations are
formed using the adaptive expectations hypothesis we examined in Chapter
12. New classical economists have presented this argument assuming price
expectations are formed using rational expectations, again as we saw in
Short-Run Aggregate Supply 549
Chapter 12. With rational expectations, the deviation of the actual price level
from the expected price level must be unpredictable, and this leads to the
presumption that the short-run aggregate supply curve holds for very short
periods of time.
Determinants of Short-Run Aggregate Supply

The determinants of short-run aggregate supply are the factors that cause it
to shift. As noted earlier, these determinants include the nominal wage rate,
the prices of other inputs, the stock of physical capital available for use in
production, and the state of technology. It may also include the expected
price level. We now look at how each of these factors affects the position
of the aggregate supply curve.
Nominal Wages. The nominal wage (wL) is the payment to labor, and
labor is the major input for most firms. Holding other things constant (in¬
cluding the price level), an increase in the nominal wage rate raises the cost
of production and leads firms to cut back on the amounts they are willing to
produce at each price level. Graphically, the reduction in short-run aggregate
supply induced by an increase in nominal wages is depicted by a leftward
shift in the short-run aggregate supply curve. A reduction in the wage rate
has the opposite effect: It increases short-run aggregate supply (shifts the
curve to the right).
Prices of Other Inputs. Changes in other input prices, including the

prices of raw material inputs such as iron ore and bauxite, and the prices of
energy inputs, such as the price of electricity per kilowatt hour or the price
of natural gas, also affect short-run aggregate supply. Increases in the prices
of these inputs affect short-run aggregate supply in the same manner as an
increase in the wage rate. Thus, increases in input prices (Z) lead to a
reduction in short-run aggregate supply, shown by a leftward shift from
SRAS to SRAS2 in Figure 16.9. Decreases in the prices of inputs lead to an
increase in short-run aggregate supply.
Taxes on Labor. Taxes on labor, which include the income tax and
social security taxes, can also affect short-run aggregate supply. Social se¬
curity taxes, for example, are paid by both employers and employees. An
increase in this tax makes labor more expensive to firms and also lowers the
aftertax wage workers earn. Consequently, an increase in such taxes reduces
the quantity of labor demanded by firms and may also decrease workers’
willingness to provide labor services. (In the extreme case of 100 percent
taxation, it is clear that few, if any, will choose to work.) For these reasons,
increases in the tax rate on labor (TL) reduce the amount of labor used in
production, which in turn shifts the short-run aggregate supply curve to the
left. Reducing tax rates on labor has the opposite effect: It shifts the short-
run aggregate supply curve to the right.
Stock of Physical Capital. The economy’s physical capital stock (K)

is a variable that is fixed in the short run, because it takes time to add new
structures or machinery to the nation’s productive capital stock. Nonetheless,
the capital stock is an important determinant of short-run aggregate supply.
A greater capital stock increases the productive capacity of the economy and
leads to higher output at every price level. An increase in the capital stock
also exerts an indirect effect on output through its effect on labor. In partic¬
ular, a rise in the capital stock increases labor demand because the additional
capital makes any given quantity of labor more productive. This causes an
increase in the demand for labor, and the increase in the equilibrium level
of employment also contributes to higher output at every price level. Thus,
an increase in the capital stock causes the short-run aggregate supply curve
to shift to the right. A decrease in the capital stock works in the opposite
direction, leading to a reduction in short-run aggregate supply.
The state of technology (Tech) is the method by which a

firm combines inputs of labor, capital, raw materials, and energy to produce
output. Improvements in technology occur when a firm can take a given
amount of inputs and produce more output than previously. For example,
the recent widespread adoption of personal computers is a technological
advance that allows substitution of the inexpensive and flexible computing
power of PCs for expensive and inflexible mainframe or human computing
power. Technological advances allow more output to be produced at each
price level and hence shift short-run aggregate supply to the right.
Summary 1 Explain the impact of the following events on the position of the short-run
Exercise 16.2% aggregate supply curve, (a) A war in the Persian Gulf breaks out, drastically
increasing the price of crude oil. (b) Congress raises the social security tax
to 20 percent to finance the social security trust fund in the wake of an aging
population.
(a) The rise in the price of crude oil—a raw material used to
produce gasoline—decreases short-run aggregate supply (shifts it to the left).
Changes in the overall price level also may occur, but they lead to a move¬
ment along the short-run aggregate supply curve, (b) An increase in the
social security tax decreases short-run aggregate supply.
Short-Run Macroeconomic Equilibrium
Our next step is to describe short-run macroeconomic equilibrium—the

determination of the equilibrium interest rate, price level, and level of real
GDP for the economy. As we saw in Chapter 15, the equilibrium interest
rate can be analyzed using either money supply and money demand in the
Short-Run Macroeconomic Equilibrium 551
money market or the loanable funds market developed in Chapter 4.3 Figure
16.10 shows the equilibrium interest rate determined in the money market,
i * as the nominal interest rate that equilibrates money supply and money
demand. For a given exogenous expectation of inflation, ire, this nominal
rate implies a real interest rate, r* such that r* = /* — i;e. Thus, the
nominal rate determined in the money market minus the expected inflation
rate gives us the real rate of interest, r *
The real interest rate, in turn, determines the position of both the con¬
sumption and investment demand curves, as we discussed earlier. Given the
real interest rate determined in the money market, we can graph aggregate
demand (which depends on the real interest rate) and short-run aggregate
supply as in Figure 16.11. The short-run equilibrium price level and the
level of real GDP are determined in the goods market by the intersection of
the aggregate demand and short-run aggregate supply curves. Equilibrium
3 In Chapter 15, we saw that these approaches are equivalent but there are advantages to using the
money market to discuss the effect of changes in the money supply while using the loanable funds
approach to discuss the effects of other changes, such as changes in the government deficit.
552 Chapter 16 A S i m p l e M ac r o e co n o m / c M o d e l
Figure 16.11
Short-Run Macroeconomic Equilibrium
Short-run equilibrium real GDP is determined by the intersection of short-run aggregate

supply and aggregate demand at point /A. The equilibrium price level and real GDP are P0
and Y0l respectively. Notice that at a price level higher than P0l such as P2, there is excess
supply and inventories build up. The excess supply will be eliminated only if the price level
falls, which will reduce the quantity supplied while increasing the quantity demanded for
reasons described in the text. Similarly, if the price level is lower than P0, such as Pn, there
is excess demand for goods and services. In this case the price level rises, stimulating the
quantity supplied while reducing the quantity demanded until the price rises enough that
short-run aggregate supply equals aggregate demand.
is at point A in Figure 16.11, so the equilibrium price level is P0 and

equilibrium real GDP is y0-
To summarize, the equilibrium nominal interest rate is determined in the
money market at the intersection of the supply and demand for real money
balances. This implies a real interest rate, and the real interest rate in turn
determines consumption demand and investment demand and hence the
position of aggregate demand. Thus, equilibrium in the money market is an
important determinant of the position of the aggregate demand curve in the
goods market. The short-run equilibrium price level and real GDP are de¬
termined in the goods market by the intersection of the AD and SRAS curves.
Achieving Equilibrium
How does the economy get to equilibrium at point A in Figure 16.11? To
answer this, suppose the price level is below the equilibrium price level, say,
at P{. When the price level is Px, the aggregate quantity supplied is only Si,
while the aggregate quantity demanded is D{. There is an excess demand
for total output of Dx — Si. Some component of aggregate demand (either
consumption demand, investment demand, or government demand) is not
being satisfied at this price level; consumers, investors, and government
desire more goods at this price level than there are goods available. This
excess demand exerts upward pressure on prices in the economy until the
price level eventually reaches P0. Notice that as the price level begins to rise
above Pu aggregate quantity supplied increases and aggregate quantity de¬
manded decreases. But as long as the price level is below P0, an excess
demand will remain (and hence pressure on the price level to move toward
P0 will remain). Once the price level reaches P0, the upward pressure on the
price level is eliminated as aggregate quantity supplied equals aggregate
quantity demanded.
Had the price level started at a higher level—say, P2—there would have
been an excess supply of output. In the aggregate, firms would begin to
accumulate larger inventories than they wanted. They would begin to lower
prices, leading to a lower overall price level. As the price level declined,
inventories would begin to shrink. Once the price level reached P0, suppliers
would have no further incentive to lower prices, and the price level would
remain at P0.
Changes in Equilibrium
How do changes in aggregate demand or short-run aggregate supply affect
the equilibrium price level and real GDP? To answer this question, we look
at changes first in aggregate demand and then in short-run aggregate supply.
Changes in Aggregate Demand. During the first part of the 1980s,

the U.S. government embarked on an extended program of expanded defense
purchases. Federal government purchases for defense rose from $143 billion
in 1980 to $259 billion in 1985, an increase of 81 percent in nominal terms
and 37 percent after inflation. This increase was substantial even when
measured against total GDP. In 1980 federal defense purchases were 5.3
percent of GDP, and by 1985 they were 6.4 percent. What is the effect of
such an increase in defense spending?
To answer this question, we will hold all variables constant except
government purchases. This will show us how an increase in government
purchases affects the economy. However, we will ignore important features
of the increase in government purchases that have to do with how the
spending is financed. The government can increase purchases and thus affect
the economy, but it also must decide how that spending will be financed,
and this too will have an effect on the economy. For now we ignore the
financing decision and take it up later in the chapter.
Figure 16.12 shows an initial equilibrium, where short-run aggregate
supply (SRASq) and aggregate demand (AD0) intersect at point A. At the
initial equilibrium, the price level is P0 and the quantity of output—real
GDP—is Y0. When aggregate demand increases to AD{—as it did due to the
increased defense purchases in the early 1980s—the new equilibrium is at
the intersection of ADX and SRASq at point B. Thus, we see that in the short
run, an increase in aggregate demand will lead to a higher price level (Px)
and a higher level of real GDP (Tj). We conclude that the defense buildup
in the first half of the 1980s tended to raise both real GDP and the price
level.
We can state these results more generally as follows. Any change in a
determinant of aggregate demand that shifts AD to the right will have the
short-run effect of increasing the price level and increasing the quantity of
output. Similarly, any change in a determinant of aggregate demand that
shifts AD to the left will have the short-run effect of reducing the price level
and decreasing real GDP.
Changes in Short-Run Aggregate Supply. The summer of 1993

was marked by widespread flooding along the Mississippi and Missouri
rivers and their tributaries. The flooding caused damage to the nation’s
capital stock as houses and factories were damaged or destroyed by floods
and flooded farmland was rendered incapable of producing crops for the
1993 season. Then, in January 1994, Los Angeles and surrounding areas
were struck by a major earthquake that severely damaged houses, factories,
and the cities’ infrastructures of roads, utility lines, and sewer systems. These
events—natural occurrences that damage the nation’s capital stock and abil¬
ity to produce output—are one example of a negative supply shock, an event
that shifts short run aggregate supply to the left. What is the impact of such
a supply shock on real GDP and the price level?
To answer this question we turn to Figure 16.13, which illustrates an
initial equilibrium between short-run aggregate supply (SRAS0) and aggre¬
gate demand (AD0) at point A. The corresponding price level is P0, and the
quantity of output—real GDP—is Y0. This situation represents the economy
before the supply shock. The supply shock is a change in one of the deter¬
minants of short-run aggregate supply that causes it to decrease to, say,
A decrease in short- Figure 16.13

run aggregate supply Effect of a Decrease in Aggregate Supply on
from SRAS0 to SRASi Short-Run Macroeconomic Equilibrium
causes the equilibrium
to change from point
A to point B. Note the
new equilibrium is at a
higher price level and
a lower level of real
GDP.
SRASX. The new equilibrium between SRAS{ and AD0 is at point B, with a
higher price level of P{ and lower real GDP of Yx. Thus, a decrease in short-
run aggregate supply results in a higher price level and lower equilibrium
real GDP. Our model predicts that the flood of 1993 and the Los Angeles
earthquake of January 1994 caused real GDP to fall and the price level to
rise relative to what they otherwise would have been.
More generally, any change in a determinant of short-run aggregate
supply that decreases SRAS—that is, shifts short-run aggregate supply to
the left—will have the short-run effect of increasing the price level and
decreasing real GDP. Any change in a determinant of short-run aggregate
supply that causes it to increase—shifts short-run aggregate supply to the
right—will decrease the price level and increase real GDP.
Summary (a) Suppose the nominal equilibrium interest rate falls. What happens to the
Exercise 16.3 short-run equilibrium price level and real GDP assuming all other determi¬
nants of AD and SRAS remain the same? (b) The computer industry is
constantly improving the speed and power of personal computers. Intel’s
486 and Pentium processors are able to outperform the original 8088 chips
of a decade ago by orders of magnitude. Holding other things constant, what
is the impact of this technological advance on the equilibrium price level
and the level of real GDP?
Answer: (a) Holding other things constant, a reduction in the nominal

interest rate leads to a lower real interest rate. This increases investment and
consumption demand, which shifts the aggregate demand curve to the right
as in Figure 16.12. The reduction in the interest rate thus leads to a higher
price level and higher real GDP in the short run. (b) Technological advances
in general shift aggregate supply to the right. The technological advances in
the computer industry shift short-run aggregate supply to the right. Holding
everything else constant, including the determinants of aggregate demand,
the result is an increase in real output and a lower price level than would
otherwise be the case.
Vertical Long-Run Aggregate Supply

and Long-Run Equilibrium
So far we have discussed short-run aggregate supply, a situation in which

nominal wages and other input prices are fixed and price expectations have
not adjusted to fully account for any changes in the price level. In the long
run, however, nominal wages typically adjust to changes in the price level,
as do the prices of other inputs, and price expectations adjust to match
movements in the price level. When wages, other input prices, and price
expectations rise in response to a higher price level, the short-run aggregate

supply curve shifts to the left, eliminating some of the increase in output
that occurs when the price level increases.
To see this, we will focus on the adjustment in nominal wages. For
simplicity we will assume other input prices and price expectations adjust
instantaneously, so only nominal wages are slow to adjust to price changes.
Consider part a of Figure 16.14, which shows an initial short-run aggregate
supply curve as SRAS0(wL = 5) for a given value of the wage rate, wL =
$5. If the price level is 100, real GDP is $4 trillion, as illustrated at point A.
Figure 16.14 : .
The Long-Run Aggregate Supply Curve
The short-run aggregate supply curve is drawn with input prices constant. In part a the initial situation is
at point A with a price level of 100. As the price level rises to 120, short-run aggregate quantity supplied
rises to point B. Notice that initially wages were 5 percent of the price level. At point B, wages are only
about 4 percent of the price level. This fall in the relative input costs led to the increase in output.
However, over time wages will adjust to give workers the same purchasing power they had before the
price change; that is, wages will equal 5 percent of the price level. At a price level of 120, a wage of $6
will equal 5 percent of the price level, which shifts the short-run aggregate supply curve to the left.
Part b shows the adjustment between the short-run and long-run equilibria. The economy is initially
in equilibrium at point A. Aggregate demand increases from AD0 to AD1; moving the short-run equilib¬
rium to point B. In the long run wages will rise, thereby shifting back the short-run aggregate supply
curve. This process will continue until the purchasing power of wages has been restored to its earlier
level at point A. This occurs at point D, with wages equal to $6, a price level equal to 120, and output at
the long-run level of $4 trillion in 1987 dollars; that is, the increase in aggregate demand is ultimately
offset by a decrease in short-run aggregate supply. The vertical line through points A and D, denoted
LRAS, is the long-run aggregate supply curve.
Price
Level
(P)
120
(a)
100
4 5 4 4.4
Real GDP (Trillions of 1987 $) Real GDP (Trillions of 1987 $)

What happens when the price level increases to 120? In the short run (when
the wage remains at $5), the quantity supplied increases as we move up the
aggregate supply curve to point B, where real GDP is $5 trillion. This
movement from A to B is the increase in quantity supplied, holding constant
the nominal wage rate.
In the long run, workers observe the higher price level and must receive
higher wages if the purchasing power of their wages is to remain the same.
When wages increase, the short-run aggregate supply curve decreases, or
shifts to the left. If the price level is still at 120, the increase in wages will
shift short-run aggregate supply leftward. How much will wages rise? For
the purchasing power of wages to remain constant, wages must rise propor¬
tionally with prices. Since the price level increased from 100 to 120, or 20
percent, wages must increase by 20 percent. Thus, wages would increase
from $5 to $6, shifting the short-run aggregate supply curve to SRASF(WL
= 6).
Once wages increase to $6, we are at point C in part a of Figure 16.14,
and the wage rate stands in the same proportion to the price level as at point
A; that is, at point A wages were $5 when the price level was 100, and now
at point C wages are $6 when the price level is 120. The ratio of wages to
prices is unchanged. Thus, workers now cost the same relative to output
prices as they did at point A. Consequently, firms will hire only as many
workers as they hired in the initial situation, and output will be the same as
at point A. In other words, the increase in the wage rate to $6 shifts short-
run aggregate supply to SRASF(wL = 6). At a price level of 120, this indicates
real GDP will again be $4 trillion (point C). Thus, in the long run, after all
adjustments in input prices, the change in output is zero.
Another way to see this, and to see just how the adjustment between the
long run and the short run takes place, is illustrated in part b of Figure 16.14.
The initial equilibrium is at point A, where AD0 and SRAS0(wL = 5) intersect.
The price level is 100, and real GDP is $4 trillion. From this starting point,
consider an increase in aggregate demand to ADX. In the short run, with fixed
wages and input prices, the equilibrium moves to point B, where the price
level is 110 and real GDP is $4.4 trillion. This is the short-run impact of the
increase in aggregate demand.
At point B, the price level has increased but wages remain fixed. In
moving from A to B, prices rise from 100 to 110, or by 10 percent. Workers
will ultimately require a 10 percent increase in wages to keep their purchas¬
ing power constant, and the wage will rise to $5.5. This increase in wages
shifts aggregate supply to SRAS{(wL = 5.5). If the price level stayed at 110,
this would reduce real GDP back to the initial equilibrium value, 4. However,
this leftward shift in aggregate supply, due to higher wages, results in an
equilibrium at point C, where the price level has risen further to 115. At
point C, the price level is 15 percent above the initial price level at point A,
but wages and other input prices are only 10 percent above their initial
values. Wages have still fallen in real terms, though not by as much as they
did at point B. At point C real GDP is still above its initial level of $4
trillion, but is now below the level it was at point B, $4.4 trillion.
Once again workers observe their wages have not kept up with the price
level, and they must receive higher wages to maintain their purchasing
power. This again shifts aggregate supply to the left. This process continues
until aggregate supply shifts to SRASF(wL = 6), where it intersects ADX at
point D. In the long run, real GDP is $4 trillion, the same output level as at
point A. Points A and D, and all points on the line labeled LRAS, are called
the long-run aggregate supply curve to indicate that changes in the price
level do not affect real GDP when wages and other input prices are not fixed.
Thus, we see that increases in aggregate demand can raise output in the
short run provided wages, price expectations, and other input prices do not
adjust to increases in the price level. But in the long run, wages, price
expectations, and other input prices increase with the price level, and real
GDP will fall to its long-run equilibrium level. Of course, as the adjustment
process illustrated in part b of Figure 16.14 indicates, the adjustment to the
long-run equilibrium may take a considerable amount of time, depending on
how quickly nominal wages adjust at each step. Also, although Figure 16.14
focuses on nominal wages, input prices and price expectations must also
adjust.
Keynesian economists and others who stress nominal wage rigidity—
including many so-called new Keynesians who stress the infrequent adjust¬
ment of nominal wages in labor contracts—tend to stress that this adjustment
process (the movement from point B to D in part b of Figure 16.14) can be
very slow. Keynes himself spoke disparagingly of economists who looked
only at the long run and is reported to have said that in the long run, we are
all dead. Other economists—notably monetarists (who stress the role of
money in economic activity) and new classical economists (who stress
rational expectations and market clearing)—are much more comfortable with
long-run analysis and think wages adjust fairly quickly to get the economy
to long-run equilibrium. New classical economists in particular often stress
that only unexpected increases in the price level cause differences between
short-run and long-run aggregate supply and that with rational expectations,
these differences should be both unpredictable and temporary. Thus, new
classical economists tend to be very comfortable with quick adjustment to
the long run.
Does the preceding analysis mean real GDP cannot change in the long
run? Not at all! It means real GDP does not change in the long run due to
aggregate demand factors. The output level indicated by the long-run aggre¬
gate supply curve is also called the natural rate of output, the amount of
goods and services the economy can produce with available technology and
the full employment of its resources, including capital and labor. In the long
run, firms can produce more only if the capital stock or the size of the work
force increases or technological advances occur. For instance, increases in
the capital stock shift long-run aggregate supply to the right, as does any
560 Cha pter16 A Simple Macroeconomic Model
improvement in technology. Lowering the tax rate on labor also shifts long-
run aggregate supply to the right, since it encourages more people to enter
the work force. Similarly, decreases in the capital stock, technological re¬
gress, or increasing the tax rate on labor all shift long-run aggregate supply
to the left.
When economists talk about full employment, they do not mean a situ¬
ation in which the unemployment rate is literally zero. Some unemployment
is frictional unemployment, which occurs as people seek their first jobs,
quit one job to seek another, or are displaced from jobs due to technological
changes and must seek other jobs. If only frictional and structural unem¬
ployment exist, we say the economy is in a state of full employment. Another
way to say this is that the economy is at the natural rate of employment,
the level of employment that exists when there is only frictional unemploy¬
ment. The natural rate of output is the level of output that occurs when
employment is at the natural rate of employment so that only frictional
unemployment exists. Thus, the level of output indicated by the position of
the long-run aggregate supply curve is the natural rate of output.
More generally, the vertical long-run aggregate supply curve means that
in the long run, economic growth—increases in the real output produced
by an economy—requires a rightward shift in long-run aggregate supply.
This fact has important implications for policy. For example, policies that
directly encourage investment, or that encourage saving and indirectly en¬
courage investment, will lead to faster growth of the capital stock and thus
of long-run output. Policies that encourage technological change, such as
subsidies or tax breaks for research and development activities, can also
increase long-run aggregate supply. Finally, policies that encourage greater
labor force participation by workers and higher labor demand by firms will
lead to higher levels of employment and hence increase long-run aggregate
supply. Such policies might be reductions in the tax rates on labor income
to encourage an increase in labor supply or reductions in taxes paid by
employers hiring labor, such as social security taxes.
Summary Prior to elections, government often pursues policies that stimulate the econ¬
Exercise 16.4 omy, such as increasing government purchases or increasing the money
supply. For example, it is claimed that in the United States, real output tends
to rise just before presidential elections and to decline after those elections.
Some analysts claim this is due to expansionary policy actions undertaken
by incumbent presidents who are trying to get reelected or to get their parties’
nominees elected. This phenomenon has been called the political business
cycle. Describe how our model of aggregate demand and supply can explain
how this political business cycle works.
In the short run, when wages are fixed, expansionary government

policies increase aggregate demand, thus leading to greater real GDP (like
Fiscal Policy and the Government Budget Constraint 561
the movement from A to B in part b of Figure 16.14). If the government

times the expansionary policy correctly, this should occur right before the
election. This would explain why the economy flourishes just prior to an
election: Incumbents stimulate the economy to increase real GDP, thus
enhancing their chances for reelection. In the long run, however (sometime
after the election is over), wages and other input prices begin to rise, which
shifts short-run aggregate supply to the left. This corresponds to a movement
from B to D in part b of Figure 16.14. As we move from B to D, real output
falls until real GDP returns to the natural rate.
Fiscal Policy and the Government Budget Constraint
Summary Exercise 16.4 considered the impact of an increase in government

purchases on short-run and long-run equilibrium. To keep the analysis sim¬
ple, we didn’t consider where the government obtained funds for the in¬
creased spending. In reality, to pay for additional spending the government
must either raise taxes, borrow, or print new money. We will see that the
way government chooses to finance its spending has important consequences
for the economy. To understand the workings of fiscal policy—government
spending or tax policy—we first look at the concept of the government
budget constraint.
What Is the Government Budget Constraint?

