Macroeconomics Simplified Understanding
Macroeconomics Simplified Understanding
Macroeconomics Simplified Understanding
Macroeconomics Simplified
Understanding Keynesian and
Neoclassical Macroeconomic Systems
Nicoli Nattrass
G. Visakh Varma
Dedication
This book is dedicated to the memory of Visakhs parents,
K. R. Goda Varma and Malathi Goda Varma, and Nicolis mother, Jill Nattrass.
Thank you for choosing a SAGE product! If you have any comment,
observation or feedback, I would like to personally hear from you.
Please write to me at [email protected]
Vivek Mehra, Managing Director and CEO,
SAGE Publications India Pvt. Ltd, New Delhi
Bulk Sales
SAGE India offers special discounts for purchase of books in bulk.
We also make available special imprints and excerpts from our
books on demand.
For orders and enquiries, write to us at
Marketing Department
SAGE Publications India Pvt. Ltd
B1/I-1, Mohan Cooperative Industrial Area
Mathura Road, Post Bag 7
New Delhi 110044, India
E-mail us at [email protected]
CONTENTS
List of Tablesxi
List of Figuresxiii
Acknowledgementsxvii
1. Introduction
Technical and Ideological Issues
2.
1
3
5
5
7
10
11
11
12
13
15
18
viii
Macroeconomics Simplified
The Determination of Output and Employment
26
The aggregate supply (AS) function26
Says law31
The Quantity Theory of Money
34
New Classical Stories about the Business Cycle
36
Review Questions and Answers
38
4.
40
41
44
48
49
58
60
62
5.
63
63
65
66
70
71
72
72
76
77
Contents
ix
Macroeconomics Simplified
The NAIRU
The Low Road/High Road Counter Narrative
Review Questions and Answers
163
164
166
193
194
195
197
Index199
About the Authors204
LIST OF TABLES
2.1
2.2
2.3
2.4
2.5
8
10
13
14
15
168
LIST OF FIGURES
2.1
2.2
2.3
6
8
9
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
3.11
3.12
22
23
24
24
25
27
28
28
29
30
33
34
4.1
4.2
4.3
4.4
Keyness Economics
The Consumption and Investment Functions
The Keynesian Cross
An Increase in Government Spending
47
50
56
57
5.1
5.2
5.3
5.4
5.5
66
69
73
74
76
6.1
6.2
81
84
xiv
Macroeconomics Simplified
119
120
122
123
126
127
129
130
132
136
137
List of Figures
148
149
150
152
169
170
172
175
176
177
178
179
180
154
157
160
161
162
181
182
183
184
188
191
192
xv
ACKNOWLEDGEMENTS
G. Visakh Varma would like to acknowledge the help of his teacher Dr P. H. Rayappa,
retired Professor and Head, Population Research Centre, Institute for Social and Economic Change, Bangalore, and of his former colleagues C. K. Hareendranathan,
retired Professor of Economics, Government Arts and Science College, Calicut
(Calicut University), Dr Martin Patrick, retired Professor of Economics, Maharajas
College (Government), Ernakulam (Mahatma Gandhi University), and Dr S. Muraleedharan, retired Professor of Economics, Maharajas College (Government), Ernakulam
(Mahatma Gandhi University). We would both like to acknowledge the many students
we have taught and who have helped us understand and appreciate macroeconomics.
CHAPTER 1
Introduction
This book draws on some material previously published in Nattrass, N. 2000. Macroeconomics: Theory
and Policy in South Africa. Cape Town: David Philip.
2
Krugman, P. 2009. How Did Economists Get It So Wrong? New York Times, 2 September. http://www.
nytimes.com/2009/09/06/magazine/06Economic-t.html?pagewanted=all
1
Macroeconomics Simplified
The Indian national accounts are introduced in Chapter 2. Chapter 3 opens the theoretical discussion with an outline of the neoclassical general equilibrium model. This is
followed by an introduction to Keynes economics and to the simple Keynesian model
of income determination (Chapter 4). Chapter 5 introduces the relationship between
investment and the interest rate. Chapters 6 and 7 show how the real and financial sectors can be brought together in the same (ISLM) model as a tool for discussing the
impact of fiscal and monetary policies.
Chapter 8 introduces the labour market by means of the ASAD model. It shows
how the basic framework can be used in a more dynamic way to discuss the relationship
between unemployment and inflation. Unemployment and related issues in the Indian
economy are discussed in this chapter.
Quoted in Snowdon, B., H. Vane and P. Wynarczyk. 1994. A Modern Guide to Macroeconomics: An
Introduction to Competing Schools of Thought. p. 4. Aldershot: Edward Elgar.
4
Ibid.
5
Krugman, How Did Economists Get it so Wrong?
3
Chapter 1 Introduction
Chapter 9 extends the conventional Keynesian ISLM framework to the open economy, and discusses the role of prices, exchange rates and interest rates in an open economy context. Chapter 10 concludes the book with a brief discussion of the proclivity
of capitalism to financial crisis and the insights offered by Hyman Minsky and Joseph
Schumpeter.
In the final analysis, ones choice of theory depends on how convincing it seems (on
its own terms), on how it gels with ones political attitudes and how the theory fits the
facts at hand. Understanding the basic theoretical constructs lying behind rival understandings is an essential pre-condition to forming good judgements in this regard. But it
is just a beginning. As John Maynard Keynes (the father of macroeconomics) argued the
master economist must possess a rare combination of gifts... He must be a mathematician, historian, statesman, philosopherin some degree ... as aloof and incorruptible as
an artist, yet sometimes as near to earth as a politician.7
Quoted in Snowdon et al. A Modern Guide to Macroeconomics. p. 174.
Quoted in Skidelsky, R. 1992. John Maynard Keynes: The Economist as Saviour, 19201937. p. 411.
London: Macmillan.
6
7
Macroeconomics Simplified
And even if one does not aspire to become such a master economist, getting to grips
with the language economists speak is an important skillif only to prevent them from
pulling the wool over your eyes. To quote Keynes once again:
Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually the slaves of some defunct economist. Madmen in authority, who hear
voices in the air, are distilling their frenzy from some academic scribbler of a few years
back. I am sure that the power of vested interests is vastly exaggerated compared with the
gradual encroachment of ideas. Not indeed, immediately, but after a certain interval; for
in the field of economic and political philosophy, there are not many who are influenced
by new theories after they are twenty-five or thirty years of age, so that the ideas which
civil servants and politicians and even agitators apply to current events are not likely to
be the newest. But sooner or later, it is ideas, not vested interests, which are dangerous
for good or evil.8
So, now is the time to make sure that you are not enslaved later by some halfunderstood idea you picked up in economics lectures or in the newspaper. It is also the
time to equip yourself with the tools for understanding and criticising the economic
policies which may be thrust upon you by madmen in authority.
8
Keynes, J. M. 1936 (Reprinted in 1964). The General Theory of Employment, Interest and Money. pp.
38384. London: Macmillan.
CHAPTER 2
Macroeconomic models deal with aggregate economic data. Most of these come from
the national income and product accountsusually referred to simply as the national
accounts. Before moving on to the terrain of macroeconomic theory, it is useful to
revise some basic