The government budget constraint is the restriction that the government
cannot purchase goods and services without having a source of funds to pay
for these purchases. Of course, you are probably thinking the federal gov¬
ernment continually purchases goods and services without immediately pay¬
ing for them with tax revenue, and you are partially correct. However, the
government does have to obtain funds to pay for programs.
One way it does so is to borrow funds in the loanable funds market,
much as you might use a credit card to borrow funds for your expenses. The
fact that you have a credit card doesn’t mean you get to buy goods and
services without paying for them, just as the government’s ability to borrow
doesn’t mean it obtains goods and services without paying for them. How¬
ever, the government has two ways to finance current spending that are not
available to you: It can collect funds through taxes, and it can print money
to pay for its purchases. Thus, the basic idea behind the government budget
constraint is that to pay for increased spending, the government must either
(1) raise tax revenues, (2) borrow funds by selling government bonds, or (3)
print money. We will see next that when the government raises taxes or
borrows funds to pay for spending increases, it dampens the short-run stim¬
ulating effects of higher government purchases on the economy.
Government Purchases Financed by Taxes

If an increase in government purchases is matched by an equal increase in
taxes to pay for the spending, the government does not need to borrow any
additional funds or print additional new money. We can analyze this case as
a simultaneous increase in government purchases and in taxes. We first
consider taxes that affect only aggregate demand and then look at taxes that
affect both demand and supply.
Consumption Haxes. Periodically there is talk in Washington about

using a consumption tax—in particular, a value-added tax, or VAT—to raise
additional tax revenue. For example, this idea was floated in the first year
of President Clinton’s term during the discussions surrounding his budget
proposal. How would such a tax affect the economy? While this is a very
complicated issue, we can use our aggregate demand and supply model to
get some idea of the effect on real GDP and the price level.
Point A in Figure 16.15 illustrates the initial equilibrium in the economy.
Suppose the government raises consumption taxes to finance an increase in
government purchases of goods and services. Such a tax does not affect
workers’ take-home pay, so to keep things simple let us suppose it does not
affect the position of the aggregate supply curve.4 The increase in govern¬
ment purchases of goods and services raises aggregate demand from AD0 to
ADX. But the rise in consumption taxes discourages consumption, which
tends to shift aggregate demand back to the left. Does the increase in taxes
totally cancel the increase in government purchases so that the net effect on
aggregate demand is zero? Most economists say the answer is no. Every $1
increase in government purchases causes government demand, and hence
aggregate demand, to shift to the right by $ 1. Every increase in taxes reduces
consumers’ aftertax income and thus reduces consumption demand—but by
less than $ 1, since consumers might pay the $ 1 in taxes partly by reducing
consumption and partly by reducing saving. Hence, the net effect is that the
tax increase only partly counteracts the increase in government purchases,
and the aggregate demand curve ends up in a position such as AD2. Thus,
the net short-run effect is that the equilibrium moves to point C, where the
price level and real GDP are higher than at point A (but lower than at point
B, which ignores the effect of the consumption tax on aggregate demand).
This example illustrates that in the short run, an increase in government
purchases financed by a consumption tax may actually stimulate the econ¬
omy, that is, cause real GDP to rise.
In the long run, however, nominal wages and other input prices will rise
due to the increase in the price level. This causes short-run aggregate supply
4 In reality, all taxes are distortionary and thus are likely to affect aggregate supply. Even a lump¬
sum head tax reduces the incentive to have children, which will decrease the size of the work force
and decrease aggregate supply.
Figure 16.15
Tax-Financed Government Purchases That Do Not
Affect Long-Run Aggregate Supply
This graph shows the effect of an increase in government purchases financed by an in¬
crease in taxes when the taxes do not affect aggregate supply. The initial equilibrium is at
point A. The increase in government purchases causes aggregate demand to rise from AD0
to ADV The increase in taxes, however, causes aggregate demand to decrease to AD2.
Thus, short-run equilibrium is at point C, where the price level and real GDP are higher
than at point A. In the long run wages rise, which results in a long-run equilibrium at point
E. Thus, in the long run, aggregate output is unaffected by the increased government
purchases financed by taxes, but the price level rises.
to shift to the left and ultimately results in a new long-run equilibrium at

point E. At point E, real GDP is the same as it was before the increase in
government purchases and taxes.
The primary source of federal government tax revenue

is the income tax. What happens if the government raises income taxes to
fund an increase in government purchases of goods and services? The in¬
crease in government purchases tends to raise aggregate demand. But the
higher income taxes tend to reduce aggregate demand, for two reasons. First,
Did Indexation Cause the Budget

Deficits of the 1980s?
It has become popular in some quarters to one's social security check. Now suppose there
blame the indexation of the tax code during the is 20 percent inflation. If your earnings keep up
Reagan tax cuts for the ongoing federal budget with inflation, you now earn $12,000. With in¬
deficits. The argument is that indexing the tax dexed tax brackets, you still pay only 10 percent
brackets to inflation, coupled with indexing tax on this income, which now comes to
spending to inflation, caused the budget deficit $1,200. (Without indexing, you would be
to expand. pushed into a higher tax bracket and pay
Does this argument make sense? It is true $1,400 in taxes—10 percent on the first
that if the tax brackets had not been indexed, $10,000 of income and 20 percent on the next
the U.S. Treasury would have had more revenue $2,000.) Social security is also indexed to infla¬
solely because inflation pushed taxpayers into tion, and the payment to recipients rises from
higher brackets. But the argument is stronger $1,000 to $1,200. Thus, before inflation your
than this. It claims that this indexing means in¬ taxes just paid for the social security payment,
flation caused an increase in the deficit, with and after inflation your taxes just paid for the
spending rising faster than tax collections. Is this social security payment. Thus, indexing of tax
correct? brackets and of spending does not cause a
Consider a very simple example. You earn budget deficit. Indeed, the increase in your tax
$10,000 and pay 10 percent tax. Every dollar bili increased by just enough to cover the in¬
over $10,000 is taxed at a higher rate, say, 20 creased social security spending.
percent. Your tax bill, $ 1,000, pays for some¬
the higher income tax leads to a reduction in consumption, as households

have less to spend on consumer goods. Second, investment declines since
firms will find fewer profitable investments at the higher tax rates.* * * * 5 This
not the end of the story, however. Higher income taxes also affect aggregate
supply. The increase in the income tax reduces people’s incentive to enter
the work force, since they get to keep less of each additional dollar they
earn. This lowers the natural rate of employment, which results in a reduction
in both short-run and long-run aggregate supply.
5 These decreases in consumption and investment may be further reduced if the higher income tax
rates decrease the supply of loanable funds. In this case, interest rates rise, which further reduces
both consumption and investment.
The ultimate magnitude of these effects is a source of much political

debate, and we can see why by using our aggregate supply and demand
framework. In Figure 16.16, the initial equilibrium is at point A at the
intersection of AD0 and SRAS0. The increase in government purchases shifts
aggregate demand to the right to ADX, but the increase in taxes reduces
investment and consumption, which shifts aggregate demand to the left, to
AD2. The rise in the income tax also decreases short-run aggregate supply
to SRASi and decreases long-run aggregate supply to LRASi.
If the rise in government purchases of goods and services more than
offsets the decline in investment and consumption, the short-run equilibrium
will be at point B, where AD2 and SRAS{ intersect. In this case, the “tax and
spend” policy has the short-run effect of increasing real GDP and the price
level compared with the initial situation at point A. However, if the reduction
in investment and consumption exactly offsets the rise in government pur¬
chases, aggregate demand will remain at AD0. In this case, the short-run
equilibrium is at point C, where real GDP is lower than at point A. This
example illustrates that a rise in government purchases financed by an in¬
crease in the income tax may either increase or decrease real GDP in the
short run, depending on how much investment declines. Obviously those
who believe the short-run effect of the “tax and spend” policy is to move
the economy to point B are more likely to favor the policy than those who
view point C as the likely equilibrium. Most income tax legislation has
special provisions, called investment tax credits, that are designed to keep
investment from drastically declining due to higher taxes, thereby increasing
the likelihood that the actual short-run equilibrium will occur at point B.
The long-run effect of government purchases financed with income taxes
is clearer. If taxes reduce long-run aggregate supply, then, regardless of the
short-run effect, the long-run effect is to reduce real GDP and raise the price
level. We can readily see this in Figure 16.16. If long-run aggregate supply
declines from LRASq to LRASi, then, whether the short-run equilibrium is at
B (along AD2) or C (along AD0) is of little importance in the long run. In
long-run equilibrium, the economy will be at either F (along AD2) or E
(along AD0), and in either case real GDP will fall to YLR, since both AD0
and AD2 intersect LRAS\ at the same level of real GDP. Thus, we see that
in the long run, if higher income taxes decrease long-run aggregate supply
by reducing incentives to enter the work force, this reduction in the natural
rate of employment will decrease the natural rate of output of the economy.
Will taxes reduce long-run aggregate supply? Much debate exists over
how much the natural rate of output declines as a result of higher income
taxes. The popular press often uses the label supply sider to describe those
who view the effect on aggregate supply as being important. Some econo¬
mists dismiss the notion that tax rates have much to do with long-run
aggregate supply. They claim long-run supply is mostly unaffected by tax
rate changes of the sort usually proposed for the U.S. economy.
The Tradeoff: Higher Real GDP or Greater Income Equal¬

ity. Our analysis of the long-run effects of tax-financed government pur¬
chases suggests income taxes may have an adverse effect on long-run GDP.
You may wonder why there is any debate at all about abolishing the income
tax and replacing it with other forms of taxation, like a lump-sum or con¬
sumption tax. The answer is surprisingly simple. First, some argue income
taxes have little effect on long-run GDP. Second, and probably equally
important, is the fact that the U.S. income tax serves a dual purpose: to fund
government programs and to decrease the degree of income inequality. Those
who favor the consumption tax tend to stress that it results in higher long-
run real GDP than under an income tax. Proponents of the income tax may
even acknowledge its adverse effects on long-run real GDP but view the
effect as small compared to the benefits of greater income equality.
Since the progressive income tax in the United States taxes the rich at
higher rates than the poor, it tends to reduce the relative difference in incomes
between the classes. Consumption taxes, on the other hand, tend to be
regressive, meaning the poor actually pay a greater fraction of their income
in taxes than the rich do. For the poor who spend all of their incomes on
consumer goods, a consumption tax is an income tax. The rich, who spend
a smaller fraction of their incomes on consumer goods, end up paying taxes
only on the smaller percentage of income actually spent on consumption
goods. For these reasons, consumption taxes tend to increase the level of
income inequality, whereas the progressive income tax tends to reduce it.
Advocates of consumption taxes believe the benefits of higher long-run
real GDP more than outweigh the negative effects on the distribution of
income, whereas advocates of the income tax do not. Which view is “cor¬
rect?” We cannot provide an answer to this question. We can only point out
the tradeoff: The cost of using an income tax to equalize the distribution of
income is lower long-run GDP.
Government Purchases Financed

by Borrowing: Deficit Finance
In the 1980s, the U.S. government ran deficits of huge magnitudes when
judged against its own history of peacetime government budgets, and fi¬
nanced these shortfalls with government bond sales. Did it make any differ¬
ence to the economy that the government chose to finance its spending by
borrowing instead of raising taxes? Two schools of thought exist on this
question. The traditional Keynesian view is that deficit financing is more
expansionary than tax financing, because taxpayers don’t recognize that
government debt is simply deferred tax liabilities and that taxes will someday
be raised to pay for the accumulated debt. Another view, called the Ricar¬
dian view, is that borrowing and tax financing are equivalent. According to
the Ricardian view, our analysis of tax-financed government purchases in
the previous section also applies to deficit-financed government spending.
We will examine the traditional Keynesian analysis here, deferring further
discussion of Ricardian ideas until Chapter 21.
What happens when the government borrows to finance an increase in
purchases of goods and services? In this case, government purchases rise
(increasing aggregate demand), and there is no decline in consumption from
raising taxes. However, government borrowing will increase the interest rate,
which reduces investment and consumption. To see why, consider the loan¬
able funds market introduced in Chapter 4. This market is illustrated in
Figure 16.17, where the initial equilibrium is at point A, with an equilibrium
interest rate of 5 percent. When the government increases its borrowing to
This figure shows how Figure 16.17

a deficit-financed in¬ Effect of Government Borrowing in the Bond Market
crease in government
purchases affects the
market for loanable Interest
funds. When the gov- Rate (%)
ernment borrows
funds, the demand for
loanable funds rises
from D0 to D-The
equilibrium then
moves from points to
point B, and the inter¬
est rate rises from 5 to
7 percent. 7
Do
Quantity of Loanable Funds
finance increased expenditures on goods and services, an increase occurs in

the demand for loanable funds, represented by a shift from D0 to Dx. This
changes the equilibrium to point B and raises the equilibrium interest rate to
7 percent. This example illustrates that government borrowing raises the
nominal interest rate.
However, this is not the end of the story. The increase in the nominal
interest rate is, holding other things constant, a rise in the real interest rate.
With no change in expected inflation, the increase in the real interest rate
represents a rise in the real cost of borrowing funds for investment purposes
or to finance the purchase of consumer durables. This, in turn, tends to
reduce consumption and investment spending and hence reduce aggregate
demand. This effect of government borrowing on interest rates and hence
on nongovernment spending is called crowding out. Due to crowding out,
the increase in government purchases financed by borrowing results in a
smaller increase in aggregate demand than would at first appear to be the
case. The case of partial crowding out is illustrated in Figure 16.18, where
the increase in government purchases shifts aggregate demand from AD0 to
AD{, but the increase in the interest rate from government borrowing reduces
investment and consumption demand and hence decreases aggregate demand

to AD2. In this case, the net short-run effect of an increase in government
purchases financed by borrowing is the movement from point A to point C
in Figure 16.18, and real GDP increases. In the long run the economy moves
to point D, where output returns to its initial level and the price level rises
further.
We note that for a large enough reduction in consumption and investment
due to the higher interest rate, it is possible that the interest rate effect
completely offsets the increase in government purchases—an effect known
as complete crowding out. In this case, government purchases of goods and
services increase aggregate demand to ADX, but the higher interest rates
caused by government borrowing lead to reductions in consumption and
investment that shift it back to AD0. Here both the short-run and long-run
equilibria remain at point A in Figure 16.18. It is important to realize that
crowding out (partial or complete) reduces the impact of deficit-financed
government purchases on short-run GDP.
Differing views exist on the magnitude of this crowding-out phenome¬
non, and this tends to spice up political debates about the merits of using
fiscal policy to stimulate the economy. Keynesians believe crowding out is
not a major detriment to deficit-financed fiscal policy, whereas monetarists
and new classical economists tend to believe crowding out is more nearly
complete. Hence, Keynesians are advocates of fiscal policy, whereas mone¬
tarists and others are not.
One interesting claim made by several influential commentators such as
Edwin Yoder and Sam Donaldson is that indexing the Federal tax code was
a leading cause of the budget deficits of the 1980s. Box 16.3 discussed this
claim (see p. 564).
Government Purchases Financed by Money Creation

Government purchases can also be financed by money creation, and this
approach has very strong short-run effects on macroeconomic equilibrium.
While the U.S. government did not choose to finance government purchases
with large-scale money creation in the 1980s, other countries have at times
resorted to money creation to finance large portions of government pur¬
chases. Historical examples include the German hyperinflation of the 1930s.
A current example is the government in Belgrade, Serbia (once the capital
of Yugoslavia), where government money creation has led to inflation rates
that reached 300 million percent per month in 1994!
What are the effects of financing government purchases with money
creation? First, the increase in government purchases increases aggregate
demand directly. Second, the increase in the money supply itself has two
effects. First, the increase in the nominal money supply increases the real
money supply and tends to lower the interest rate. Holding everything else
constant, including the expected inflation rate, this fall in the interest rate
translates into a decline in the real interest rate, which stimulates investment
spending and consumption spending on durable goods and hence further
increases aggregate demand. Second, the rise in the money supply increases
the wealth of consumers, which directly stimulates consumption demand via
the wealth effect and again increases aggregate demand.
Figure 16.19 illustrates government purchases financed by money crea¬
tion. The initial equilibrium is at point A, with aggregate demand given by
AD0. The increase in government purchases raises aggregate demand to ADX.
The increase in the money supply, then, has two effects. First, it lowers
interest rates and thus stimulates investment spending and consumer spend¬
ing. Second, it increases consumer wealth, which exerts a positive effect on
Using Aggregate Demand and Supply:

The Recession of 1990-1991
The U.S. economy experienced negative real est rates responded as expected. The interest
GDP growth—a recession—during the third and rate on three-month T-bills was 6.7 percent in
fourth quarters of 1990 and the first quarter of 1988, and with an inflation rate of 3.9 percent,
1991. Real GDP declined from $4,878 billion (in this was a 2.8 percent ex post real interest rate.
1987 dollars) in 1990 to $4,821 billion in 1991. The slow growth in the monetary base in 1989
What caused this recession? Several answers had its intended effects: the three-month T-bill
have been offered. rate rose to 8.1 percent in 1989, and with an
First, monetary policy was somewhat restric¬ inflation rate of 4.4 percent, this was a 3.7 per¬
tive in 1989, and this restricted growth in aggre¬ cent ex post real interest rate. Thus, the Fed
gate demand. The monetary base grew at 7.1 took actions to restrict money growth and in¬
percent in 1988 but slowed to 4.2 percent in crease interest rates in 1989, and our aggregate
1989. The Ml measure of the money supply demand and supply model suggests aggregate
grew by only 1 percent in 1989 after growing demand would grow more slowly because of
by 5 percent in 1988. M2 still grew at 5.1 per¬ these measures. You may wonder why a restric¬
cent in 1989, down slightly from 5.5 percent in tive monetary policy in 1989 would cause a re-
1988, but slowed to 3.5 percent in 1990. Inter¬ Continued on p. 572
Continued from p. 571 of crude oil skyrocketed. From the spring to fall
cession in the last half of 1990. The reason is of 1990, crude oil prices doubled. This increase
that it takes time for monetary policy actions to in the price of an important input to production
affect the economy. These lags are not incor¬ caused aggregate supply to shift to the left.
porated in our model, but students of monetary What was the effect of all this? In the ac¬
history have observed repeatedly that it seems companying figure, AD90 and SRAS90 intersect
to take about 12 months or more for monetary at a real GDP of $4,878 billion (in 1987 dollars)
policy to have its desired effects. We discuss lags and a price level of 113. In 1991, aggregate de¬
in policymaking in more detail in Chapter 19. mand increased, but only to AD9]l and aggre¬
The second restrictive action was the reduc¬ gate supply shifted left to 5/?A591. The intersec¬
tion in federal government purchases for na¬ tion occurs at a real GDP of $4,821 billion and a
tional defense. With the winding down of the price level of 118. Had monetary policy and gov¬
Cold War, these purchases actually declined be¬ ernment purchases been less restrictive, aggre¬
tween 1988 and 1990 from $287 billion (in gate demand might have grown to ADNR (where
1987 dollars) to $283 billion. Total federal gov¬ NR stands for "no recession"). If oil prices had
ernment purchases (defense plus nondefense) not doubled, aggregate supply might have
also declined in 1989, from $377 billion to $376 shifted only to SRASnr. Had these shifts in¬
billion, before rising slightly in 1990. This reduc¬ creased real GDP by its postwar average growth
tion in federal government purchases, G, also rate of about 2.7 percent, real GDP in 1991
restricted growth in aggregate demand. might have been $5,010 billion. Instead, 1990—
Finally, aggregate supply also played a role. 1991 was a period marked by a recession—one
When Iraq invaded Kuwait and at least implicitly with widespread causes.
threatened Saudi Arabia in early 1990, the price
consumer spending. Both effects tend to increase aggregate demand further,

to AD2 in the figure. Thus, the short-run effect of government purchases
financed by money creation is the movement from point A to point C, which
implies a higher price level (Px) and greater real GDP. Monetarists view the
shift from B to C as relatively large, whereas Keynesians view it as rather
small. The movement from A to C illustrates that government purchases
financed by an increase in the money supply has a strong stimulating effect
on real GDP in the short run. Note, however, that the price level also rises
considerably in the short run due to the increase in the money supply.
Of course, only in the short run is aggregate supply responsive to changes
in the price level. In the long run, the increase in aggregate demand will
result in a higher price level but no change in real GDP. In Figure 16.19,
for example, the long-run impact of the increase in aggregate demand to
AD2 will be to change the equilibrium to point D, where the price level has
risen to P2 but real GDP has returned to Y0. Thus, the long-run effect of
paying for government purchases by increasing the money supply is to raise
prices. Moreover, since the short-run effect is a large shift in aggregate
demand, the long-run effect is a large increase in the price level.
An increase in govern¬ Figure 16.19

ment purchases of Government Purchases Financed by Money Creation
goods and services in¬
creases aggregate de¬
mand from AD0 to
AD-[. The increase in
the money supply that
financed the increase
in government pur¬
chases leads to lower
interest rates, which
further raises aggre¬
gate demand from
AD^ to AD2. Thus, the
short-run equilibrium
moves from point A to
point C. The long-run
result is an increase in
the price level but no
change in output
(point D).
Box 16.4 shows how our aggregate demand and supply model can be
used to explain the U.S. recession of 1990-1991.
Monetary Policy: Open Market Operations
In the previous section, we saw that government spending may be financed

by increases in the money supply. Can the government change the money
supply without changing taxes or spending? In other words, is it possible
for government to pursue an independent monetary policyl The answer is
yes; it can do so through open market operations. In this section, we examine
the effect of an expansionary monetary policy conducted via an open market
operation and how this policy relates to the government budget constraint.
Recall from Chapters 13 and 14 that an open market operation is the
sale or purchase of government bonds by the Federal Reserve System. Con¬
sider an open market purchase. The Fed prints money and uses this money
to purchase government bonds. Thus, the public now holds more money but
fewer bonds, and in fact the increase in money holdings is equal in value to
When the Fed pur¬ Figure 16.20

chases bonds, it in¬ Effect of an Open Market Purchase on the Interest Rate
creases the monetary
base. Thus, the supply
of money increases
from Mq/P to M\/P.
The equilibrium in the
money market moves
from point /\ to point
B. The result is a de¬
crease in the equilib¬
rium interest rate from
/0 to q.
the reduction in bond holdings; thus, the wealth of the public is unchanged.
How does this influence aggregate demand? As Figure 16.20 illustrates,
holding other things constant the increase in the money supply reduces the
interest rate.6 Here the initial equilibrium is at point A, with an equilibrium
interest rate of i0. The increase in the supply of money from Ms0 to M\
changes the equilibrium to point B, where the interest rate falls to q. Thus,
the effect of an open market purchase is that the Federal Reserve System
increases the money supply, lowering the equilibrium interest rate.
The effect of open market operations on aggregate demand, then, follows
from the change in the interest rate. Holding the expected inflation rate
constant, the decrease in the nominal interest rate is a decrease in the real
interest rate and leads to a rise in both consumption and investment demand
and hence in aggregate demand. Figure 16.21 illustrates this effect. Aggre¬
gate demand is initially at AD0, and the initial equilibrium is at point A. The
6 We could also see this effect in the loanable funds market, where the Fed in effect acts as a
lender, increasing the supply of loanable funds and reducing the interest rate.
open market purchase lowers the interest rate, shifting aggregate demand to
ADi and moving the equilibrium to point B in the short run, increasing the
price level to Px and the level of real GDP to Yx. In the long run, however,
the economy moves to point C, where the price level rises further to P2 and
real GDP return^ to its natural rate, Fo¬
llow does this open market purchase relate to the government budget
constraint? It turns out that an open market operation is in perfect accord
with the government budget constraint. The government budget constraint
merely restricts the change in government spending to equal changes in tax
revenue, plus changes in bonds, plus changes in money. If government
spending and tax revenue are unchanged, as they are in an open market
operation, the government budget constraint merely says that any increase
in bonds must be met by a reduction in the money supply, and vice versa.
But that is exactly what happens in an open market operation: The stock of
bonds changes by an equal but opposite amount from the stock of money.
Thus, an open market operation is a monetary policy action that takes place
with no change in government spending or tax collection.
Summary (a) Suppose the government increases purchases of goods and services and
Exercise 16.5 finances the spending by issuing bonds. What will be the short-run and long-
run effects of this policy? (b) What will be the effect of this policy if the
government funds the increase in purchases by raising the money supply?
Answer: (a) Government borrowing—deficit finance—leads to an in¬

crease in the demand for loanable funds, which in turn raises the interest
rate. This increase in the interest rate decreases the level of investment and
consumption in the economy, which tends to shift aggregate demand to the
left. The increase in government purchases shifts aggregate demand to the
right. If there is complete crowding out, these two effects perfectly cancel
each other out, and no change in the price level or aggregate output occurs,
although the interest rate does rise. If only partial crowding out exists, the
increase in government purchases will more than offset the reduced con¬
sumption and investment, and aggregate demand will increase. In the short
run, this will lead to greater output and a higher price level. In the long run,
however, wages and other input prices will adjust, which will shift short-run
aggregate supply to the left. In the long run, the policy will have no effect
on real output; it will only raise the price level, (b) An increase in the money
I supply decreases interest rates, which for given inflationary expectations

leads to greater investment and consumption demand. Likewise, the increase
in government purchases raises aggregate demand. Thus, the short-run effect
of this policy is to increase the price level and real GDP. In the long run,
wages and other input prices rise, leading to a higher price level and returning
real GDP to its natural rate.
Conclusion
In this chapter, we looked at a simple macroeconomic model that links

together the determination of interest rates, the price level, and real output.
Interest rates are determined in the money (or loanable funds) market, while
real GDP and the price level are determined by aggregate demand and
supply. We saw that government purchases of goods and services and mone¬
tary policy can stimulate the economy in the short run, but these policies do
not have long-term effects on real GDP. Keynesians tend to advocate fiscal
policy as a means of stimulating the economy in the short run, whereas
monetarists advocate monetary policy. In the next chapter, we turn to a
discussion of the open economy macroeconomic model and see how inter¬
national trade affects the domestic economy.
KEY TERMS
closed economic model new classical economists

consumption goods natural rate of output
KEY TERMS continued

investment goods frictional unemployment
physical capital natural rate of employment
government purchases economic growth
transfer payments fiscal policy
net exports government budget constraint
aggregate demand curve Keynesian view
substitution effect Ricardian equivalence
income effect crowding out
short-run aggregate supply (SRAS) partial crowding out
nominal wage rigidity complete crowding out
short-run macroeconomic equilibrium independent monetary policy
long-run aggregate supply curve

1. Show how the aggregate demand curve is (d) . Increased consumer optimism
derived from the investment demand, con¬ (e) . Technological improvements
sumption demand, and government de¬ (f) . A tax on labor income
mand curves. (g) * A decrease in government purchases
2. Explain why investment demand is 5. Holding other things constant, determine

graphed as downward sloping in Figure the impact (if any) of the following on in¬
16.4. vestment demand.
(a) . An increase in the real interest rate
3. Holding other things constant, determine (b) . A decrease in the money supply
the impact (if any) of the following on (c) . The elimination of investment tax
consumption demand. credits
(a) . An increase in the interest rate (d) . An income tax
(b) . An increase in household taxes
(c) . Increased consumer optimism 6. Determine the impact of the following on
(d) . Technological improvements aggregate demand.
(a) . An increase in the real interest rate
4. Holding other things constant, determine (b) . A decrease in the money supply
the impact (if any) of the following on (c) . An increase in household taxes
government demand. (d) . An investment tax credit
(a) . An increase in the interest rate
(b) . A decrease in the money supply 7. Holding everything else constant, deter¬
(c) . An increase in household taxes mine the short-run impact of the following
on the equilibrium price level and real ing borrowing. What are the short-run and
GDP. long-run macroeconomic effects on the in¬
(a) . An increase in the interest rate terest rate, the price level, and real GDP?
(b) . A decrease in the money supply
12. Suppose the government reduces spending
(c) . An increase in the income tax
and finances the spending cuts by printing
(d) . Technological improvements
less money. What are the short-run and
8. Explain why the short-run aggregate sup¬ long-run macroeconomic effects on the in¬
ply curve slopes upward. terest rate, the price level, and real GDP?
9. Explain why the long-run aggregate sup¬ 13. Explain why open market operations can
ply curve is vertical. have a short-run effect on the economy
even if government spending remains
10. Suppose the government cuts spending
fixed.
and finances the spending cuts by lower¬
ing taxes. What are the short-run and 14. Explain how the possibility of complete
long-run macroeconomic effects assuming crowding out and partial crowding out
(a) the tax solely affects consumption, gives rise to potential disagreements
(b) the tax solely affects aggregate supply, among policymakers about the desirability
and (c) the tax affects both aggregate de¬ of using deficit-financed government pur¬
mand and aggregate supply? chases to stimulate the economy.
11. Suppose the government cuts spending 15. How can real GDP be increased in the
and finances the spending cuts by reduc¬ long run?
Selections for Further reading
Allen, S. D., J. M. Sulock, and W. A. Sabo. “The Economic Models of Direct Foreign Investment:
Political Business Cycle: How Significant?” Pub¬ Toward a Synthesis.” Hitotsubashi Journal of Ec¬
lic Finance Quarterly, 14 (January 1986), 107- onomics, 25 (June 1984), 1-20.
112. Koray, F., and P. Chan. “Government Spending and
Beard, T. R., and W. D. McMillin. “Government the Exchange Rate.” Applied Economics, 23
Budgets and Money: How Are They Related?” (September 1991), 1551-1558.
Journal of Economic Education, 17 (Spring 1986), McNelis, P. D. “Irrepressible Monetarist Conclusions
107-119. from a Non-Monetarist Model.” Journal of Mone¬
Cecchetti, S. G. “Seigniorage as a Tax: A Quantita¬ tary Economics, 6 (January 1980), 121-127.
tive Evaluation: Comment.” Journal of Money, Ratti, R. A. “A Descriptive Analysis of Economic In¬
Credit, and Banking, 23 (August 1991), 476-480. dicators.” Federal Reserve Bank of St. Louis
Fackler, J. S., and W. D. McMillin. “Federal Debt Review, 67 (January 1985), 14-24.
and Macroeconomic Activity.” Southern Eco¬ Ratti, R. A. “The Effects of Inflation Surprises and
nomic Journal, 55 (April 1989), 994-1003. Uncertainty on Real Wages.” Review of Econom¬
Kandil, M. “Variations in the Response of Real Out¬ ics and Statistics, 67 (May 1985), 309-314.
put to Aggregate Demand Shocks: A Cross- Raynold, P., W. D. McMillin, and T. R. Beard. “The
Industry Analysis.” Review of Economics and Sta¬ Impact of Federal Government Expenditures in the
tistics, 73 (August 1991), 480-488. 1930s.” Southern Economic Journal, 58 (July
Kojima, K., and T. Ozawa. “Micro- and Macro- 1991), 15-28.
The IS Curve 579
Salemi, M. K. “On Teaching a Fractured Macroeco¬ and Nominal Shocks on Union-Firm Contract Dura¬
nomics: Thoughts.” Journal of Economic Educa¬ tion.” Journal of Monetary Economics, 27 (June
tion, 18 (Spring 1987), 227-231. 1991), 361-380.
Scaperlanda, A. “Factors Affecting Private Invest¬ Wulwick, N. J. “The Phillips Curve: Which? Whose?
ment.” Business Economics, 15 (May 1980), To Do What? How?” Southern Economic Journal,
73-74. 53 (April 1987), 834-857.
Wallace, F. H., and H. Blanco. “The Effects of Real
Appendix An IS-LM Model of Aggregate Demand

In the text, we develop a model of aggregate demand from four components
of aggregate spending: consumption, investment, government purchases, and
net exports. In this model aggregate demand is downward sloping, for two
reasons. First, there is a wealth effect on consumption from any decrease in
the price level that leads to an increase in consumer expenditures. Second,
any decrease in the price level leads to an increase in the real money supply
that will cause a decline in the interest rate, thus stimulating investment and
consumption spending. Due to these two effects, consumption demand and
investment demand depend on the price level, and hence aggregate demand
depends on the price level.
The relationship between changes in the price level and changes in the
quantity demanded of aggregate real goods and services can be derived more
formally using what is called the IS-LM model. To develop this model, we
first derive the IS curve.
The IS Curve
The IS curve is defined as the set of real GDP and real interest rate values
consistent with equilibrium between aggregate spending and aggregate in¬
come. (The name IS signifies that this curve is the set of points at which
Investment equals Saving, an alternative way of looking at equilibrium
between spending and income.) Recall that in Figure 16.7 we derived a
relationship that we labeled aggregate demand, or AD, from the relationship
Y = Cd + Id -F Gd + NXd. We showed that the requirement that the income
variable affecting Cd be the same value as aggregate expenditures (or real
GDP) allowed us to derive an aggregate demand curve that held whenever
income equals aggregate spending.
The curve we derived, which we labeled AD, is very similar to the curve
we call IS here. The IS curve is essentially our aggregate demand curve
graphed as a function of the real interest rate: IS = IS (r, P, ire, W, TH, TF,
EE, G, NX). This function is the IS equation, and we can graph this curve
as we do in Figure 16A.1, with the real interest rate on the vertical axis and
Figure 16A.1
The IS Curve
real output, Y, on the horizontal axis. The IS curve is downward sloping

because increases in the real interest rate lead to reductions in investment
and consumption spending. Increases in the other variables cause shifts in
the IS curve. For example, an increase in government purchases, G, causes
income to increase at every level of the real interest rate, so IS shifts to the
right. An increase in the price level, P, while holding nominal wealth con¬
stant will cause real wealth to decline for every level of the real interest rate,
which reduces consumption spending and hence real GDP, so IS shifts to
the left.
The LM Curve
The LM curve is the set of real interest rates and real output values consistent
with equilibrium in the money market. (The name LM signifies that this
curve is the set of points at which liquidity demand—i.e. money demand—
The LM Curve 581
equals the supply of money.) We derive LM from the money market equi¬
librium as in Figure 16A.2. Here money demand, a function of real income
F0, intersects real money supply at point A, where the interest rate is i0. For
a given expected inflation rate tt6, the real interest rate is then r0 = i0 —
7Te. Thus, holding everything else constant, the money market equilibrium
at point A has a real interest rate of r0 and real output of F0-
What happens if income increases to YX1 Money demand increases to
Ld(Yi), and money market equilibrium moves to point B, with an interest
rate of q and a real interest rate of rx = q — tts. Thus we continue to hold
everything else constant and find that the higher income level Yx and higher
real interest rate rx occur in the equilibrium at point B.
We can continue to change real income and observe what happens to
the interest rate—and the real interest rate—due to the money market. By
doing so, we can find all combinations of real income and the real interest
rate consistent with equilibrium in the money market when we hold constant
all other determinants of either money supply or money demand. This gives
us the LM curve in part b of Figure 16A.2.
The LM curve is upward sloping, reflecting the relationship between
equilibrium real income and equilibrium real interest rates in the money
market. The LM curve holds constant a number of variables, including the
expected inflation rate, the nominal money supply, and the price level. An
increase in the nominal money supply, a reduction in the price level, or an
increase in the expected inflation rate will shift LM to the right.
IS-LM Equilibrium
IS tells us the set of equilibrium real interest rates and real income levels
that can occur when aggregate demand (or aggregate spending) equals ag¬
gregate income. The LM curve tells us the set of equilibrium real interest
rates and real income levels that can occur when real money supply equals
real money demand. The intersection of IS and LM when both curves are
placed on the same graph tells us the real interest rate and real income level
that are consistent with equilibrium both between expenditures and income
and between money supply and money demand. This is shown in Figure
16A.3. Changes in any of the determinants of IS or LM will result in changes
in both the real interest rate and real income. For example, an increase in
the nominal money supply will shift LM to the right, reducing the real
interest rate and increasing real income. An increase in government pur¬
chases will shift IS to the right, increasing both the real interest rate and real
income.1
Aggregate Demand
One criticism of the IS-LM model is that it holds the price level constant.
To counter this criticism, we can use IS-LM to derive an aggregate demand
curve. We do this by asking how the equilibrium between IS and LM might
change with changes in the price level. For example, in part a of Figure
16A.4 we graph both LM(P0) and IS(F0) and note their intersection at point
A, with a real interest rate of r0 and real income of Y0. What happens if the
price level increases to PX1 This reduces real money balances to M/Px, so
LM shifts to the left. It also reduces real wealth, which shifts IS to the left.
The new intersection is at point B, where real output has fallen to Y\. At
point B, the real interest rate may rise or fall depending on the magnitude
of the shifts in IS and LM, but we have drawn the shifts so that the real
interest rate rises to rx. In any case, the key is that the increase in the price
level to Pi has resulted in a fall in real output to Yx. In fact, every time the
price level increases, IS and LM shift leftward, reducing real output. Every
time the price level decreases, IS and LM shift rightward, and real output
increases. In part b of Figure 16A.4, we draw this inverse relationship
between the price level and real output as the aggregate demand curve, AD.
Why does this AD curve slope downward? There are two reasons. First,
as the price level rises, real money balances decline, and this reduces real
money supply and leads to a higher real interest rate, which tends to reduce
investment and consumption spending and hence lower aggregate demand.
This is the leftward shift in LM. Second, as the price level rises, real wealth
declines, which directly reduces real consumption spending. This is the
leftward shift in IS. In terms of changes in real income, the shifts in IS and
LM reinforce each other, reducing real income to Yx. Thus, AD slopes
downward in the IS-LM scenario for the same reasons given in the text.
This AD curve can be used, along with the aggregate supply curves
developed in the text, to analyze the effects of changes in the determinants
of IS or LM on the price level and real income. One advantage of the
IS-LM approach is that it more clearly ties together the equilibrium rela¬
tionship between real income and real spending—the IS curve—and the
1Notice that these results for interest rates and for real output are the same as we find in the body
of the text for an increase in government purchases or an increase in the money supply.
money market equilibrium—the LM curve. This allows us to see how

changes in the price level and real output can lead to changes in the equilib¬
rium real interest rate. However, the results we got in the text are basically
the same as those that can be derived using the IS-LM approach, at least in
terms of the directions of movement in the variables due to changes in the
determinants of AD.
IS-LM is not without its critics or disadvantages. One disadvantage is
that the IS-LM model relies on the money market equilibrium for any anal¬
ysis of changes in the interest rate. However, we saw in Chapter 15 and in
this chapter that the loanable funds market is often more convenient for
analyzing the effects on the interest rate of changes in variables other than
the money supply. For example, the effects on real interest rates of increases
in the government deficit are most easily analyzed in the loanable funds
market, as are the effects of increases in the expected inflation rate.
v
CHAPTER
17 Open Economy
Macroeconomics
n Chapter 16, we examined a simple but useful model of the macro

economy. The model presented there is a model with exogenous net exports,
essentially a modest extension of a closed economy model, one that does not
consider the impact of changes in international trade on the economy. For
many years it was considered acceptable to model the United States in this
way, since international trade was only a small part of the U.S. economy.
However, international trade and investment flows have become increasingly
important to the United States. In 1990, gross domestic product was $5,465
billion, while U.S. exports to other countries were valued at $653 billion, or
more than 10 percent of output. Imports of foreign goods into the United
States totaled $723 billion, an even larger percentage of GDP. Thus, foreign
trade in goods and services is an important component of the U.S. economy
today. Foreign investment is also becoming increasingly important. In 1990,
U.S. direct investment in other countries totaled $421 billion, while foreign
investment in the United States totaled $404 billion. This amounts to about
8 percent of GDP, indicating a substantial flow of funds between the United
States and other countries.
Because of the growing importance of international trade and finance in
the United States, we now extend the model of Chapter 16 to take account
of international trade. The model we build in this chapter has endogenous
net exports and is an open economy model, a model of an economy that is
open to international trade. In Chapter 16, aggregate demand consisted of
consumption demand, investment demand, government demand, and exog¬
enous net exports. In this chapter, our model includes endogenous net ex¬
ports. This not only affects the slope of the aggregate demand curve, giving
yet another reason for downward-sloping aggregate demand, but also adds
to the list of variables that shift aggregate demand.
We also look at the relationship between aggregate demand and aggre¬
gate supply in the open economy, with the particular aim of seeing how
international trade affects macroeconomic equilibrium. We will explore the
importance of open economy considerations in the loanable funds market
and see how international borrowing and lending of loanable funds has
important implications for the interest rate effects of government finance.
We will also see that the effectiveness of monetary and fiscal policy in
586
stimulating the economy depends on whether exchange rates are fixed or

flexible. Throughout this chapter, we will pay close attention to the effects
of policies that spring from special features of the open economy.
The Balance of Payments
The balance of payments summarizes the value of payments between house¬

holds, firms, and governments in one country and the rest of the world. For
the United States, this includes an accounting of imports and exports of
goods and services, financial capital flows that include lending to and bor¬
rowing from other countries, and transfers and gifts between U.S. house¬
holds, firms, or governments and the rest of the world. U.S. imports are
goods and services produced in other countries and purchased by U.S. res¬
idents. When you buy a TV made in Taiwan, you are purchasing an import,
a good or service that is made overseas but is purchased by a U.S. resident.
U.S. exports are goods and services produced in the United States and
purchased by the rest of the world. When a French business purchases
computer software produced in the United States, this transaction represents
an export from the United States.
In the balance of payments calculations, any transactions that result in
a receipt of payments by a U.S. firm, household, or government is a positive
entry—a credit. Likewise, any transaction that results in a payment by a U.S.
firm, household, or government to a foreign party is a negative entry—a
debit. Thus, exports are a positive entry, since the sale of U.S. products
overseas results in the receipt of funds by the United States. Imports are a
negative entry, since payment for imports by parties in the United States
results in funds being received overseas.
Flows of financial assets also affect the balance of payments. When U.S.
firms, households, or governments borrow from overseas, funds flow from
abroad to the United States, resulting in a positive entry in the balance of
payments. U.S. residents receive funds, while the rest of the world receives
an asset, an IOU payable in the future. Similarly, when the United States
lends funds to the rest of the world, the result is a negative entry in the
balance of payments, because funds flow out of the United States into the
hands of overseas parties.
How do purchases of stocks or bonds in other countries affect the balance
of payments account? When a U.S. investor purchases stock on the London
stock exchange, a negative entry in the U.S. balance of payments results
because funds flow from the United States to the United Kingdom. Similarly,
when a Japanese investor purchases a U.S. government bond, the result is a
positive entry in the U.S. balance of payments because funds enter the United
States from Japan. The key in all these examples is to remember that it is
the direction of the flow of funds that determines whether a transaction is a
positive or a negative entry in the balance of payments.
588 Chapter17 Open Economy Macroeconomics
The balance of payments accounts are constructed so that the total of all
funds flowing overseas should equal the total of all funds flowing to the
United States. The idea is the same as any double-entry bookkeeping system.
In the balance of payments accounts, an inflow of funds from the sale of
U.S. exports will be matched by an outflow of funds as U.S. producers buy
assets or goods in other countries. Thus, the overall balance of payments
should be zero. In practice, however, the overall balance is not zero due to
measurement errors and omissions, called the statistical discrepancy, in the
balance of payments accounts. The statistical discrepancy can be rather
substantial, indicating the difficulty governments have in keeping track of
international transactions.
The balance of payments accounts are divided into two major sub¬
accounts. The first is called the current account and the second the capital
account. The current account records the net flow of funds due to trade in
goods and services, investment income, and gifts (unilateral transfers). As
always, actions that generate a flow of funds to the United States are credits
(positive entries), and actions that create a flow of funds out of the United
States are debits (negative entries). The capital account records the net flow
of funds due to changes in the levels of capital and financial assets between
the United States and the rest of the world. When U.S. residents or the U.S.
government increase their holdings of foreign assets, funds flow from the
United States overseas to purchase those assets, so a debit or a negative
entry in the capital account occurs. When foreign residents or governments
increase their holding of U.S. assets, funds flow to the United States to
purchase those assets, and a credit or a positive entry in the capital account
results.
We cannot stress too much that aside from the statistical discrepancy,
the sum of the current account and the capital account should be zero by
construction. Thus, when news commentators report a balance of payments
deficit or surplus, they cannot be referring to the overall balance of payments.
Instead they are referring to one of the subaccounts, most likely the current
account balance or even the merchandise trade balance (net exports of
goods).
The balance of payments are reported in a table such as Table 17.1. Here
we see the main categories in the balance of payments—the current account
and the capital account—as well as details of the components of these
accounts. We look at these accounts in more detail in the following sections.
The Current Account

The current account consists of the merchandise trade balance, the balance
on services, the net flow of income on investments, and the net flow of
unilateral transfers. The merchandise account consists of exports and imports
of goods, where exports are a positive entry (funds flow to the United States)
and imports a negative entry (funds flow abroad). The difference between
The balance of pay¬ Table 17.1

ments accounts con¬ The Balance of Payments Accounts
sist of the current ac¬
count balance and the
capital account bal¬ 1. Current Account
ance. A. Merchandise Trade
i. Exports (+)
ii. Imports (—)
B. Services
i. Exports (+)
ii. Imports (—)
Bl. Net Military Transactions
(included in B)
C. Income on Investments
i. Receipts (+)
ii. Payments (—)
D. Unilateral Transfers
2. Capital Account
A. U.S. investment overseas (increase of capital outflows)
i. U.S. private assets, net
ii. U.S. government assets other than official reserve assets
iii. U.S. official reserve assets (ORAs)
B. Foreign investment in the United States (increase in capital inflows)
i. Private and other (non-ORA) foreign assets, net
ii. Foreign official reserve assets (ORAs)
merchandise exports and merchandise imports is called the merchandise

trade balance.1 A positive trade balance simply means the United States
exported more goods than it imported; a negative balance of trade indicates
the United States imported more goods than it exported.
The balance on services consists of exports minus imports of services,
including such items as military transactions, tourism, and transportation
services. The category net military transactions includes military grants as
well as sales, which we include as a separate category for later use.
Net income on investments includes the net flow of investment income
such as dividends, interest income, and royalties. Income on foreign assets
owned by U.S. residents is an inflow of funds to the United States—a credit.
Income paid to foreign owners of U.S. assets is an outflow of funds from
the United States—a debit.
Finally, the category unilateral transfers is the net flow of funds from
gifts between U.S. and foreign residents. These are transfers of funds or
1News commentators sometimes mistakenly refer to this as the balance of payments. As discussed
earlier, the balance of payments is always zero by construction.
590 Chapter 17 Open Economy Macroeconomics
goods for which no payment is expected in return. They include government

grants (essentially foreign aid), private gifts sent to and received from other
countries, and pension payments made to households overseas. If U.S. res¬
idents are net recipients of such gifts, this entry will be positive. In the usual
case this entry is negative for the United States, representing that on net the
United States is a donor of gifts to overseas residents.
The current account balance is the sum of the merchandise trade balance,
the balance on services, net income on investments, and unilateral transfers.
The current account balance is closely related to but not exactly the same as
what we call net exports in aggregate demand. The category net exports
includes net exports of merchandise—the merchandise trade balance—and
net exports of services. However, the balance on services includes one
category, net military expenditures, which consists of grants of military
equipment and supplies from the U.S. government. Since this category does
not represent actual sales of military goods abroad, we would exclude it
from the calculation of net exports. Hence, net exports is the sum of the
merchandise trade balance and the balance on services, minus net military
transactions. Thus, net exports in aggregate demand differs from the current
account balance in that net exports does not include net military transactions,
net income on investments, or unilateral transfers.
The Capital Account

The second major category in the balance of payments is the capital account.
The capital account keeps track of investments by U.S. firms, households,
and governments overseas and investments by foreign residents in the United
States. These international movements of loanable funds are called inter¬
national financial capital movements, or simply international capital move¬
ments. They represent the lendings and borrowings by U.S. households,
firms, and government. If U.S. households and firms lend to a foreign coun¬
try, they receive foreign securities (such as stocks and bonds) in exchange
for funds sent overseas. Since funds flow out of the United States, these
movements are a negative entry in the capital account. Borrowing by U.S.
firms and households from foreign sources is a positive entry in the capital
account, because funds flow into the United States in exchange for securities
flowing to overseas lenders. A positive capital account balance indicates the
United States borrows more funds from abroad than it lends overseas; there
is a net inflow of loanable funds into the United States.
One set of entries in the capital account is changes in U.S. official reserve
assets and changes in foreign official reserve assets. U.S. official reserve
assets are assets held by the U.S. Treasury or by the Federal Reserve System
that consist primarily of gold and holdings of foreign currencies. Foreign
currency is obtained when the Federal Reserve or the Treasury purchases it
to influence exchange rates. When the Federal Reserve buys foreign currency
and official reserve assets increase, this is a debit—a negative entry—because
dollar funds flow out of the United States. Similarly, if the Federal Reserve
sells foreign currency, obtaining dollars but reducing its holding of official
reserve assets, the result would be a credit, because dollar funds would flow
into the United States.
Recent U.S. Balance of Payments Accounts

Table 17.2 presents the U.S. balance of payments accounts for 1990, 1991,
and the first nine months of 1992. Notice the U.S. current account was in
deficit in all three years. The current account balance was — $90.4 billion in
1990 but moved dramatically toward balance in 1991, when it was — $3.7
billion. A large portion of this move was due to the change in unilateral
transfers from — $32.9 billion to + $8.0 billion as countries around the world
made unilateral transfers to the United States to help pay for the Persian
Gulf War. In 1992 the current account balance moved back to a larger deficit
position, —$37.9 billion in the first nine months.
The components of the current account provide interesting information
about the U.S. trade situation. Notice in Table 17.2 that the merchandise
trade balance was negative in all three years, meaning the United States
imported more goods than it exported. However, the balance on services
was positive; the United States exported more services than it imported.
Income on overseas investments was also positive, indicating U.S. residents
received more income from their investments abroad than foreign residents
received from their investments in the United States. Finally, with the ex¬
ception of the Persian Gulf War payments in 1991, the United States was a
net donor of unilateral transfers to foreign residents.
What about the capital account? The U.S. capital account was in surplus
each year, especially in 1990 and 1992. In each year, overseas residents
increased their assets held in the United States by more than the United
States increased its assets held overseas. This resulted in an inflow of funds
to the United States and a positive capital account balance as foreign resi¬
dents purchased U.S. assets.
One special type of official reserve asset is called SDRs, or special
drawing rights. Special drawing rights are an international reserve asset
created by the International Monetary Fund (IMF) as a substitute for gold.
A country in need of international reserves could obtain an allocation of
SDRs from the IMF that other countries would agree to honor as an inter¬
national reserve asset, valid for trade in the currency markets. Entry 3 in
Table 17.2 shows that SDRs were zero in these three years, indicating that
the U.S. did not obtain international reserves from the I.M.F.
Entry 4 shows a discouraging feature of the U.S. balance of payments
accounts: the size of the statistical discrepancy. This measure was attributed
to such factors as illegal international trade in drugs and other contraband,
unreported international trade in goods and services (to avoid tariffs or other
trade restrictions), or unreported international borrowing and lend-
Table 17.2
U.S. Balance of Payments Accounts for 1990, 1991,
and 1992 (partial) (billions of dollars)
1992
1990 1991 (Jan.-Sept.)
1. Current Account -90.4 -3.7 -37.9

A. Merchandise Trade Balance -108.9 -73.4 -68.3
i. Exports (+) 388.7 416.0 326.2
ii. Imports ( — ) -497.6 -489.4 -394.5
B. Services 32.0 45.3 42.1
i. Exports (+) 148.5 163.6 133.7
ii. Imports (—) -116.5 -118.3 -91.6
Net Military Transactions -7.8 -5.5 -1.8
(included in B)
C. Income on Investments 19.3 16.4 10.0
i. Receipts (+) 143.5 125.6 84.7
ii. Payments (—) -124.3 -108.9 -74.8
D. Unilateral Transfers -32.9 8.0 -21.7
2. Capital Account 43.1 4.8 58.9
A. U.S. investment overseas -56.3 -62.2 -24.4
i. U.S. private assets, net -56.5 -71.4 -26.0
ii. U.S. government assets
other than ORAs 2.3 3.4 2.4
iii. U.S. ORAs -2.2 5.8 -0.7
B. Foreign investment in U.S. 99.4 67.0 83.3
i. Private and other 65.4 48.6 48.9
ii. Foreign ORAs 33.9 18.4 34.3
3. Allocation of SDRs 0.0 0.0 0.0
4. Statistical Discrepancy 47.4 -1.1 -21.0
5. Net Exports of Goods and Services -69.1 -22.6 -24.4
6. Official Settlements Balance 31.7 24.2 33.6
Note: Numbers may not add to reported sums due to rounding. Data for 1992 is for first
nine months only.
Sources: Economic Report of the President, (Washington, D.C.: U.S. Government Printing Office,
January 1993), pp. 462-463; U.S. Commerce Department, Survey of Current Business, various
ing. The size of the statistical discrepancy was large relative to the other
entries in the balance of payments accounts, indicating the difficulty gov¬
ernment accountants have in keeping track of international transactions. In
1990, for instance, the statistical discrepancy was $47.4 billion, larger than
the entire surplus on the capital account!
Entry 5 in Table 17.2 also reports net exports of goods and services, the
variable that enters aggregate demand calculations. This measure is the sum
of the merchandise trade balance and the balance on services, minus net
military transactions (which are largely grants and aid without repayment
and hence not sales of goods and services). Net exports were negative but
grew closer to balance, from —$69.1 billion in 1990 to —$22.6 billion in
1991, before moving further from balance, to —$24.4 billion over the first
nine months of 1992.
Finally, Entry 6 shows that the official settlements balance, the sum of
the change in U.S. official reserve assets and the change in foreign official
reserve assets, was positive in each year.
Summary How would the following transactions affect the balance of payments? (a)
Exercise 17.1 A U.S. household purchases a car from Germany, (b) A U.S. firm sells a
bond to a Dutch investor, (c) A Japanese citizen tours the United States,
spending funds on lodging, food, and souvenirs, (d) The U.S. government
makes a gift of wheat to Somalia.
Answer: (a) This is a negative entry in the current account, since funds
flow from the U.S. household to Germany in exchange for the car. (b) In
this case, the sale of the bond leads to an inflow of funds to the United
States, so this sale is recorded as a positive entry in the capital account, (c)
The Japanese tourist is buying goods and services from the United States,
creating a flow of funds into the United States, so this is a positive entry in
the current account, (d) The dollar value of the gift of wheat shows up as a
negative entry in the unilateral transfers component of the current account.
The Demand for Net Exports

We have seen that net exports have fluctuated in recent years. Why do these
movements occur, and how do they affect aggregate demand and macroeco¬
nomic equilibrium? To answer these questions, we now look at the demand
for net exports—the demand for exports minus the demand for imports. The
demand for net exports has two components: the demand for U.S. exports
(the demand by foreign residents for U.S. goods and services) and the
demand for imports (the U.S. demand for foreign goods and services). We
begin with two examples that illustrate why the domestic price level, the
foreign price level, and the exchange rate are primary deteminants of import
and export demand.
Suppose a Japanese student is considering the purchase of word proc¬
essing software made by a U.S. producer. The list price is $400 (including
shipping). If the exchange rate in yen per dollar is 110, this software costs
$400 X ¥110/$ = ¥44,000 for the Japanese student. In deciding whether to
purchase this software or another product, the student is concerned about
the price in Japanese yen, which is the dollar price times the exchange rate.
The higher either the exchange rate or the dollar price, the less likely the
Japanese student is to buy U.S.-made software. More generally, a rise in
either the exchange rate or the U.S. price level will lower the demand for
U.S. exports, since it will make U.S. goods more expensive relative to foreign
goods.
Similarly, suppose you are considering purchasing a car imported from
Japan. You look at the price of the Japanese car relative to the price of cars
made in the United States. In this case, any increase in the price of the
Japanese car will tend to reduce your desire to buy it, thus reducing U.S.
imports. But the price you have to pay for the Japanese car depends on the
price in Japan, in yen, divided by the exchange rate in yen per dollar. For
example, consider a car that costs ¥1,100,000 in Japan and an exchange rate
of ¥110 per dollar. The price in dollars is ¥1,100,000/(¥110/$) = $10,000.
If the price in yen rises or the exchange rate falls, the dollar price of the
Japanese car will rise, making it less attractive for you to purchase it. More
generally, the higher the foreign price level or the lower the exchange rate,
the lower will be the U.S. demand for imports.
These examples explain why changes in price levels and exchange rates
affect imports and exports. Since the demand for net exports is defined as
export demand minus the demand for imports, the examples also suggest
that net export demand depends on price levels and exchange rates. This is
indeed the case, and in fact there are several other important determinants
of net export demand. We summarize all of these determinants next.
The Domestic Price Level. If we hold everything constant except the

domestic price level and net exports, we obtain a relationship between the
price level and expenditures for net exports called the net export demand
curve. This relationship is the curve labeled NXd in Figure 17.1 and indicates
that, holding other things constant, an increase in the domestic price level
will reduce the amount of real spending on net exports demanded. This is
represented by the movement from A to B along net export demand curve
NXd. The downward slope reflects the impact of a rise in the domestic price
level on exports and imports: Exports fall (since U.S. goods and services are
more expensive for overseas residents) and imports rise (since foreign goods
and services are now cheaper relative to U.S. goods and services). The
decline in exports and rise in imports reduce net exports.
One unusual feature of the net export demand curve is that it can be
negative. Usually we think demand must be positive or zero. However, net
exports is exports minus imports. When imports exceed exports, net exports
is negative, which simply means that (ignoring net income on investments
and transfers) there is a current account deficit in the balance of payments.
In contrast, when exports exceed imports, a current account surplus exists:
Exports of U.S. goods and services exceed imports.
Figure 17.1 reveals that, holding other things constant, the magnitude of
the current account deficit depends on the domestic price level. When the
domestic price level is P0, net exports are zero on the curve labeled NXd. If
the domestic price level falls below P0 and we move to point A, U.S. goods
and services become cheaper relative to foreign ones. At point A net exports
are + $100 billion, indicating a positive current account balance: The United
States exports $100 billion more in real goods and services than it imports.
Similarly, if the domestic price level rises above P0 to point B on the NXd
curve, U.S. goods and services become more expensive relative to foreign
ones. At point B net exports are — $100 billion, indicating a current account
deficit.
Whenever there is a change in a determinant of net export demand—
say, a change in the foreign price level—the entire net export demand curve
shifts to the left or to the right. This is called a change in net export demand
and is illustrated in Figure 17.1 by shifts from NX' to either NXf or NXC{.
Next we look at how changes in the foreign price level and other determi¬
nants of net export demand can shift the net export demand curve.
The Foreign Price Level. Suppose the price level in Japan rises so
that the yen prices of Japanese goods increase. For a given exchange rate
and U.S. price level, this makes Japanese goods more expensive to both
Japanese and U.S. buyers. Japanese buyers will therefore tend to substitute
toward U.S. goods, and U.S. exports will rise. Meanwhile you and other
U.S. consumers will find it more costly to buy cars and other goods imported
from Japan and will likely substitute toward U.S. goods, so imports from
Japan will fall. Together the rise in exports and the decline in imports mean
the difference, net exports, increases. More generally, a rise in the foreign
price level increases the demand for net exports. This is represented by a
rightward shift from NXd to NXd in Figure 17.1. A decrease in the foreign
price level shifts the net export demand curve to the left to NXd.
The Exchange Rate. Changes in the exchange rate function in much

the same way as changes in the foreign price level. A rise in the exchange
rate (given as the number of units of a foreign currency needed to buy a
single U.S. dollar, such as 100 yen per dollar) means foreign residents need
more of their currency to buy a dollar. We say their currency has depreciated
relative to the dollar. Holding everything else constant, including the U.S.
price level, this means they need more of their currency to buy any U.S.
good, making it less desirable to do so. Thus, their demand for U.S. goods
and services will decline when the exchange rate rises, and export demand
will fall. In like fashion, U.S. consumers compare the prices of U.S. and
foreign goods. For a given U.S. price level and foreign price level, the
increase in the exchange rate means it takos fewer U.S. dollars to buy foreign
goods, so U.S. demand for imports increases. We say that the dollar has
appreciated relative to the foreign currency.
For example, suppose a student in Britain is considering the purchase of
a sweater that costs $20 in the United States. If the exchange rate is .5
English pounds per dollar, then, ignoring shipping costs, the student can buy
this sweater for 10 pounds ($20 X .5 pounds/dollar = 10 pounds). If the
exchange rate increases to 1 pound per dollar, the British student must pay
20 pounds ($20 X 1 pound/dollar — 20 pounds) for the sweater. Thus, as
the exchange rate increases (i.e., as the pound depreciates relative to the
dollar), U.S. goods become more expensive in Britain and British consumers
will buy fewer of them, reducing export demand.
A similar analysis holds for imports. Consider the price you would have
to pay for a British car costing 10,000 pounds. If the exchange rate is .5
pounds per dollar, you will need $20,000 to purchase the car (since $20,000
X .5 pounds/dollar = 10,000 pounds). If the exchange rate rises to one
pound per dollar (i.e., the dollar appreciates relative to the pound), your cost
will be only $10,000. The increase in the exchange rate reduces the price of
the car to you, a U.S. resident, which leads to an increase in imports.
Thus, we see that an increase in the exchange rate (an appreciation of
the dollar) will reduce export demand and raise import demand, which means
net export demand must decline. This is illustrated by leftward shift in the
entire net export demand curve. Likewise, a decrease in the exchange rate
(a depreciation of the dollar) will increase net export demand (shift the entire
net export demand curve to the right). How well does this prediction hold
up in the historical record? Box 17.1 shows U.S. net exports and a weighted
average exchange rate between the currencies of the major U.S. trading
partners and the dollar.
Domestic Wealth. An increase in domestic wealth decreases net ex¬

port demand through its effect on the demand for imports. A rise in domestic
wealth increases the ability of U.S. residents to buy imported goods, thus
increasing the demand for imports. Consequently a rise in domestic wealth
leads to a decrease in net export demand, as represented by a leftward shift
in the net export demand curve.
Foreign Wealth. An increase in foreign wealth increases net export

demand through its effect on the demand for U.S. exports. An increase in
foreign wealth increases overseas residents’ demand for U.S. products, which
raise their demand for our exports. Thus, a rise (decline) in foreign wealth
leads to an increase (decrease) in net export demand, as represented by a
rightward (leftward) shift in the net export demand curve.
The Real Domestic Interest Rate. Holding other things constant,

a higher real domestic interest rate induces people in the United States to
save more, which leaves them less for purchasing imports. As a consequence
net exports rise, which is represented by a rightward shift in the net export
demand curve. A decline in the domestic real interest rate shifts the net
export demand curve to the left.
The Foreign Real Interest Rate. When the foreign real interest rate
increases, foreign households tend to save more and thus have less to spend
on U.S. goods. Their demand for U.S. exports will decline, as will U.S. net
export demand. Similarly, the lower the foreign real interest rate, the greater
will be foreign spending on consumer goods and services, including U.S.
products, and this will increase U.S. net export demand. Graphically, an
increase in the foreign real interest rate shifts the net export demand curve
to the left, whereas a decrease in the rate shifts the curve to the right.
Tariffs. Tariffs are taxes on imports and can be levied by both the United
States and other countries. When the United States imposes a tariff, it raises
598 Cha pter17 Open Economy Macroeconomics
Net Exports and the Exchange Rate
Do U.S. net exports respond to exchange rate left-hand scale, and the exchange rate is meas¬
movements? The answer is yes, but they do so ured on the right-hand scale. Notice that when
slowly and with lags. The accompanying graph the dollar appreciated in the early 1980s, peak¬
shows U.S. net exports and an exchange rate ing in 1985, net exports were declining, and
measure that averages together the exchange reached a record deficit of about -$140 billion
rate between the dollar and the currencies of in 1987. The exchange rate sank quickly from its
the United States' major trading partners. Net 1985 peak. After 1987 net exports began rising,
exports are given in billions of dollars on the and by 1991 they were around - $20 billion.
the aftertax price of imports to U.S. consumers. This reduces the relative
attractiveness of imported goods and lowers import demand. In turn, net
export demand increases. Graphically, an increase in tariffs levied by the
United States shifts the demand for net exports to the right, whereas a
reduction in U.S. tariffs shifts net export demand to the left.
When another country imposes a tariff on U.S. goods, it raises the prices
of those goods to its citizens. This tends to reduce their demand for U.S.
exports, making U.S. net exports decline. Thus, increases in foreign tariffs
on U.S. goods will reduce net export demand, shifting the net export demand
curve to the left. Decreases in foreign tariffs will increase net export demand
and shift the net export demand curve to the right.
Quotas and Other Nontariff Barriers. Quotas are one of several

possible nontariff barriers to trade. A tariff raises the price of imported goods
and thus reduces demand for them. A quota also reduces imports, but by
directly restricting their quantity. For example, the United States negotiated
a “voluntary” quota with Japanese auto companies on the quantity of Jap¬
anese cars allowed into the United States. This quantity restriction reduced
the number of Japanese autos imported to the United States. Other nontariff
barriers are restrictions or regulations on the import of certain goods for a
variety of reasons, ranging from restrictions on imports of medicines without
Food and Drug Administration approval to restrictions on imports of goods
from certain countries, such as cigars from Cuba, for political reasons. These
restrictions may serve legitimate public policy or national security concerns,
but they also restrict imports. Indeed, some restrictions that were once con¬
sidered important for these purposes have become difficult to remove even
after these justifications are no longer warranted, because some domestic
industry has prospered under the regulation and doesn’t want to face in¬
creased competition. An example from overseas is Japan, which has quotas
and other restrictions to protect its domestic rice farmers from foreign com¬
petition. The Japanese government claims these measures are important for
national security reasons, although there has been a move recently to loosen
some of these restrictions.
Quotas and other nontariff barriers imposed by the United States tend
to reduce imports to the United States and thus to increase net export demand.
This is illustrated by a rightward shift in net export demand. Quotas and
nontariff barriers imposed by other countries on United States goods tend to
reduce U.S. exports. This causes a decline in net export demand, or a leftward
shift.
One way tariffs and nontariff barriers such as quotas are relaxed is
through treaties among countries designed to free up trade among them.
Recently the United States, Canada, and Mexico signed the North America
Free Trade Agreement (NAFTA) to establish freer trade among these three
countries, as discussed in Box 17.2.
Tastes and Expectations. Finally, tastes and expectations of both

U.S. and foreign residents can affect net export demand. During the 1980s,
for example, a popular campaign in TV commercials was “buy American.”
This slogan was designed by U.S. workers and producers to encourage U.S.
consumers to shun foreign products. If successful, such advertisements re¬
duce import demand, thus increasing the demand for net exports.
600 Cha pter17 Open Economy Macroeconomics
The Impact of NAFTA on Output

and Employment
The North America Free Trade Agreement, or now produced more efficiently, thus lowering
NAFTA, is a treaty between Canada, the United the costs of final goods and services to con¬
States, and Mexico to establish a free trade zone sumers. Flence free trade will push the long-run
in North America. NAFTA will require that the aggregate supply curve to the right.
three partners lower the tariffs and trade barri¬ If free trade is such a good thing, why was
ers that apply to one another's goods and ser¬ there so much opposition to NAFTA? The prob¬
vices. What will be the result for the U.S. econ¬ lem is that while free trade may be good for the
omy? The standard theory of international trade country as a whole, in terms of a higher long-
presented in economics textbooks is that free run natural rate of output, it may well have ad¬
trade will make all countries better off. This re¬ verse effects on individual firms or households.
sult is based on the idea that free trade allows For example, NAFTA will result in some changes
greater specialization and thus increases overall in the locations of various firms. Some U.S. firms
economic efficiency. As each country specializes and workers will benefit from increased access
in the production of those goods in which it has to Canadian and Mexican markets, but others
a comparative advantage, the greater quantity will be hurt by increased competition from these
and quality of goods available will benefit all countries. During the adjustment period, as pro¬
countries. duction moves to the country with the compara¬
In the economy as a whole, an increase in tive advantage, those firms and workers dis¬
specialization and economic efficiency raises the placed by more efficient production elsewhere
natural rate of output. A rise in international will have to find new jobs or new goods and
trade allow firms to use domestic factors of pro¬ services to produce. Even though the final result
duction more efficiently. This means an econ¬ is more efficient production and expanded con¬
omy can produce a larger quantity of output sumption possibilities, during the adjustment
than previously. Moreover, if labor is now more process some serious economic difficulties may
efficient in production, the increase in labor de¬ arise. Those who see their jobs or firms in dan¬
mand should translate into higher employment. ger from increased competition abroad were
Free trade also increases the consumption op¬ understandably reluctant to see NAFTA enacted.
portunities for a country, because goods are
Summary Holding other things constant, determine the impact of each of the following
Exercise 17.2 on net export demand: (a) a decrease in the exchange rate, (b) a decrease in
the U.S. interest rate, (c) an increase in tariffs on Japanese goods, and (d) a
European quota limiting the number of personal computers U.S. firms can
sell in Europe.
(a) A decrease in the exchange rate means it takes fewer units

of a foreign currency to buy a dollar or, equivalently, more dollars to buy a
unit of the foreign currency. This makes U.S. exports less expensive to
foreign residents, thus increasing demand for U.S. exports. It also makes
imported goods more expensive in the United States. Thus, net export de¬
mand increases, (b) A decrease in the U.S. interest rate will increase net
export demand, (c) An increase in tariffs on Japanese goods will reduce U.S.
demand for imports and hence increase net export demand, (d) A European
quota on U.S. goods will reduce demand for U.S. exports and hence reduce
U.S. net export demand.
Open Economy Macroeconomic Equilibrium^

We are now ready to see how international trade affects macroeconomic
equilibrium. We will want to know about (1) U.S. (or domestic) interest
rates, (2) the exchange rate between U.S. dollars and other currencies, (3)
the U.S. price level, and (4) U.S. real GDP. The equilibrium interest rate is
determined in the money market or, equivalently, in the market for loanable
funds. The equilibrium exchange rate is determined in the foreign exchange
market, which we examined in detail in Chapter 11. Here we focus on
determining the final two elements of macroeconomic equilibrium: the U.S.
price level and real GDP.
As in a closed economy, in an open economy the equilibrium U.S. price
level and real GDP are determined in the market for U.S. goods and services
by the intersection of the aggregate demand and supply curves. Figure 17.2
shows such an equilibrium at point A, where the equilibrium price level is
P0 and the equilibrium level of real GDP is Y0. While this picture looks
identical to that for a closed economy, we need to address several aspects
of these aggregate demand and supply curves before we examine the effects
of fiscal and monetary policy in an open economy.
Aggregate Demand in an Open Economy

Aggregate demand in our open economy includes consumption demand,
investment demand, government purchases, and an endogenous net export
demand; that is, aggregate demand for U.S. goods in an open economy may
be written as
AD = Cd + Id + Gd + NXd.
The aggregate demand curve for an open economy is downward sloping due
not only to the wealth effect and interest rate (or intertemporal substitution)
effect described in Chapter 16 but also to the effect of the price level on net
export demand. We saw in Figure 17.1 that net export demand is negatively
related to the price level in our open economy model. Changes in the do-
mestic price level cause changes in the relative attractiveness of domestic

and foreign products, and this leads to changes in the quantity demanded of
domestic consumption (Cd), investment (Id), and net exports (NXd) when the
price level changes. This is the third reason for downward-sloping aggregate
demand, which we alluded to in Chapter 16 but postponed discussion of
until this chapter. Net export demand may also change because changes in
the price level alter the interest rate, as we saw in Chapter 16, and these
changes in the interest rate can lead to changes in exchange rates and hence
in net export demand. Thus, an increase in the price level will reduce real
money supply and hence lower the interest rate. As we saw in Chapter 11,
this will lead to a reduction in the exchange rate as international investors
find the United States a less rewarding place in which to invest. This reduc¬
tion in the exchange rate, in turn, leads to a decrease in net exports. This is
yet another reason net export demand is downward sloping. Thus, the slope
of net export demand and hence the slope of aggregate demand already
incorporates the effect of changes in the interest rate and the exchange rate
caused by changes in the price level. Of course, the effect of changes in the
interest rate or the exchange rate not caused by changes in the price level
will still shift aggregate demand.
As in a closed econ¬ Figure 17.2

omy, macroeconomic Macroeconomic Equilibrium in the Open Economy
equilibrium in the
goods market occurs
at point A where ag¬
gregate demand
equals aggregate sup¬
ply. In our open econ¬
omy, however, net ex¬
ports are endogenous,
and aggregate de¬
mand consists of con¬
sumption demand, in¬
vestment demand,
government pur¬
chases, and endoge¬
nous demand for net
exports.
What are the determinants of aggregate demand? These include not only
the determinants of consumption demand, investment demand, and govern¬
ment purchases (all of which we discussed in detail in Chapter 16) but also
the determinants of net export demand just discussed. For instance, in an
open economy, an increase in U.S. tariffs increases net exports and thus
leads to an increase in aggregate demand. More generally, a change in any
determinant of C*, ld, Gd, or NXd will change aggregate demand. Table 17.3
summarizes each of these determinants, along with their impact on aggregate
demand.
Short-Run Aggregate Supply in an Open Economy

In our model of the closed economy, we saw that the major determinants of
short-run aggregate supply are (1) the wage rate, (2) prices of other inputs,
(3) taxes, (4) the stock of physical capital, and (5) the level of technology.
In the open economy, two additional determinants come into play: (6) prices
of foreign inputs and (7) tariffs.
Prices of Imported Inputs. Not all inputs are from domestic sup¬
pliers. U.S. firms import a variety of inputs from around the world, including
oil from the Middle East; industrial diamonds from South Africa and the
former Soviet Union; computer chips from Japan; steel from Europe, Japan,
and South Korea; transistors and other electronic components from Taiwan
or Hong Kong; and so on. Increases in the prices of these imported inputs
shift short-run aggregate supply to the left.
The major imported U.S. input is oil and, interestingly, oil trades on
world markets in U.S. dollars. Some inputs, however, are priced in foreign
currencies, and in those cases the price to a U.S. firm is the foreign currency
price divided by the exchange rate (in foreign currency per U.S. dollar). For
example, if a computer chip costs 1,000 yen and the exchange rate is 125
yen per dollar, the price of each computer chip to a U.S. computer manu¬
facturer is 1,000 yen/(125 Yen/US$) = $8. Changes in the exchange rate
obviously can affect the dollar price to U.S. firms of such inputs, even if the
foreign price remains constant.
For simplicity, we will assume either that imported inputs trade in the
world market at dollar prices (like oil) or that firms obtaining inputs from
abroad hedge against short-run movements in exchange rates by using the
futures market (as discussed in Chapter 9). Under either of these scenarios,
short-run changes in the exchange rate will not affect the dollar price of
foreign inputs, and the short-run aggregate supply curve will not vary with
changes in the exchange rate.
Tariffs. We saw earlier in the chapter that tariffs on imported finished

goods raise the goods’ prices. However, not all tariffs are on imports of
finished goods. Many are on imported intermediate goods, such as steel or
604 C hapter 77 Open Economy Macroeconomics
|
Table 17.3
V :< .
Determinants of Aggregate Demand in an Open Economy

of Consumption Demand Consumption Demand Aggregate Demand
Increase in domestic wealth Shifts right Shifts right

Increase in domestic real interest rate Shifts left Shifts left
Increase in domestic expected inflation rate Shifts right Shifts right
Increase in taxes on domestic households Shifts left Shifts left

of Investment Demand Investment Demand Aggregate Demand
Increase in domestic real interest rate Shifts left Shifts left

Increase in taxes on domestic firms Shifts left Shifts left

of Government Purchases Government Demand Aggregate Demand
Exogenous increase in domestic government purchases Shifts right Shifts right

of Net Export Demand Net Export Demand Aggregate Demand
Increase in foreign price level Shifts right Shifts right

Increase in exchange rate Shifts left Shifts left
Increase in domestic wealth Shifts left Shifts left
Increase in foreign wealth Shifts right Shifts right
Increase in domestic real interest rate Shifts right Shifts right
Increase in foreign real interest rate Shifts left Shifts left
Increase in domestic tariffs or quotas Shifts right Shifts right
Increase in foreign tariffs or quotas Shifts left Shifts left
computer chips. In addition, the Clinton administration at one time consid¬

ered a tariff on oil imports. Such tariffs raise the prices of inputs to U.S.
firms, which in turn shifts the short-run aggregate supply curve to the left.
Reducing tariffs on imported inputs has the opposite effect: It shifts the
short-run aggregate supply curve to the right.
Long-Run Aggregate Supply in an Open Economy

The final feature of macroeconomic equilibrium that is affected by interna¬
tional trade is long-run aggregate supply—the supply of domestic goods and
services after wages and other input prices have adjusted to any change in
the domestic price level. As we discussed in Chapter 16, in the short run
input prices tend to be sticky, or slow to adjust. Consequently, as the price
level rises in the short run, the real cost of inputs to firms declines, and firms
increase real output as the price level increases. This gives the short-run
aggregate supply curve its upward slope. Over time, however, wages and
input prices increase, shifting the short-run aggregate supply curve to the
left until long-run equilibrium is achieved. In the long run, after all input
prices have fully adjusted to changes in the price level, the ratio of input to
output prices remains constant. This, in turn, implies that in the long run the
quantity of inputs used in production will be determined by the economy’s
natural rate of employment, giving rise to a vertical long-run aggregate
supply curve at the natural rate of output.
If foreign inputs are priced in dollars, the same reasoning gives rise to
a vertical long-run aggregate supply curve in a closed economy. In the long
run, increases in the U.S. price level will lead to proportional changes in
contracted prices of foreign inputs, leaving their relative prices unchanged.
Thus, when the prices of foreign inputs are in U.S. dollars, the long-run
aggregate supply curve in the open economy is also vertical, determined by
the natural rate of output.
What if foreign input prices are in foreign currency units? The long-run
aggregate supply curve will still be vertical, provided purchasing power
parity holds. Recall from Chapter 11 that purchasing power parity is the idea
that the exchange rate adjusts so that the U.S. price level is proportional to
the foreign price level divided by the exchange rate. According to purchasing
power parity, the exchange rate between the German mark and the U.S.
dollar would adjust in the long run to keep the ratio of the German price
level to the U.S. price level constant.
How would this work for an imported input price? Suppose a U.S.
computer manufacturer buys computer chips from a Mexican firm under a
contract specified in Mexican pesos—say, 100 pesos per chip. The cost of
the chips to the U.S. firm is the peso price divided by the exchange rate in
Mexican pesos per U.S. dollar—say, 3 pesos per dollar. Thus, the dollar cost
to the U.S. firm is 100 pesos/(3 pesos per dollar) = $33.33. The real cost
to the U.S. firm is this dollar cost divided by the price of the firm’s output.
For the overall economy, we simply divide the dollar cost of the inputs by
the U.S. price level, which might be taken to be 1.
What happens if the U.S. price level rises—say, to 2? In the short run,
if the Mexican price level and the exchange rate are held constant, the real
cost of the Mexican chips will fall, and U.S. firms will tend to purchase
more of them. This will lead to an increase in U.S. output and partly explains
why short-run aggregate supply is not vertical. If purchasing power parity
holds, however, in the long run the exchange rate will adjust to keep the
ratio of the Mexican price level to the U.S. price level constant.
Suppose the Mexican price level remains constant and the price of the
chips remains a constant 100 pesos per chip. Since the U.S. price level
doubled, the exchange rate will have to change from 3 to 1.5 pesos per
dollar. But this means the dollar cost to the U.S. firm changes to 100 pesos/
(1.5 pesos per dollar) = $66.67. Thus, the dollar cost doubles. However,
since the U.S. price level doubled, this means the real cost to the U.S. firm
has just returned to its original level. Purchasing power parity leads to a
vertical long-run supply curve in an open economy even if some input prices
are quoted in foreign currency units.
Finally, we note that in an open economy, quotas and tariffs on inputs
affect the natural rate of output, because they change the long-run level of
productive resources that can be used to produce in the United States. In¬
creases in U.S. tariffs or quotas on inputs reduce long-run aggregate supply,
whereas reductions in these trade barriers increase it.
Summary I Suppose the U.S. government imposes quotas on all items imported into the
Exercise 17.3 United States. Why might real GDP first increase and then decrease?
Answer: As the figure shows, the imposition of quotas affects both ag¬
gregate supply and aggregate demand. Quotas reduce imports and hence
increase net export demand and aggregate demand from AD0 to ADX. The
quotas reduce the availability of imported inputs, and this decreases both
short-run and long-run aggregate supply, from SRAS0 to SRASi and LRAS0
to LRASi, respectively. The increase in aggregate demand and the reduction
in short-run aggregate supply will definitely raise the domestic price level,
and in the short run it may also increase real GDP as the economy moves
from point A to point B. In the long run, however, wages and other input
prices will begin to rise in response to the increase in the price level, shifting
short-run aggregate supply to the left. If long-run aggregate supply were
unaffected, the long-run equilibrium would eventually return to the original
Real GDP (Domestic)

Fixed Exchange Rates, Flexible Exchange Rates, and Policy Effectiveness 607
level of real GDP. However, because long-run aggregate supply itself shifts
to the left due to the quotas, the long-run effect will be to reduce real GDP,
as the economy utimately moves to point D.
Fixed Exchange Rates, Flexible Exchange

Rates, and Policy Effectiveness
One important feature of the open economy that has large implications for
policy is the choice between fixed and flexible exchange rates. In Chapter
11 we learned that under fixed exchange rates, monetary policy is dedicated
to keeping the exchange rate fixed relative to some other currency. This
means monetary policymakers must adjust the money supply to keep the
exchange rate at the desired level instead of to achieve other macroeconomic
goals such as full employment or price stability. When the demand for dollars
is high, the Fed must increase the money supply to keep the exchange rate
from rising; when demand is low, it must decrease the money supply to keep
exchange rates from falling. A government participating in a fixed exchange
rate system gives up its ability to use monetary policy to stimulate the
economy. However, it can use fiscal policy.
In contrast, if the exchange rate is flexible, monetary policy is not tied
to maintaining the exchange rate. In this case, the Fed can use monetary
policy to achieve other macroeconomic goals or to stimulate the economy.
However, while having flexible exchange rates frees up monetary policy, it
restricts the use of fiscal policy. This restriction is not direct but is due to
the way fiscal policy works in the economy. In a closed economy, the
crowding-out effect discussed in Chapter 16 tends to at least partially coun¬
teract the effects of any increase in government purchases financed with
borrowing, because the increase in demand for loanable funds used to finance
the spending drives up the interest rate and reduces consumption and in¬
vestment. In an open economy, this crowding out occurs in consumption,
investment and net exports. Increases in the interest rate reduce consumption
and investment just as in a closed economy. Increases in the interest rate
attract foreign funds, which increases the capital account. These increases in
the capital account are balanced by reductions in the current account, mean¬
ing net exports decline. This occurs because increased foreign capital inflows
result in a rising demand for dollars, driving the exchange rate in foreign
currency per U.S. dollar upward (an appreciation of the dollar) and reducing
net exports. Thus, any increase in the interest rate reduces not only con¬
sumption and investment but also net exports. These three reductions coun¬
teract the increase in government purchases. The result is called crowding
out in the open economy and is most severe when exchange rates are flexible.
If exchange rates are fixed, net exports will not decline because of a higher
exchange rate, and crowding out is less severe.
Thus, a country must choose between fixed and flexible exchange rates,
608 C HAPTER 17 O P E N E C O N O M Y M A C R O E C O N O MI C S
and this choice will determine whether monetary or fiscal policy is more
effective. Under flexible exchange rates, monetary policy can be effective at
changing aggregate demand, while the effects of fiscal policy are diminished.
Under fixed exchange rates, monetary policy is tied to exchange rate man¬
agement and is not available for use in pursuing other policy goals. Fiscal
policy, however, is less subject to crowding out and therefore more effective.
Thus, the choice of fixed or flexible exchange rates has important implica¬
tions not only for international traders and financiers themselves but also for
the effectiveness of various macroeconomic policy tools. The United States
has chosen to have a flexible exchange rate for a number of reasons, one
being the additional flexibility it gives to the use of monetary policy. Box
17.3 describes the choice between fixed and flexible exchange rates in several
European countries.
Fiscal Policy in an Open Economy
As we discussed in Chapter 16, fiscal policy actions are directed by the

government budget constraint: The government must obtain funds to finance
any government purchases. In an open economy, the only difference is that
the sources of funds have expanded to include both taxes on international
trade and international borrowing. However, we can still classify the methods
of financing government purchases into three categories: (1) Raise tax rev¬
enues, including revenues from tariffs; (2) borrow funds by selling govern¬
ment bonds to U.S. or foreign residents; or (3) print money. The method of
financing can affect the exchange rate; the interest rate, the price level, and
real output. We now look at how different methods of financing government
purchases affects the open economy.
Tax- and Tariff-Financed Increases

in Government Purchases
Taxes available to the government include taxes such as the income tax and
the social security tax and taxes such as tariffs on imports. We considered
government purchases financed by an income tax in Chapter 16. Now we
look at an increase in government purchases financed solely by a rise in
tariffs. Such a consideration would have been very realistic in the early days
of the United States, when tariffs were a major source of revenue and income
taxes were unconstitutional. If an increase in government purchases is
matched by an equal increase in tariffs, the government does not borrow any
additional funds or print additional new money. We can analyze this situation
as a simultaneous increase in government purchases, G, and in tariffs. The
increase in government purchases has the direct effect of raising aggregate
demand. The increase in tariffs raises the costs of imports, reducing demand
for imports and raising net export demand. Thus, a tariff increase raises both
net exports and aggregate demand! This is quite the opposite of the increase
Fixed Versus Flexible Exchange Rates:

The European Experience
The European Economic Community (EC) has the U.S. Federal Reserve System in the early
been moving toward greater economic, political, 1980s to reduce the U.S. inflation rate. But
and monetary union for more than two dec¬ when Germany raised its interest rates, other
ades. During this time, many plans have been ERM nations had to respond by raising their in¬
proposed for establishing a common currency terest rates. If they didn't, the higher German
for Europe or, more accurately, for the EC mem¬ rates would lead to capital inflows to Germany,
bers. In 1971, a plan called the Werner Report and, as we saw in Chapter 11, the increased de¬
called for a common currency by 1980. Most re¬ mand for German marks would cause the mark
cently, the 1991 Maastricht treaty calls for a to appreciate and the other currencies to depre¬
common currency by as early as 1997 and no ciate, thus violating the ERM bands. The fixed
later than 1999. As a step in this direction, the exchange rate system hence transmitted Ger¬
European Monetary System has had in place an many's restrictive monetary policy to the rest of
exchange rate system known as the exchange the EC. Moreover, other countries, such as
rate mechanism (ERM) since March 1979. In this France and the United Kingdom, found them¬
system, members of the European Monetary selves unable to respond to domestic macroeco¬
System pledged to maintain exchange rates nomic events with monetary policy, since they
within relatively narrow bands of target levels. were forced to maintain high interest rates—
At first, there were many revaluations of the tar¬ which reduced investment and consumption—to
gets as the various members adjusted to the maintain the exchange rate system. The problem
ERM, including a series of major realignments in was exacerbated by a worldwide recession in
the early 1980s and again in 1986 and 1987. 1991 and 1992. Late in the summer of 1992,
From then until late in the summer of 1992, the the United Kingdom pulled out of the ERM, re¬
ERM was fairly stable, with most member coun¬ fusing to suffer further under high interest rates
tries' currencies keeping within narrow bands of to maintain the exchange rate with the German
±2.25 percent around the target (or central) mark. Then, in the summer of 1993, the bottom
rates and others within ±6 percent bands. The fell out: The currencies of many countries were
United Kingdom, long a holdout from the ERM, near the bottom of their bands against the
finally joined in 1990. mark. The European Monetary System's re¬
However, in 1992 the effects of German re¬ sponse was to widen the bands to ± 15 per¬
unification were being felt in Germany, which cent, which effectively ended, at least temporar¬
had experienced large budget deficits and infla¬ ily, the fixed exchange rates of the ERM.
tion. The German central bank, the Bundesbank, What had gone wrong? Basically the world¬
responded by restricting money growth, which wide recession put pressure on European gov¬
caused interest rates to rise. This is the time- ernments to lower interest rates at the same
honored response to inflation and was used by Continued on p. 610
Continued from p. 609 Will the European Monetary System return

time Germany was raising its rates to counter in¬ to the tight ERM bands and to its goal of even¬
flation caused by increased government spend¬ tual monetary union? The answer is not clear. It
ing due to reunification. Germany was unwilling will take some time to move back to ERM bands
to lower its interest rates, and other ERM mem¬ of 2 or 3 percent, and the goal of monetary
bers were unwilling to maintain high enough in¬ union by 1997 is extremely unlikely to be
terest rates to keep the exchange rates within achieved. Whether this experience will dampen
the agreed-upon bands. This example illustrates the enthusiasm for eventual monetary union re¬
the cost of fixed exchange rates: They deprive a mains to be seen.
country of the ability to use monetary policy to
achieve domestic policy goals.
in the consumption tax we analyzed in Chapter 16. But in an open economy,

an increase in tariffs on imported inputs also might reduce aggregate supply.
In this case, the increase in aggregate demand will be offset to some extent
by the decrease in aggregate supply.
Figure 17.3 illustrates a tariff-financed increase in government purchases
for the case where the tariffs do not affect aggregate supply. This case is
most likely if the tariffs are imposed on consumer goods but not on imported
inputs. Point A is the initial equilibrium. An increase in government pur¬
chases raises aggregate demand from AD0 to ADX, and the increased tariffs
also raise it, from AD{ to AD2. In this case, the short-run equilibrium moves
from point A to point C, illustrating a short-run rise in real GDP and prices.
In the long run, however, wages and other input prices adjust, and the
economy moves to point D. The long-run effect of a tariff that does not
affect aggregate supply is to raise the domestic price level but leave real
GDP at its natural rate.
Figure 17.4 illustrates the effect of an increase in government purchases
financed by tariffs imposed on all foreign goods, including inputs. In this
case the initial equilibrium is at point A, at the intersection of AD0 and
SRAS0. The increase in government purchases shifts aggregate demand to
the right to AD{, and the increase in tariffs raises net exports, which further
increases aggregate demand to AD2. But the increase in tariffs on inputs
raises production costs and reduces short-run aggregate supply from SRAS0
to SRASi, and long-run aggregate supply declines from LRAS0 to LRAS
The short-run effect is that the equilibrium moves to point B, with a higher
price level and a higher level of real GDP than at point A. (Of course, this
depends on the magnitude of the shifts in aggregate demand and short-run
aggregate supply; real GDP could fall even in the short run.) In the long run,
the adverse effects are compounded as the economy moves to point C. Prices
rise even more, and output falls below the initial level at point A. Thus,
tariffs on inputs reduce the natural rate of output; tariff-financed government

purchases result in lower long-run GDP. This raises a dilemma for policy¬
makers, particularly politicians: The short-run temptation is often to stimu¬
late the economy now by imposing such tariffs, but the long-run effect will
be a decline in real output.
Government Purchases Financed by

Borrowing: Deficit Finance
What happens when the government borrows to finance spending in an open
economy? As we saw in Chapter 16, the United States borrowed large
amounts to finance government purchases in the early 1980s. How does our
analysis of this policy change when we take account of the open economy?
When government purchases rise, aggregate demand increases. There is no
increase in taxes to reduce consumption. However, there is an increase in
government borrowing, which results in crowding out. In the loanable funds
market, the rise in government borrowing is an increased demand for loan-
able funds, which leads to an increase in the interest rate. This reduces
investment and consumption just as we saw in the closed economy in Chapter
16. Furthermore, in an open economy, the increased demand for loanable
funds is met in part by foreign lending. But the increase in the U.S. interest
rate attracts foreign investors, who must first obtain additional U.S. dollars
to purchase U.S. assets. This drives up the exchange rate, leading to a
reduction in net exports and a worsening of the current account balance.
Thus, the rise in aggregate demand due to the increase in government pur¬
chases is countered by two effects in an open economy: (1) an interest-rate-
induced decrease in investment, consumption, and net exports and (2) a
decline in net exports due to a higher exchange rate. This is crowding out
in the open economy.
The case of partial crowding out is illustrated in Figure 17.5, where the
increase in government purchases shifts aggregate demand from AD0 to ADU
but the increase in the interest rate from government borrowing and the rise
in the exchange rate jointly cause a reduction in aggregate demand to AD2.
In this case, the short-run effect of an increase in government purchases
financed by borrowing is the movement from point A to point C, where the

price level and real GDP rise. In the long run, wages and other input prices
increase, which moves us to point D. At point D, output returns to its natural
rate, but the price level is higher.
It is important to remember that crowding out in an open economy occurs
because of increases in both interest rates and the exchange rate. In contrast,
in a closed economy crowding out occurs due only to rising interest rates.
The magnitude of the crowding out depends on whether exchange rates are
fixed or flexible. Figure 17.5 is drawn for the case of a flexible exchange
rate, in which the increased exchange rate reduces net exports and further
Figure 17.5
Crowding Out in the Open Economy
mtMim i urn I II in
This graph shows the effect of a deficit-financed increase in government purchases in an

open economy. The increase in government purchases causes aggregate demand to rise
from AD0 to ADV However, the rise in the interest rate that accompanies the additional
government borrowing will have a negative impact on aggregate demand, both because it
reduces consumption and investment and because higher interest rates increase the
exchange rate, which lowers net exports. These adverse effects reduce aggregate demand
from ADi to AD2. The short-run equilibrium moves from point A to point C, while the
long-run equilibrium moves from point A to point D.
crowds out the increase in government purchases. If the exchange rate were
fixed, the Fed would intervene in the exchange market by supplying U.S.
dollars to keep the exchange rate, and thus net exports, unchanged.
Government Purchases Financed by Money Creation

Just as in the closed economy, one of the most expansionary ways to increase
government purchases is to finance the spending by money creation. How¬
ever, this method of finance is available only in the case of flexible exchange
rates. With fixed exchange rates, the money supply cannot be increased to
finance government purchases because doing so would lower the exchange
rate. Thus, the analysis in this section pertains only to a regime of flexible
exchange rates, as currently exists in the United States.
Government purchases financed by money creation has several effects
on the economy. First, the increase in government purchases raises aggregate
demand directly. Second, the increase in the money supply reduces the
domestic interest rate. This increases domestic consumption and investment,
further raising aggregate demand. Finally, the lower domestic interest rates
reduce foreign demand for U.S. assets, which in turn reduces their demand
for U.S. dollars. The result is a decrease in the exchange rate, making
domestic goods cheaper relative to foreign goods, which raises net exports
and improves the balance of trade. Thus, all three effects tend to increase
aggregate demand.
Figure 17.6 illustrates these effects. The initial equilibrium is at point A,
with aggregate demand given by AD0. The increase in government purchases
raises aggregate demand to ADX. The increase in the money supply lowers
interest rates and the exchange rate, which increases aggregate demand to
AD2. Thus, the short-run effect of government purchases financed by money
creation is the movement from point A to point C, which implies a higher
price level and higher short-run real GDP. In the long run input prices rise,
and equilibrium moves to point D. Thus, the long-run effect is merely to
raise the price level.
The Twin Deficits: The Budget

and Current Account Deficits
During the 1980s, the United States had large government budget deficits,
accompanied by large current account deficits. These deficits were labeled
the twin deficits to indicate that they seemed to move together much of the
time. These twin deficits are graphed in Figure 17.7, where we see that they
both increased greatly from 1980 to 1985. From 1985 through 1989 there
was a trend toward a reduction in both deficits, but in the 1990s the govern¬
ment budget deficit again increased sharply while the current account con-
tinued to improve. In this section, we explore the relationship among in¬

vestment, saving, the government budget deficit, and the current account
deficit and see why these twin deficits have moved together in recent years.
Investment, Saving, and the Budget Deficit

In an open economy, total spending on domestic output, Y, can be divided
into consumption purchases (C), investment purchases (/), government pur¬
chases (G), and net exports (NX). Thus, we have the following relationship
between total output and the various spending categories:
Y = C + I + G + NX. (17.1)
Equation 17.1 provides one relationship between economic output and
spending. Now consider individuals in the economy. Individuals receive
income that equals the value of total output, Y, and they dispose of that
income by buying consumption goods, C, by saving it, S, and paying taxes
with it, T. Thus, we also have the relationship
Y = C + S + T. (17.2)
Substituting Y in equation 17.2 for Y in equation 17.1 gives us

C + S+ T= C + I+ G +NX,
which can be simplified to
I + (G - T) = S - NX. (17.3)
Equation 17.3 gives a relationship stating that investment spending (/) plus
the government deficit (G — T) must equal private sector saving (S) minus
net exports (NX). At first this relationship may look a bit strange, but in a
closed economy, where net exports are zero, this relationship merely states
that the demand for funds for investment and for financing the government
deficit must equal the supply of funds available from private sector saving.
In a closed economy, any increase in investment spending or in the govern¬
ment deficit must be matched by an increase in private saving. The left-hand
side of equation 17.3 is essentially the demand for loanable funds, and the
right-hand side is the supply of loanable funds, which in a closed economy
is simply private sector saving.
In an open economy, equation 17.3 subtracts net exports from private
sector saving. If we ignore income flows and unilateral transfers, net exports
are essentially the current account in the balance of payments. Furthermore,
the current account equals the opposite of the capital account. Thus, ignoring
income flows, unilateral transfers, and statistical discrepancies, net exports

indicate the size and magnitude of the capital account. If net exports are
— $100 billion, the capital account must be + $100 billion. In effect, negative
net exports—a current account deficit—are an outflow of funds to purchase
goods and services produced abroad. This outflow is balanced by a capital
account surplus, an equal-sized inflow of funds from abroad. This is an
inflow of loanable funds used to finance both private investment and the
government deficit. For given levels of private saving and investment, the
greater the government deficit, the greater the amount of loanable funds
needed to finance it and the greater the current account deficit. In an open
economy, a current account deficit provides an additional source of loanable
funds to the U.S. financial markets.
The Supply of Loanable Funds in an Open Economy

There is another way to look at the relationship between a current account
deficit and the supply of loanable funds in the United States. Negative net
exports means the United States is buying more goods from abroad than it
sells abroad. There will be a net outflow of dollars from the United States,
which will accumulate in the hands of overseas producers. These producers
do not want to hold idle dollar balances, so they use these accumulated funds
to purchase U.S. assets; that is, they trade them to U.S. borrowers in return
for stocks, bonds, or other assets. In essence, they supply loanable funds in
the U.S. loanable funds market.
Figure 17.8 provides a graphical view of the difference between an open
and a closed economy in the loanable funds market. The graph explicitly
indicates that the demand for loanable funds, DLF, is to finance investment
(/) and the government deficit (G — T). The supply of loanable funds in
the closed economy (S\f ) is simply private domestic saving, S. The supply
of loanable funds in the open economy is private domestic saving plus the
supply of foreign funds as indicated by the balance on the capital account.
Since the negative of the current account, or — NX, equals the capital account
balance, the total supply of loanable funds to the United States in the open
economy is S — NX, or S\F.
Loanable funds equilibrium in the closed economy is at point A at the
intersection of SFF and DLF, whereas equilibrium in the open economy is at
point B, at the intersection of S\F and DLF. Notice that when there is a surplus
on the capital account (or, equivalently, a deficit on the current account), net
exports are negative. In this case additional loanable funds enter the United
States from abroad, and the interest rate in the open economy (iB) is below
that in the closed economy (iA).
International trade also has a more subtle effect on the total supply of
loanable funds. Because both the private saving schedule and the supply of
foreign funds are sensitive to interest rate movements, the slope of loanable
funds supply in the closed economy is steeper than that in the open economy.
V
Figure 17.8
The Loanable Funds Market in an Open Economy
wmmM,
In a closed economy the supply of loanable funds is private saving, 5, whereas in an open
economy the supply is private saving, 5, plus the foreign supply of loanable funds. Equilib¬
rium in a closed economy is thus at point A where the interest rate is iA. In an economy
such as the United States, where net exports are negative, equilibrium is at point B. Thus,
interest rates are lower because the capital account surplus provides an additional supply
of loanable funds. The foreign supply is given by the capital account, which is -NX in the
absence of central bank intervention. Hence, in the open economy, the supply of loanable
funds is S - NX.
This means any increase in the U.S. demand for loanable funds will result
in a larger increase in the interest rate in the closed economy than in the
open economy. This implies, among other things, that a government deficit
financed by borrowing will result in a smaller increase in the interest rate in
the open economy, because foreign sources of loanable funds will help meet
the increased demand for loanable funds. The flip side of this coin is that
increases in the government deficit will result in a greater surplus on the
capital account and hence a greater deficit on the current account, or greater
negative net exports.
Explaining the Movement in the Twin Deficits

What can explain the close movements in the budget deficit (G — T) and
the current account deficit ( — NX) shown earlier in Figure 17.1? Many have
argued that the high government budget deficit in the 1980s caused the high
current account deficit during those years. Equation 17.3 does suggest that
the budget deficit and the capital account deficit will move together, but only
if both investment and private saving are held fixed, or at least if the differ¬
ence between investment and private saving fails to adjust to the government
budget deficit. For example, if we hold investment and private saving con¬
stant, any increase in the U.S. government budget deficit must be matched
by a reduction in net exports (an increased deficit position in the current
account). Note, however, that the validity of this argument depends on the
crucial assumption that private saving and investment do not adjust to
counter the government budget deficit.
In reality, domestic investment and private saving were not constant in
the early 1980s, and in fact investment grew almost twice as fast during
this period as private saving. This increase in real domestic investment
was caused in part by drastic changes in tax laws that created a more
favorable climate for investing in the United States. Domestic private sav¬
ing did not keep up with investment; instead the funds required to finance
this increase in investment came from abroad. To obtain dollars to lend in
the United States, overseas producers increased their sales of goods in the
United States, which increased the current account deficit. Concurrently,
government spending in the 1980s increased at a faster rate than tax revenues
(due largely to changes in tax laws), which led to an increase in the budget
deficit.
Thus, we see that the high current account deficits in the early 1980s
were caused by two factors: an increase in the government deficit and a rise
in domestic investment relative to domestic private saving. Tax changes in
the early 1980s contributed to both of these factors. Hence it is probably
misleading to say that the increased government budget deficits were the
sole cause of the current account deficits. Instead government budget deficits
played a role in the rise of the current account deficits, a role shared by the
rise in investment demand.
Monetary Policy: Open Market Operations

What is the effect of open market operations in an open economy? With
fixed exchange rates, monetary policy must focus on keeping the exchange
rate fixed; the Fed supplies U.S. dollars when the demand for dollars is high
and buys U.S. dollars when demand for dollars is low. Such monetary policy
620 Chapter 1 7 Open Economy Macroeconomics
keeps the exchange rate constant, or at least within some narrow band. This
means, however, that the Fed cannot freely use open market operations as a
means of stimulating the domestic economy.
The story differs with flexible exchange rates. In this case, the exchange
rate is determined by market forces, and the Fed does not intervene to keep
it from rising or falling. In other words, with flexible exchange rates, the
Fed can use monetary policy to affect the domestic economy. Let us see
how this works in an open economy.
In an open market purchase, the Federal Reserve System prints money
and uses it to purchase bonds. This lowers the interest rate. The effect of
open market operations on aggregate demand then follows from the change
in the interest rate. The decrease in the interest rate leads to an increase in
domestic consumption and investment and hence a rise in aggregate demand.
Furthermore, in an open economy the reduction in the interest rate leads to
a decrease in the foreign demand for U.S. assets and hence in the foreign
demand for dollars on the market for foreign exchange. This tends to lower
the exchange rate, making U.S. goods cheaper relative to foreign goods.
This effect increases net exports (and improves the balance of trade), which
further raises aggregate demand. Thus, we see that in an open economy,
monetary policy has an effect on net exports, which tends to reinforce the
effects present in a closed economy.
Figure 17.9 illustrates this effect. Aggregate demand is initially at AD0,
and the initial equilibrium is at point A. The open market purchase lowers
the interest rate and the exchange rate, shifting aggregate demand to ADX
and moving the equilibrium to point B in the short run, increasing both the
price level and level of real GDP. Keynesians view the shift to point B as
relatively small, whereas monetarists see it as more pronounced. In the long
run, the economy moves to point C, where the price level is higher and real
output returns to its natural rate.
Summary | (a) Suppose the government reduces spending and finances the spending by
Exercise 17.4 I issuing fewer bonds. What will be the short-run and long-run effects of this
policy on the interest rate, the exchange rate, the price level, real GDP, and
the balance of trade? (b) How will the impact change if the government
funds the spending cuts by decreasing the money supply?
Answer: (a) A reduction in government spending leads to a decrease in

the demand for loanable funds, which in turn lowers the interest rate. This
decrease in the interest rate increases the level of investment and consump¬
tion in the economy, which tends to shift aggregate demand to the right. In
addition, the decrease in the interest rate lowers the exchange rate and tends
to increase net exports (and the balance of trade), which further shifts ag¬
gregate demand to the right. The decrease in government purchases, how-
ever, shifts aggregate demand to the left. If there is complete crowding out,
these effects perfectly cancel each other out, and no change in the price level
or aggregate output occurs, although the interest rate and exchange rate fall.
If there is only partial crowding out, the decrease in government purchases
will more than offset the increase in consumption, investment, and net ex¬
ports, so aggregate demand will decrease. In the short run, this will lead to
a lower real GDP and price level. In the long run, however, wages and other
input prices will fall, shifting aggregate supply to the right. In the long run,
the policy has no effect on real output; it only lowers the price level, (b) A
decrease in the money supply increases domestic interest rates and thus the
exchange rate. This leads to a reduction in consumption, investment, and net
export demand (worsening the balance of trade). The decrease in government
purchases further lowers aggregate demand, so net aggregate demand shifts
to the left. The short-run effect of this policy is to decrease the price level
and real GDP. In the long run, wages and other input prices fall, leading to
a lower price level and the original level of output.
Conclusion
In this chapter, we explored the implications of international trade and

international finance for the macro economy. We saw how features of the
open economy modify our analysis of Chapter 16 and alter the effectiveness
of monetary and fiscal policy. In the final three chapters of this book, we
build on these basic models to look more closely at monetary and fiscal
policy. But first, in Chapter 18, we look at the data illustrating the relation¬
ship between money and the economy.
key terms
open economy model statistical discrepancy
balance of payments current account
import net exports
export capital account

1. Suppose a U.S. resident buys a case of 4. Determine the impact of the following on
French wine for $1,000. The owner of the the demand for net exports:
French vineyard uses this money to buy (a) . An increase in the U.S. interest rate
$1,000 worth of IBM stock. (b) . An increase in the foreign price level
(a) . How do these transactions show up in (c) . An increased taste for U.S. goods
the balance of payments? sold in Japan
(b) . If these are the only transactions in (d) . A decrease in the exchange rate
the economy, will there be a balance of
5. Determine the impact of the following on
payments deficit? A balance of trade
the demand for net exports:
deficit?
(a) . An increase in both the U.S. and for¬
eign interest rates
2. Explain why net exports can be positive
(b) . An increase in both U.S. and foreign
or negative and why a higher (U.S.) price
wealth
level results in a reduction in the quantity
demanded of net exports. 6. Determine the impact of the following on
aggregate demand:
3. Determine the impact of the following on (a) . An increase in the foreign interest
the demand for imports: rate
(a) . An increase in the U.S. price level (b) . A decrease in the money supply
(b) . An increase in the foreign price level (c) . An increase in tariffs on imported
(c) . A decrease in the exchange rate consumer goods
(d) . Improved consumer optimism tariffs. What are the short-run and long-
(e) . Technological improvement run macroeconomic effects assuming the
(f) . A tariff on imported raw materials tariff increase (a) affects only consump¬
for industry tion, (b) affects only aggregate supply,
and (c) affects both aggregate demand and
7. Holding everything else constant, deter¬
aggregate supply?
mine the short-run impact of the following
on the equilibrium price level and real 11. Suppose the government increases spend¬
GDP: ing and finances the spending by borrow¬
(a) . An increase in the exchange rate ing. What are the short-run and long-run
(b) . A decrease in the foreign interest rate macroeconomic effects on (a) the interest
(c) . An increase in U.S. tariffs on im¬ rate, (b) the exchange rate, (c) the price
ported household goods level, and (d) real GDP?
(d) . Technological improvement
12. Suppose the government increases spend¬
(e) . A tariff on imported raw materials
ing and finances the spending by printing
8. Explain why the exchange rate might af¬ money. What are the short-run and long-
fect the position of short-run aggregate run macroeconomic effects on (a) the in¬
supply. Under what conditions will terest rate, (b) the exchange rate, (c) the
changes in the exchange rate not affect price level, and (d) real GDP?
short-run aggregate supply?
13. Assume the U.S. government is concerned
9. Explain why the tariffs on imported about the fact that net exports in the
raw materials will affect the positions of United States are negative. What types of
both short-run and long-run aggregate policies could improve the balance of
supply. trade?
10. Suppose the government increases spend¬

ing and finances the spending by raising
Ahmed, S., B. Ickes, P. Wang, and S. Yoo. “Inter¬ Economics and Statistics, 71 (May 1989), 337-
national Business Cycles.” American Economic 341.
Review, 83 (June 1993), 335-359. Himarios, D. “The Impact of the Exchange Rate on
Beladi, H., B. Biswas, and G. Tribedy. “Growth of U.S. Inflation and GNP Growth: Comment.”
Income and the Balance of Payments: Keynesian Southern Economic Journal, 55 (April 1989),
and Monetary Theories.” Journal of Economic 1044-1051.
Studies, 13 (1986), 44-45. Honohan, P., and P. McNelis. “Is the EMS a DM
Davila, A. E., and J. P. Mattila. “Do Workers Earn Zone?—Evidence from the Realignments.” Eco¬
Less Along the U.S.-Mexico Border?” Social Sci¬ nomic and Social Review, 20 (January 1989), 97-
ence Quarterly, 66 (June 1985), 310-318. 110.
Deyak, T. A., W. C. Sawyer, and R. L. Sprinkle. “An Kim, B. J. C. “A Time-Series Study of the Employ¬
Empirical Examination of the Structural Stability ment Real Wage Relationship: An International
of Disaggregated U.S. Import Demand.” Review of Continued on p . 624
.
Continued from p 623 Reichenstein, W. R., and F. J. Bonello. “Aggregate
Comparison.” Journal of Economics and Business, 40 Supply Considerations and the St. Louis Equa¬
(February 1988), 67-78. tion.” Journal of Economics and Business, 34
Lowinger, T. C., C. Wihlborg, and E. S. Willman. (1982), 253-262.
‘‘OPEC in World Financial Markets: Oil Prices Romans, J. T., and S. A. Warren. “A Balance of Pay¬
and Interest Rates.” Journal of International ments Analysis of the Latin American Debt Cri¬
Money and Finance, 4 (June 1985), 253-266. sis.” Review of Income and Wealth, 36 (June
Ozawa, T. “Japan in a New Phase of Multinational¬ 1990), 207-213.
ism and Industrial Upgrading: Functional Integra¬
tion of Trade, Growth and FDI.” Journal of World
Trade, 25 (February 1991), 43-60.
CHAPTER
Money and Economic Activity:

A Look at the Evidence
magine that you work for a major Wall Street investment firm. Your
boss calls you in and asks you about the likely impact on the economy of a
recent rise in the money supply. How will you reply? The aggregate demand
and supply model we developed in Chapters 16 and 17 suggests that in the
short run (when input prices are fixed), an increase in the money supply may
cause real output (real GDP) and the price level to rise. However, the model
also suggests that in the long run (after input prices fully adjust), increases
in the money stock will not change real output; it will only raise the price
level. In the long run, an economy cannot produce more real output simply
by printing more money.
These results are only theoretical, however, and you may wonder
whether your boss would understand the diagrams you might use to illustrate
these effects. You may even wonder yourself whether these theoretical ideas
line up with the real-world facts. This chapter provides the answer as we
look at the data on money and other economic variables to see the empirical
relationship between money and such variables as real output, inflation, and
interest rates.
The Long-Run Relationship among

Money, Prices, and Real Output
Our model of aggregate demand and supply makes predictions about the
relationship between money and other economic variables for both the long
run and the short run. In this section, we examine three propositions that
summarize the predictions about the long-run relationship among money,
output, and prices. Then we look at some empirical evidence on these rela¬
tionships, from both the United States and abroad, to see how well they hold
up in reality.
Theoretical Relationships Between Money

and Economic Activity
Recall from Chapters 16 and 17 that our long-run macroeconomic model
has a vertical aggregate supply curve. This indicates that in the long run,
625
626 Chapter18 Money and Economic Activity: A Look at the Evidence
after all input prices fully adjust to changes in the price level, changes in
aggregate demand do not affect output since any increase or decrease in
aggregate demand will merely result in movements up or down the vertical
long-run aggregate supply curve. Changes in aggregate demand do affect
the price level in the long run. Recall too that one important determinant of
aggregate demand is the money stock. Changes in the money stock cause
changes in aggregate demand, which in turn cause changes in the price level
or output in the short run and changes in the price level in the long run. In
this section, we elaborate on the relationship among money, prices, real
output, and nominal GDP in the long run.
To begin, remember that aggregate demand is the sum of consumption
demand, investment demand, government demand, and net export demand;
the determinants of aggregate demand are simply the determinants of these
components. Thus, the determinants of aggregate demand include wealth,
the interest rate, taxes and tariffs, and the exchange rate. Anything that causes
these determinants to change causes changes in aggregate demand, and thus
changes in the price level or in real output.
Concentrating on the long run, we want to consider which of these
determinants might be responsible for long-run movements in aggregate
demand, that is, which determinant could create increases or decreases in
aggregate demand that persist over years or even decades. A number of
possibilities exist, but a traditional answer to this question that extends back
to the classical economists such as Adam Smith and Alfred Marshall is that
ongoing changes in the money supply cause ongoing changes in aggregate
demand. In fact, this idea is formalized in the equation of exchange, first
discussed in Chapter 3:
MV = PY.
The left-hand side of this equation is the product of the money stock, M,
and the velocity of money, V (the number of times the average unit of money
changes hands in one year). The right-hand side is the product of the price
level, P, and real output (or real GDP), Y. The product of the price level and
real GDP is also known as nominal GDP, or simply GDP.
We can use the equation of exchange to develop the quantity theory of
money. As we saw in Chapter 3, this involves making assumptions about
the behavior of velocity and real GDP. In particular, we assumed velocity
to be fairly stable and changes in velocity were not related to changes in the
money stock. Using the quantity theory, we can develop a model of aggregate
demand that greatly simplifies the model developed in Chapters 16 and 17.
The quantity theory emphasizes the importance of the money stock, which
we recognized as being but one of a number of factors that could shift
aggregate demand. The quantity theory consolidates all other determinants
of aggregate demand into velocity so that changes in taxes, interest rates, or
tastes are all considered changes in velocity. The classical economists
thought these other determinants of demand are relatively stable and, more
important, were not likely to have sustained movements in one direction or
The Long-Run Relationship Among Money, Prices, and Real Output 627
another over long periods of time. Thus, over the long run these other
components may occasionally cause a change in aggregate demand, but they
would not continually increase or decrease year after year and thus would
not be responsible for ongoing movements in aggregate demand.
Changes in the money stock, on the other hand, do tend to be systematic;
governments regularly print more money over time. The equation of
exchange emphasizes the importance of such changes in the money stock.
It also allows us to easily discuss another idea: that increases in the money
stock lead to changes in nominal income (nominal GDP). For example, the
equation of exchange can be written in terms of growth rates as
8 m + 8v — ^nominal GDP — 77 + 8y-
This equation indicates that the growth rate of the money stock (gM, or AM/
M) plus the growth rate of velocity (gv, or A V/V) equals the growth rate of
nominal GDP (gn0minai gdp> or AGDP/GDP). Furthermore, it separates the
growth of nominal GDP into the inflation rate (it, or A P/P) and the growth
of real output (gr, or A Y/Y).
As we mentioned earlier, one assumption the classical economists made
was that the growth rate of velocity is fairly stable and is unaffected by
changes in the growth rate of the money stock. We can simplify this to say
that the growth rate of velocity is on average equal to zero. As we will see,
this is approximately true for the velocity of M2, although it is not true for
the velocity of Ml. This is one reason this chapter stresses looking at M2
instead of Ml: The velocity of M2 more closely corresponds to the classical
assumptions.
If velocity growth is zero, we have the classical relationship between
money growth and nominal income growth:
8m gnominal GDP ^ 8y-
This equation, along with the vertical long-run aggregate supply curve, gives
us a host of propositions about money growth, nominal income growth, real
output, and inflation. It links money growth (gM) to nominal income growth
(^nominal gdpX which itself is the sum of output growth (gF) and inflation (tt).
The vertical long-run aggregate supply curve means that changes in aggre¬
gate demand, including those caused by changes in the money stock, cannot
lead to changes in real output in the long run. Thus, increases in money
growth in the long run must lead to increases in the inflation rate. We
summarize the implications of these ideas in the following three propositions:
Proposition 1: In the long run, an increase in the growth rate of money
will lead to an equal increase in the growth rate of nominal GDP.
will lead to an equal increase in the inflation rate.
will not lead to any change in the growth rate of real GDP.
628 Chapter 18 Money and Economic Activity: A Look at the Evidence
We stress that these are long-run propositions. They need not apply to
the short run. In the short run, aggregate supply is not vertical, so proposi¬
tions 2 and 3 need not hold. Furthermore, in the short run the other deter¬
minants of aggregate demand continually lead to rises or falls in velocity,
and these results detract from the relationship specified in proposition 1. We
will discuss the short-run relationships between money and economic activity
later. For now, we look at the evidence for these long-run propositions. We
first look at evidence in the United States and then see whether or not the
experience in other countries is similar.
U.S. Evidence on the Long-Run Relationships

Between Money and Economic Activity
To investigate our three propositions, we first look at the evidence from 120
years of U.S. history on the impact of changes in the growth rate of the
money stock on the inflation rate, the growth rate of real GDP, and the
growth rate in nominal GDP for the 1869-1989 period. We divide these data
into 10-year periods. For each period, we calculate the average growth in
the money supply and compare it to the corresponding decade average
inflation rate, nominal GDP growth rate, and real GDP growth rate. This
allows us to see whether money growth over a 10-year period affected
economic activity over the same decade. We will use a technique that allows
us to visualize the relationship between money growth rates in these different
decades to the growth in the price level, nominal GDP, and real GDP over
the same period.
The first proposition states that increases in the growth rate of the money
stock should lead to equal increases in the growth of nominal GDP. Is this
proposition confirmed by 120 years of U.S. history? Part a of Figure 18.1(a)
shows the growth rates of nominal GDP and the M2 money stock for each
of 12 decades ranging from 1869-1879 to 1979-1989. This plot of points is
called a scatter diagram; each point corresponds with the growth in nominal
GDP and money growth in a given decade. This scatter diagram reveals that
during decades in which money growth was high, the growth in nominal
GDP also tended to be high. This is consistent with proposition 1: An
increase in the growth rate of money leads to an increase in the growth rate
of nominal GDP. The line drawn through these points is called a best fit
line; it is chosen by a statistical procedure to fit the best straight line to the
scatter of points. The slope of this line is 1, indicating that increases in the
rate of money growth over these decades tended to lead to an equal rise in
the rate of nominal GDP growth. Clearly the points do not lie exactly on the
line, so proposition 1 did not hold exactly for every decade. Instead, the
figure indicates that over time, increases in money growth tended to raise
nominal GDP growth by an equal amount.
There are two rather large deviations from the best fit line in Figure
18.1. The one that lies farthest above this line has nominal GDP growth of
6.56 percent and money growth of 3.74 percent and represents the decade
Sources: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, various issues; U.S. Department of
Commerce, Bureau of Economic Analysis, Survey of Current Business, various issues; Citibase electronic database.
1949-1959. The one farthest below the line has nominal GDP growth of
2.99 percent and money growth of 8.00 percent and represents the decade
1879-1889. Remember, however, that the proposition concerns the long run,
and over our sample of 120 years of U.S. history, the scatter of points in
Figure 18.1 indicates that in the long run, increases in the growth rate of
money led to increases in the growth of nominal GDP.
The second proposition states that increases in the growth rate of the
money stock should lead to equal increases in the inflation rate. This prop¬
osition builds on proposition 1 and on the idea of a vertical long-run aggre¬
gate supply curve. Figure 18.2 shows the inflation rate and the growth rate
of the money stock in the United States for 12 decades, ranging from 1869—
1879 to 1979-1989. This scatter of points is consistent with proposition 2,
since decades with higher growth rates of money tend to have higher inflation
rates.
The line drawn through the scatter of points is again the best fit line,
and the slope of this line is about .9, which is not statistically different from
1. This suggests that increases in the rate of money growth over these decades
led to equal or nearly equal increases in the rate of inflation.1 As in Figure
The appendix to this chapter contains the regressions used to estimate the best fit lines.
Each point on this V

Figure
'
18.2
scatter diagram repre¬ Long-Run Money Growth and Long-Run Inflation
sents the growth in
the money supply
(M2) and the inflation 7.5 - * /
rate in a given decade * S*
between 1869 and * //
1989 for the United 5.0 -
States. The slope of

the best fit line is 0.9, * sk
— 2 5-
which is not statisti- CD
+->
/ *
ro
cally different from 1. cc
This implies that in the o 0.0 -

yY *
4—>
long run, increases in _ro
■k /
4-
*
k
the growth of money c
led to equal increases -2.5 -
in the inflation rate.

Thus, U.S. data sup- r r\
-5.U ^ 1 1 1 1 T
port proposition 2.
0.0 2.5 5.0 7.5 10.0 12.5
Growth in M2 (%)
18.1, it is clear that the points in Figure 2 do not lie exactly on the line, so
proposition 2 did not hold exactly for every decade. Instead, over time
increases in the rate of money growth caused equal or nearly equal increases
in the inflation rate, with possible substantial deviations in any one decade.
An interesting feature of the relationship between inflation and money
growth is that the best fit line does not pass through the point with zero
inflation and zero money growth. This is because the equation of exchange,
in growth rate form, links the rate of money growth to the rate of inflation
plus the rate of output growth. Since output growth averaged more than 3
percent during 1869-1989, the inflation rate was on average about 3 percent
less than the rate of money growth. Of course, output growth varied consid¬
erably over our sample due to changes in the capital stock and technology
and to periods of recession, especially the Great Depression of the 1930s.
The third proposition states that increases in the growth rate of the money
stock should have no effect on the growth rate of output, which we measure
as growth in real GDP. Like proposition 2, this proposition is derived from
proposition 1 and the idea of a vertical long-run aggregate supply curve. In
the case of real GDP, the vertical long-run aggregate supply curve means
that in the long run, real GDP was unaffected by changes in determinants
of aggregate demand in general and by the money supply in particular.
Figure 18.3 graphs the rate of growth of real GDP and the growth rate
of the money stock for 12 U.S. decades. This scatter of points is consistent
with the notion that increases in the growth rate of money do not lead to
increases in the rate of growth of real GDP. Indeed, except for the point at
the origin, the points actually show a slight negative relationship between
real GDP growth and money growth. We do not draw a best fit line because
the slope cannot be distinguished statistically from zero. We conclude that
proposition 3 held, since the figure shows no link between the rate of money
growth and the rate of growth of real GDP. As in Figures 18.1 and 18.2, it
is clear that the points do not lie on a straight line and that real GDP was
not constant over this period. Real GDP growth was almost zero during the
decade that included the Great Depression, but even leaving this point aside,
the real GDP growth rates varied from just over 2 percent to more than 5
percent during this period, with no systematic relationship to the rate of
money growth.
We conclude that U.S. evidence generally supports propositions 1
through 3. Data from 12 decades of U.S. history show that: first, decades
with higher money growth had higher growth in nominal GDP, and further¬
more this relationship was one to one. For example, decades with 5 percent
higher money growth on average had 5 percent higher growth in nominal
GDP. Second, decades with higher money growth rates tended to also have
higher inflation rates. Moreover, this relationship was one to one; decades
632 ChapterIS Money and Economic Activity: A Look at the Evidence
with 5 percent higher money growth generally had inflation rates that were
5 percent higher. Third, the growth rate of real GDP and the growth rate of
money were not related. Decades with high money growth had, on average,
the same growth in real GDP as did decades with lower money growth.
Again we stress that these are long-run relationships that did not hold in any
one year or, as we have seen, even in any one decade. Instead, they held
over more than a century of U.S. history.
International Evidence on the Long-Run Relationships

Before looking at short-run evidence on money and economic activity, we
will examine another type of long-run evidence on our three propositions.
Our previous evidence from the United States compared money growth and
measures of economic activity over 12 consecutive decades. An alternative
is to use data from a single timespan and gather them for several countries.
This method gives us data on a large number of different experiences of
money growth rates, all from a comparable period of time. The advantage
of this approach over looking at only one nation is that we can gather much
more information and examine all countries at the same point in time. A
disadvantage is that we must compare nations with different institutional
details and different values for velocity and the other determinants of aggre¬
gate demand. But in a way this is a plus, because our intent is to see whether
differences in money growth can explain differences in nominal income and
inflation rather than differences in real GDP growth.
Gerald P. Dwyer, Jr., and Rik W. Hafer conducted just such a study.
Using data from 62 countries gathered for the period 1979-1984, they em¬
ployed the scatter technique used in Figures 18.1 through 18.3. The differ¬
ence is that each point represents a different decade in U.S. history, while
each point in their figures represents the experience of a different country
over the same time span.
Recall that our first proposition is that increases in the growth rate of
the money stock should lead to equal increases in the growth rate of nominal
income. Figure 18.4 graphs the average growth rate of nominal GDP and
the average growth rate of the money stock for 62 countries for the time
period 1979-1984. The scatter of points is consistent with proposition 1.
The rates of money growth and GDP growth lie along the reference line
drawn with a slope of 1, as suggested by proposition one. It is not the best
fit line, which actually has a slope of 1.01. Thus, countries with higher
money growth rates tended to have higher inflation rates.
Notice in Figure 18.4 that there is quite a bit more variation in interna¬
tional money growth rates than we saw when looking at historical U.S. data.
In the 12 most recent decades of U.S. history, the highest rate of money
growth was less than 12 percent and the lowest was above 0 percent.
The 62 countries in Figure 18.4 showed a much wider range of money
growth rates, ranging from a bit more than 0 percent to 220 percent in
Source: Dwyer, Gerald P. and R. W. Hafer, “Is Money Irrelevant?” Federal Reserve Bank of St. Louis Review. Copyright
© 1988. Used with permission.
Bolivia. Likewise, they showed a much wider range of growth rates in

nominal GDP and more evidence on whether proposition 1 held throughout
this broader range. The answer is that it did seem to hold. The scatter plot
suggests that countries with a higher rate of money growth did on average
have higher rates of nominal GDP growth. It is also clear that, just as for
the United States, the points do not lie exactly on the line, so proposition 1
did not hold exactly for every country over these five years. Instead, the
figure indicates that those with 5 percent higher money growth tended to
have 5 percent higher nominal GDP growth.
Proposition 2 states that increases in the growth rate of the money stock
should lead to equal increases in the inflation rate. Figure 18.5 graphs the
inflation rate and the growth rate of the money stock for the same 62
countries over the same time span. Also drawn is a reference line with a
slope of 1. This scatter of points is consistent with proposition 2, since
countries with higher money growth rates tend to have higher inflation rates.
This gives rise to the upward-sloping best fit line, which has a slope of 1.03.
This indicates that increases in the rate of money growth and increases in
the inflation rate occurred essentially one for one. Again it is clear that the
points do not lie exactly on the line, so proposition 2 did not hold exactly
for every country. However, there is evidence supporting the idea that in-
Source: Dwyer, Gerald P, and R. W. Hafer, “Is Money Irrelevant?” Federal Reserve Bank of St. Louis Review. Copyright
© 1988. Used with permission.
creases in money growth cause equal increases in the inflation rate, as

proposition 2 claims. Notice too that the country with the highest inflation
rate (Bolivia, with 206 percent inflation) also had the highest growth in the
money supply (220 percent).
Finally, the third proposition states that increases in the growth rate of
the money stock have no effect on the growth rate of real GDP. Figure 18.6
graphs the rate of growth of real GDP and the growth rate of the money
stock for 62 countries. This scatter of points shows little relationship between
a country’s real GDP growth and its money growth. The reference line has
a slope of zero and is drawn at the average rate of growth of real GDP across
the 62 countries, a bit more than 2 percent. The best fit line (not shown) has
a slope of — .02, which is very close to zero. As in the previous figures, it
is clear that the points do not lie on a straight line and that real GDP growth
varied over the countries in the sample. It is also clear that the money growth
rate had little effect on growth in real GDP. Indeed, even nations with
very high money growth, such as Bolivia (220 percent) and Israel (152
percent) had rates of growth in real GDP that were similar to those of other
countries.
We can conclude that international evidence also generally supports
propositions 1 through 3. The data from 62 countries for the 1979-1984
period indicate that in the long run, first, countries with higher money growth
rates tended to have higher growth rates of nominal GDP, and this relation¬
ship was one to one. Second, countries with higher money growth rates
tended to have higher inflation rates. Finally, countries with higher money
growth rates experienced roughly the same growth rates in real GDP as those
with lower ones; money growth and real output growth were not related.
Again we stress that these are long-run relationships that did not hold pre¬
cisely for any one country but did hold on average for the countries in our
sample. Box 18.1 examines how well these relationships held for historical
data from the United Kingdom.
Summary Show how aggregate supply and demand analysis can explain the empirical
Exercise 18.1 relationships between money growth and growth in real GDP, the price
level, and nominal GDP seen over the most recent 12 decades in U.S. history.
Answer: Suppose the economy is initially in long-run equilibrium at point

A. For convenience, let us suppose the initial money stock is 200, the price
level is 100, and real GDP is 4. Now suppose that over a 10-year period,
the Fed increases the money supply by 10 percent to 220, which amounts to
an average of 1 percent per year. This increase in the money supply will
raise aggregate demand to ADU and in the long run the economy will move
Source: Dwyer, Gerald P. and R. W. Hafer, “Is Money Irrelevant?” Federal Reserve Bank of
St. Louis Review. Copyright © 1988. Used with permission.
,
Money Growth GDP Growth and Inflation ,
in the United Kingdom: 1
In the text, we examined evidence on our three Part a graphs nominal GDP growth against
propositions from a series of historical U.S. data money growth in the United Kingdom. We find
and also looked at the results from Dwyer and the expected positive relationship, but the slope
Hafer's study of 62 countries. This evidence of the best fit line is only .3, which is much less
strongly supports the three propositions. Here than one.
we look at evidence on these three propositions We find a similar result for the relationship
from the United Kingdom. Our data set consists Continued on p. 637
of observations from 1963 to 1991, a much
shorter series than the U.S. data.
Q_
o
CD
03 QJ
+-1
ro
(a) £ (b) CC
o c
O
c _tp
c
$
O
V)
CL.
Q
O
03
CU
(C) DC
0 5 10 15 20 25
Money Growth (%)
Continued from p. 636 fit line did not indicate that increases in money
between monetary growth and the inflation rate growth rates would show up one for one in
in part b. There is the expected positive relation¬ nominal GDP growth and inflation. However,
ship between money growth rates and inflation, since we also find that increased money growth
but the best fit line has a slope of only about .3 did not show up in an increase in real GDP
instead of unity. growth, we must conclude that velocity in the
Part c plots real GDP growth rates against United Kingdom was responding to money
money growth rates, and we see there is very growth rate changes ever this sample. Finally,
little relationship between these two variables. we note that the results from the United King¬
The best fit line is not shown, but it has a slope dom may be due to the short span of data used
of only .02, which is insignificantly different and that a longer series of data would be more
from zero. helpful in comparing the U.K. and U.S. experi¬
We conclude that the U.K. data provide less ences.
support for propositions 1 and 2, since the best
Source for data: International Monetary Fund, International Financial Statistics, various issues.
to a new equilibrium at point B. At B the price level is 110, which indicates

prices have increased by 10 percent over this 10-year period, or by 1 percent
per year. Thus, we see that in the long run, increasing the growth rate of
money by 1 percent raises the inflation rate by 1 percent. Real output remains
at 4, indicating that increased money growth has no long-term impact on
growth in real GDP. Finally, notice that nominal GDP at point A is 100 X
4 = 400, while at point B it is 110 X 4 = 440. Nominal GDP increased
by 10 percent over this period, or by 1 percent per year. This explains why,
in the long run, an increase in the growth of the money supply leads to an
increase in the growth of nominal GDP.
4 Real GDP (Trillions $)

The Short-Run Relationship among

Money, Prices, and Real Output
We have seen that in the long run, there is substantial support for our three
propositions about the relationship of money to economic activity. What is
the evidence on the short-run relationship between money and economic
activity? To answer this question, we first ask what our aggregate demand
and supply model predicts for the short-run relationship among money, the
price level, and real output. Then we return to U.S. history to see how well
this short-run model explains actual data on these variables.
Theoretical Relationships Between

Money and Economic Activity
In Chapters 16 and 17, we learned that the long-run aggregate supply curve
is vertical, since in the long run an economy can have increased output only
if the natural rate of output rises. In the short run however, when wages and
prices of other inputs are fixed, the aggregate supply cuve is upward sloping.
This implies that in the short run, an increase in aggregate demand will lead
to an increase in both the price level and the level of output.
What can change aggregate demand? As we first saw in Chapter 16,
there is a whole list of determinants of aggregate demand, which we added
to when we considered the open economy in Chapter 17. Changes in any of
these determinants will shift the location of aggregate demand and cause
changes in the price level and real output. In the previous section, we saw
that persistent changes in aggregate demand are most frequently caused by
ongoing increases in the money stock when we examined decade-long av¬
erages of historical data.
The short run, however, has many determinants of aggregate demand,
and changes in any one of them can alter the price level and real output.
Moreover, two or more determinants (like taxes and the money supply) often
change at the same time, so the net effect on aggregate demand, and hence
on the price level and real output, is unclear. Thus, we have no reason to
believe the relationship between money and other economic variables will
be as strong in the short run as we found it to be in the long run. Nonetheless,
it is useful to see what happens to our scatter diagrams when we look not at
decade-long averages but at quarterly observations of money growth rates,
nominal GDP growth rates, inflation rates, and real output (real GDP) growth
rates. This will not only provide information on the short-run relationship
between money growth rates and these variables but also give us an idea of
the types of problems faced by monetary policymakers, who must respond
to short-run movements in the economy with changes in the money growth
rate that may be tied only loosely to policy goals such as growth in real
output.
The Short-Run Relationship Among Money, Prices, and Real Output 639
Empirical Evidence on the Short-Run Relationship

To examine the short-run empirical relationships, we will look at quarterly
data on money growth, inflation, real GDP growth, and nominal GDP growth
from 1959 to 1992. Then we will duplicate our scatter diagrams that previ¬
ously used a decade-long data set. This will allow us to see whether growth
in the money supply during a given quarter had any effect on economic
activity during the same period.
Figure 18.7 looks at the relationship on a quarter-to-quarter basis be¬
tween the rate of growth of nominal GDP and the rate of growth of the
money stock, measured using both M2 (part a) and Ml (part b). It also shows
an estimated line of best fit. It is clear from the scatter of points that the
relationship between nominal GDP growth and money growth is much more
tenuous on a quarter-to-quarter basis. In fact, the best fit line has an estimated
slope of .39 for M2, which is statistically larger than zero but also statistically
smaller than 1. This means a 10 percent increase in money growth this
quarter increased this quarter’s nominal GDP growth by only 3.9 percent.
For Ml, the estimated slope of the best fit line is even smaller—.17—
meaning a 10 percent increase in Ml money growth this quarter increased
this quarter’s nominal GDP by only 1.7 percent. Thus, we find that money
growth in a given three-month period was positively related to the rate of
nominal GDP growth during that period, but the relationship was much
weaker than the one-to-one relationship we found over the 10-year periods.
Thus, we see that short-run (quarter-to-quarter) growth in the money supply
leads to small but positive increases in the growth of nominal GDP.
The short-run effects of money growth on inflation and real GDP also
differ from the long-term effects we saw in the previous section. Figure 18.8
considers the short-run effects of increases in money growth on the inflation
rate. The money growth rates and inflation rates are for each quarter from
1959 through 1992. Several things are very apparent in this graph. First,
there is not a strong positive relationship between money growth and infla¬
tion like that in our long-run study. Second, the best fit line for M2 growth
has an estimated slope of .15, and for Ml growth the estimated slope is .04,
although the slope of Ml growth is not statistically different from zero. We
conclude that in the short run, on a quarter-to-quarter basis, the inflation rate
and the rate of money growth are not very closely related. This suggests that
short-run increases in the growth of the money supply do not lead to im¬
mediate increases in the rate of inflation. This result is quite different from
our long-run analysis, and indeed different from the predictions suggested
by our short-run aggregate supply and demand analysis.
What is the short-run relationship between the rate of growth of real
output and the rate of money supply growth? Figure 18.9 plots quarter-to-
quarter real GDP growth and short-term money supply growth. The best fit
line is actually estimated to have a positive slope of .23 for M2, which is
statistically different from zero and statistically less than 1, and .13 for Ml.
Real GDP growth was positively related to money growth. This indicates
that short-run increases in money supply growth do lead to short-run in¬
creases in real output. Together, Figures 18.8 and 18.9 suggest that in the
short run, the Fed can increase the money supply and increase real output
with very little impact on the price level. However, the results presented in
the previous section suggest that the long-run effects of such an increase
will be merely to increase the price level and inflation. This presents an
obvious dilemma for the Fed and policymakers: Increasing the growth rate
of the money supply may lead to greater economic growth in the short run,
but in the long run economic growth will fall to its natural rate and inflation
rates will rise.
To summarize our short-run findings, we have found that nominal GDP
growth, the inflation rate, and real GDP growth are positively related to the
The Short-Run Relationship Among Money, Prices, and Real Output 641
rate of money growth in the short run, and the best fit line describing this
relationship for M2 is estimated to have a slope of .39 for GDP, .15 for
inflation, and .23 for real GDP. The relationship between Ml growth and
these three variables is even weaker. Thus, in contrast to the long run, in the
short run changes in monetary growth do affect the real output produced in
the economy but have little or no effect on the price level.
Summary j Our short-run empirical evidence suggests that increases in money growth
Exercise 18.21 do not affect the short-run inflation rate but do increase the growth in real
output. Use aggregate demand and supply analysis to explain how this can
occur.
Suppose the economy is in equilibrium at point A and the Fed

increases the money supply by 3 percent during one quarter. This leads to
an increase in aggregate demand from AD0 to AD]. If the short-run aggregate
supply curve is flat, the new short-run equilibrium will be at B, where the
price level remains at P0 but real output rises from F0 1° Tj. Thus, when the
short-run aggregate supply curve is flat, increases in monetary growth will
I lead to short-run increases in real output but no change in the price level or
the rate of inflation.
Reconciling the Long-Run and Short-Run Results

U.S and international data reveal that the short-term and long-term effects
of monetary growth are quite different. In the short run, increases in the
growth rate of money are related to increases in the growth rate of real GDP
but are not related to the inflation rate. Yet in the long run, increases in the
growth rate of money are related to increases in the inflation rate but not to
increases in the growth rate of real GDP. How does this happen?
The explanation involves several elements. First, many things change in
the economy on a day-to-day basis, including many of the determinants of
aggregate demand. These short-run changes can lead to changes in the price
level, output, and nominal GDP that are not accounted for in our three
propositions, which all concern the long-run effects of changes in the money
growth rate. Some of these changes in demand will be changes in velocity,
and these changes will alter the relationship between money growth and
nominal income growth.
Second, there are lags in the effects of changes in money growth on
changes in aggregate demand and hence in nominal income. These lags are
due to the way money affects the economy. Changes in the money stock can
exert powerful influences on aggregate demand, but these influences do not
occur instantaneously. Instead, increases in the money supply first show up
as additional money balances in the hands of individuals. These increased
money holdings raise consumers’ spending power, but individuals often react
somewhat slowly to such increases. Increases in the money supply also
temporarily lower the interest rate and stimulate investment spending.
Once aggregate demand shifts, the effects on the economy also depend
on the slope of the short-run aggregate supply curve. If that curve is very
flat, a change in aggregate demand will have a larger effect on real output
than on the price level. If the short-run aggregate supply curve is very steep,
a change in aggregate demand will have a larger effect on the price level
than on real output. The short-run evidence seems to suggest that the short-
run aggregate supply curve is fairly flat, since in the short run increases in
the money growth rate affect real output growth but not the inflation rate.
Another interpretation, however, is that the quarter-to-quarter movements in
money growth are themselves responses to movements in real output. The
scatter diagrams indicate only that money growth and output growth are
related, not which causes which. It may be that in the short run, increases in
real output growth raise money demand and hence increase money growth
due to an endogenous money supply. Box 18.2 expands on this possibility.
For the above reasons, increases in the rate of money growth take time
to affect the economy. Renowned monetary economist Milton Friedman has
Does Money Create Income, or

Does Income Create Money?
A considerable amount of economic research changes in money predate changes in income.1

has concentrated on the effects of nominal However, subsequent work by Sims and others
money balances on income. One of the best- (e.g., Robert Litterman and Laurence Weiss2)
documented facts in economics is that money found that the results are not so clear when the
and income tend to move together over time. study includes interest rates along with money
Traditionally this has been explained by changes and income. In fact, when interest rates are in¬
in the money supply causing changes in income. cluded, money no longer plays a significant role
But this relationship could be the result of in explaining changes in income. Instead, it is in¬
I changes in income causing changes in the

money supply.
One of the first attempts to analyze the
terest rate changes that precede changes in
both money and income. The meaning of this
result is not clear, and many interpretations
money-income relationship was the seminal have been offered.
work by Milton Friedman and Anna Schwartz Bennett McCallum argues that the mone¬
titled A Monetary History of the United States, tary authority may be following an interest rate
1867-1960 (Princeton, N.J.: Princeton University target, so changes in the interest rate indicate
Press, 1963). By examining historical episodes of policy actions.3 According to this view, the re¬
monetary policy in the United States, they were sults of interest rate changes preceding income
able to isolate cases in which changes in money changes still favor the importance of monetary
growth had an independent origin, that is, an policy. Consider, for example, a situation in
origin that was not a response to changes in which the Fed has decided to keep the interest
macroeconomic conditions. These changes in rate constant at 6 percent. To maintain that in¬
money growth resulted from factors such as terest rate, the Fed must supply any amount of
gold discoveries (which increased the money money the economy demands at that interest
supply under the gold standard), changes in rate. Since the demand for money is a function
Federal Reserve leadership, or changes in mone¬ of the level of income, any increase in income
tary institutions. In most of these cases, Fried¬ will lead to a higher demand for money. When
man and Schwartz could document that this happens, the Fed will have to increase the
changes in money growth occurred prior to Continued on p. 645
changes in income, supporting the idea that
changes in money caused changes in income 1 Christopher Sims, "Money, Income, and Causality,"
and not vice versa. American Economic Review (September, 1972), 540-555.
More recently, there have been formal tests 2 Robert Litterman and Laurence Weiss, "Money, Real In¬
terest Rates, and Output: A Reinterpretation of Postwar
of the hypothesis that changes in the money
U.S. Data," Econometrica (January, 1985), 129-156.
supply cause changes in income. The first and 3 Bennett McCallum, "A Reconsideration of Sims' Evi¬
probably best known of these tests is probably dence Concerning Monetarism," Economics Letters
the work by Christopher Sims, who found that (1983), 167-171.
Reconciling the Lo n g-R u n an d Short-Run Results 645
Continued from p. 644 the money supply, and this could lead to a
money supply to keep the interest rate constant lower income level. Thus, the interest rate
Thus, increases in income will lead to an in¬ changes could lead to changes in income and
crease in the money supply—exactly the reverse the money supply not because money is unim¬
of Friedman and Schwartz's results. Moreover, portant but because the Fed's interest rate peg
any increase in the interest rate peg—say, from obscures the role of the money supply in caus¬
6 to 8 percent—will require the Fed to reduce ing income.
suggested that long and variable lags link changes in the money supply to
changes in other macroeconomic variables. Hence, the quarterly scatter plots
shown earlier can show no relationship between, say, money growth and
inflation, whereas the long-run scatter plots show a clear one-to-one rela¬
tionship between these two variables. The problem is that the initial increase
in money growth has little or no effect on inflation. Only after individuals
accumulate these money balances do they begin to spend these funds, and
this is the cause of the increased aggregate demand. Moreover, in the short
run aggregate supply is not vertical, so increases in aggregate demand raise
output. Only in the long run do we get the predicted link among money
growth, inflation, and growth in real output.
How can we see the movement from the short run to the long run in our
U.S. data? One way is to take account of the fact that changes in the money
growth rate affect only variables like the inflation rate or the rate of growth
of nominal GDP with a lag. To do this, we use a measure of money growth
that is not the current quarter’s value but instead is an average of money
growth rates over the previous 16 quarters. In this way, we compare average
money growth over several years, rather than the money growth rate this
quarter, to other variables.
This measure of average money growth is called a moving average. Its
name comes from the fact that it is an average—in our case, of 16 quarters—
and that the value of the average changes as we move through time. Every
period we recalculate this average, using only the 16 most recent data points.
We now look at a final set of scatter plots of money growth rates against
nominal GDP growth, inflation, and real GDP growth, using a moving-
average measure of money growth rates. The idea is to see whether using
the moving average to take account of the lags in the effect of money growth
rates on the economy can begin to reveal results more like the long-run
results from our decade-long averages in Figures 18.1 through 18.3.
Figure 18.10 looks at the relationship of nominal GDP growth to the
moving averages of M2 and Ml money growth. It also includes the estimated
line of best fit. The scatter of points clearly indicates that the relationship
between the nominal GDP growth and the moving average of M2 money
growth is more like the long-run results in Figure 18.1 than the short-run
results in Figure 18.7. Indeed, the slope of the best fit line is .98, which is
statistically indistinguishable from 1. Of course, a lot of variation occurs
Moving-Average Money Growth and Growth in

Short-Run Nominal GDP
If money growth affects nominal GDP with a lag, there will be a positive relationship between the mov¬
ing average of past money growth rates and this guarter's growth in nominal GDP. In part a this graph
for the United States uses data from 1965 through 1992 and suggests that the 16-month average of
past quarterly M2 money growth rates was positively related to this quarter's growth in nominal GDP.
This suggests that money has lagged effects on nominal GDP. The relationship of the moving average of
quarterly Ml growth rates to nominal GDP (part b) is positive but much weaker, and is not statistically
significant.
(a) (b)
Commerce, Bureau of Economic Analysis, Sur\>ey of Current Business, various issues; Citibase electronic database.
about the best fit line, but over time changes in the moving average of M2
money growth rates tend on average to be matched by changes in the rate
of nominal GDP growth. Notice that the link between Ml growth and
nominal GDP is not very striking. This is one reason many economists
consider M2 a better measure of the money supply.
What about the relationship between the moving average of money
growth and inflation? Figure 18.11 plots inflation against our moving-
average measures of M2 and Ml money growth. As in Figure 18.10, we see
a lot of variability, but we also see a positive relationship between inflation
and money growth rates. Indeed, the best fit line for M2 has an estimated
slope of .73, which is not quite 1 (and statistically less than 1) but is much
higher than the .15 slope coefficient on the link between inflation and M2
money growth in Figure 18.8. Thus, our moving average of M2 money
growth rates is much more closely related to the inflation rate than just the
current quarter’s M2 money growth rate. This suggests that the average M2
money growth over the past 16 (or more) quarters is what affects this
quarter’s inflation rate. Again, the case for Ml is less strong.
Finally, what is the relationship between the moving average of money
growth rates and this quarter’s real output growth? Figure 18.12 plots real
GDP growth and our moving-average measure of money growth. The best
fit line for M2 is estimated to have a slope of .25, which cannot be statistically
distinguished from zero. Thus, real GDP growth in any current quarter is
only weakly related to the moving average of M2 money growth over the
past 16 quarters. The moving average of M2 money growth does affect
inflation, but has a much smaller effect on the growth in real output. For
Ml, the relationship is negative. These results are more similar to those we
found for the long run
To summarize, we have found that propositions 1 through 3 hold in the
long run but not in the short run. We have also found that if we average M2
money growth rates over a number of periods, we can move a long way
toward our long-run results even by looking at quarterly data on the inflation
rate or real GDP growth rates. It seems high money growth in any one period
just doesn’t matter very much, while high growth over a year or two matters
a lot for both nominal GDP and inflation. Box 18.3 looks at the United
Kingdom data to see whether the moving-average calculations of money
growth help us better understand the relationship of money growth to nom¬
inal and real GDP and the inflation rate in the United Kingdom.
Summary Some economists have suggested that prices are sticky, or slow to adjust, in
Exercise 18.3 the short run. In that case, any increase in money growth will first raise real
GDP and nominal GDP but will not change prices. Over time, however, the
increase in money growth will raise the inflation rate but not affect real
GDP; thus, absent any other changes, real GDP will return to its initial level.
Money Growth, GDP Growth, and Inflation

in the United Kingdom: 2
In the text, we looked at evidence on the rela¬ of .22, which is significantly less than one.
tionship between a moving average of money When we plot real GDP growth against money
growth and inflation, GDP growth, and nominal growth in part b, we find a best fit line that has
GDP growth in the United States. Here we fol¬ a slope not significantly different from zero. The
low up on Box 18.1 and present similar meas¬ evidence is that the U.K. data provide less sup¬
ures from our U.K. data. Part a of the accompa¬ port for proposition 2 than did our long-run se¬
nying figure presents the scatter plot of inflation ries of U.S. data or the international data of
rates and our moving-average measure of U.K. Dwyer and Hafer. The data do, however, tend
money growth rates. Again we find a positive to support proposition 3.
relationship, but our best fit line has a slope
Moving Average of Money Growth (%) Moving Average of Money Growth (%)
(a) (b)
(a) If this is the case, how will real GDP respond to an increase in money
growth in the short run and in the long run? (b) Suppose the money supply
increased at 5 percent during the first two quarters of 1994 and by 10 percent
over the second two quarters. What was the moving average of the quarterly
money growth rate during this one-year period? (c) If money growth in¬
creases further to 15 percent in each of the first two quarters of 1995, what
happens to the one-year moving-average growth rate of the money supply?
Answer: (a) An increase in money growth will first raise real GDP growth
rates, as the question indicates. This is the short-run effect. Over time, as
prices start to react and the inflation rate rises, real GDP will return to its
initial level, so the GDP growth rate must fall. Thus, an increase in money
growth first causes an increase in growth of real GDP, but after a time real
GDP growth declines until it is restored to its initial level. After this, real
GDP growth continues at its usual pace. Money growth causes real GDP
growth to first rise, then fall, and then settle back down to its original rate,
(b) The one-year moving average of the money growth rate is the average
growth in the money supply over the most recent four quarters. Thus, the
moving average in 1994 was (.05 + .05 + .1 + .l)/4, or 7.5 percent (c)
Based on the most recent four quarters, the moving average of money growth
is now (.1 + .1 + .15 + .15)/4, or 12.5 percent.
Inflation, Money Growth, and Interest Rates
An important issue in monetary economics is the relationship among money

growth rates, inflation, and interest rates. We already examined the link
between money growth rates and inflation and concluded that in the long
run, increases in money growth rates lead to increases in the inflation rate,
while in the short run increases in money growth rates in any one period
have little or no impact on the inflation rate. Now we examine how changes
in the inflation rate affect the interest rate.
Irving Fisher, a famous economist from the early part of this century,
postulated a one-to-one relationship between the interest rate and the ex¬
pected inflation rate, a relationship now known as the Fisher equation. As
we saw in Chapter 3, the Fisher equation is
i = r + TTe,
where i is the nominal interest rate, r is the real interest rate, and ire is the
expected inflation rate.
In using this relationship, Fisher argued that the real interest rate could
be taken as constant, at least for the most part. If so, the Fisher equation
predicts that an increase in the inflation rate would show up as an equal
increase in the nominal interest rate. The link from money growth rates to
interest rates would then occur through expected inflation. Increasing the
money growth rate, at least over an extended period of time, would raise the
inflation rate and the expected inflation rate and thereby increase the nominal
interest rate. The real interest rate would remain constant by assumption.
Does this really happen when the inflation rate increases? Part a of Figure
18.13 plots the inflation rate and the nominal interest rate from 1959 to 1992.
Notice that the nominal interest rate and the inflation rate tracked fairly
closely until about 1980. After 1980 the inflation rate declined rather sharply,
but interest rates fell much more slowly, and not until the early 1990s did
the two rates approach each other. Thus, there appears to be a relationship
Sources: Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, various issues;
U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, various
between the inflation rate and nominal interest rates, but it is by no means a
steadfast rule that nominal interest rates mimic increases or decreases in the
inflation rate. Part b of Figure 18.13 provides a scatter diagram and the best
fit line for the relationship between the one-year T-bill rate and the inflation
rate. The slope of the best fit line is .58, not the 1 predicted by the Fisher
relationship. Using a moving-average measure of inflation rates would make
this slope closer to unity. However, the scatter plot also shows quite a lot of
variation around the best fit line, indicating the relationship between interest
rates and inflation rates is not the neat and tidy one the Fisher equation
would lead us to believe.
The slow decline of nominal interest rates and the quick drop in the
inflation rate in the 1980s meant the real interest rate during this time was
higher than it had been in recent history; that is, the real interest rate,
calculated as the nominal interest rate minus the actual inflation rate, was
652 Chapter 1 8 Money and Economic Activity-. A Look at the Evidence
not constant. Of course, the Fisher equation suggests the real interest rate is
the nominal interest rate minus the expected inflation rate, but expected and
actual inflation rates would have to differ for quite some time to keep the
real interest rate constant. A more likely explanation, and one that supports
other empirical research, is that real interest rates do vary over time, espe¬
cially in the short run.
Why were real interest rates higher in the 1980s? Several explanations
have been offered. One is that real interest rates were high due to the large
budget deficits of that period. Increased government borrowing in the loan¬
able funds market raised demand for loanable funds and pushed up interest
rates even as inflation rates were declining, yielding higher real interest rates.
Another explanation is that expected inflation rates did differ from actual
inflation rates for an extended period in the 1980s. According to this view,
actual inflation rates fell much more rapidly than expected, and expected
inflation rates adjusted slowly to those changes. Thus, the real interest rate
in the Fisher equation, defined as the nominal interest rate minus the expected
inflation rate, may well have remained constant.
Instead of looking at the effect of inflation on interest rates, let us
examine the effect of money growth on interest rates. Part a of Figure 18.14
plots the nominal interest rate against the rate of money growth. Notice the
best fit line is upward sloping, although the estimated slope for M2 growth
is .03, which is not statistically different from zero. (The results for Ml are
similar but are not given here.) Thus, we find that money growth has a weak
effect on interest rates.
If we look at the relationship between the interest rate and a moving
average of money growth rates (part b), we see a much stronger relationship.
The estimated slope coefficient of the best fit line is .56, which is statistically
significant. It seems that increases in the money growth rate lead to rises in
the interest rate, but the effect is not immediate; instead it requires a sustained
period of increased money growth. One puzzling feature is that the increase
in money growth rates does not cause the interest rate to decline, even when
comparing the current-quarter money growth rate to the interest rate. We
will say more about this particular result in the next section.
In summary, what do our data suggest about the effects of changes in
monetary growth on interest rates? The effect of changes in the current
quarter’s money growth rate on the expected inflation rate in the Fisher
equation is rather tenuous. This is partly because any increase in money
growth causes inflation only in the long run and partly because the link
between inflation and expected inflation may not be exact. The lesson for
monetary policy is that in the short run, changes in the money growth rate
may have little effect on the expected inflation rate and thus may not raise
interest rates. In the long run, however, a persistent increase in the money
growth rate will cause inflation, and increases in inflation will tend to cause
nominal interest rates to rise. Thus, policy actions taken in the short run may
have very different long-run effects.
Summary (a) How do considerations of income taxes affect the Fisher relationship?
Exercise 18.4 (Remember, income taxes are paid on interest income.) (b) How might this
relationship partially explain the rising real interest rate and declining nom¬
inal interest rate during the early 1980s?
Answer: The Fisher relationship states that the real interest rate equals the
nominal interest rate minus the expected inflation rate. However, this is the
relationship in the absence of taxation. If lenders are required to pay a tax
rate of t on their interest earnings, they will calculate their aftertax real
interest rate as the nominal interest rate, i, times the portion of this income
that they get to keep after taxes, (1 — t), minus the expected inflation rate;
that is, the aftertax real rate is
r = /(I — t) — tt€.
(b) Tax rates were significantly reduced during the early 1980s. The analysis
in part a reveals that a decrease in the tax rate will, holding other things
constant, increase the real interest rate. Alternatively, holding the real interest
rate constant, it will lower the nominal interest rate. Figure 18.13 reveals
that both effects occurred during the 1980s. The gap between the nominal
interest rate and the inflation rate widened during this period, even though
the nominal interest rate declined.
Real Money Balances, Velocity, and Interest Rates
In Chapter 15 we discussed money demand at length, beginning with the

equation of exchange and covering several other views on money demand.
One was the modified Cambridge view that money demand can be written
as
AT7 _ Y
(18.1)
We now investigate how well this relatively simple view of money demand
can explain the U.S. historical experience. Then we turn to some policy
implications of this view.
,
Real Money Balances Real Income,
and the Interest Rate
The modified Cambridge view of money demand has two determinants. The
first is real income, a measure of the desired amount of transactions. The
second is the interest rate, a measure of the opportunity cost of holding
money. How well do these two variables explain movements in real money
balances? Figure 18.15 presents two graphs using U.S. data from 1959 to
1992. Part a is a scatter diagram of real M2 money balances and real GDP.
This plot indicates a strong positive relationship between real money bal¬
ances and real GDP. The graph includes an estimated best fit line, and the
points lie very close to this line over the entire sample. Thus, we see that
increases in real income indeed lead to rises in real money balances, just as
our theory of money demand would suggest. Increased real income raises
the amount of purchases consumers will make, which increases their desire
to hold money to use in making transactions.
Part b of Figure 18.15 graphs real M2 balances and real GDP over time.
Notice how closely they track each other’s movements through time. The
downturns in real GDP are recessions, which are mimicked by downturns
in real money balances. It appears real GDP can explain much about real
M2 holdings, just as the modified Cambridge money demand equation would
suggest.
Real Money Balances, Velocity, and Interest Rates 655
Taking the modified Cambridge view further, we see that equation 18.1
can be rearranged by dividing both sides by income to yield
1 . Md
- = k(i) — —.
V(i) PY
Here the inverse of velocity, \IV, is also called the Cambridge k and is
simply real money balances divided by real income. Moreover, velocity, and
hence the Cambridge k, depends on the nominal interest rate. An increase
in the nominal interest rate raises the opportunity cost of holding money
balances. This causes money users to try to get by using fewer money
balances. As they attempt this, money will change hands more frequently
and velocity will increase. The Cambridge k will then decrease. Thus, ve¬
locity rises and the Cambridge k decreases with increases in the interest rate.
We investigate this interest rate effect on velocity by plotting movements
in velocity over time along with the nominal interest rate in Figure 18.16.

Michael R. Baye Dennis W. Jansen

Uploaded by

Copyright:

Available Formats

Michael R. Baye Dennis W. Jansen

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Michael R. Baye Dennis W. Jansen

Uploaded by

Copyright:

Available Formats

MONEY, BANKING *

Wr Hi) q 6? 14% 1•«*n4%+’ft'

533)4 447/8.32 S'S

80%64«16 2B2S 79 781ft

Houghton Mifflin Company Boston Toronto

Geneva, Illinois Palo Alto Princeton, New Jersey

To my parents, Firmin and Jaynet Baye,

In memory of Elmer Jansen and William Hennessey.

Sponsoring Editor: Denise Clinton

Cover design: Judy Arisman

Copyright © 1995 by Houghton Mifflin Company. All rights reserved.

Printed in the U.S.A.

PART ONE Introduction to Money and Banking i

CHAPTER 1 Money and Banks 2

PART TWO The Microeconomics of Banking and Financial Markets 91

CHAPTER 4 Supply and Demand in Financial Markets 92

PART THREE The Valuation of Financial Assets 235

CHAPTER 8 The Time Value of Money and Asset Pricing 236

PART FOUR The Creation of Money 405

PART FIVE Money and the Macro Economy 525

CHAPTER 16 A Simple Macroeconomic Model 526

PART SIX Special Issues in Monetary Economics 665

CHAPTER 19 An Introduction to Monetary Policy 666

part one Introduction to Money and Banking i

chapter i Money and Banks 2

The Evolution of Banks 14

Box 1.1 The Data Bank Components of the Money Supply 4

Box 1.3 Inside Money Cigarettes as Money in a Prisoner of War Camp

Box 1.4 Inside Money Where Are All the Dollars? 17

chapter 2 Financial Markets, Financial

Money Market Instruments 41

Official Definitions of the Money Stock 54

Box 2.3 Inside Money A Lexicography for Bonds 48

CHAPTER 3 Money, Inflation, and Interest 6i

Distinguishing Between Inflation and Economic Growth 61

The Price Level and Real Output 62

How Inflation and Taxes Affect Interest Ra tes 80

The Real Interest Rate 80

Historical Data on Interest Rates 86

PART TWO The Microeconomics of Banking and Financial Markets 9i

CHAPTER 4 Supply and Demand in Financial Markets 92

Basic Supply and Demand Analyis: A Brief Review with Applications

Demand for Fee-Based Financial Services 93

The Market for Loanable Funds 104

The Demand for Loanable Funds 104

Applications of the Loanable Funds Model 116

Box 4.3 Inside Money Nominal Interest Rates and Expected

CHAPTER 5 The Bank as a Firm: Loans no

Purely Competitive Banks 130

Banks with Market Power 139

Sources of Market Power for Banks 139

Oligopoly Banks 147

The Nature of Oligopoly Interdependence 148

Banks with Imperfect Information 153

Symmetric Information 153

Box 5.1 1 he Data Bank Economies of Scale in Banking 142

Box 5.2 International Banking Concentration Ratios in Banking 149

CHAPTER 6 The Bank as a Firm: Deposits m

Total and Marginal Product 167

The Demand for Deposits by Purely Competitive Banks I7i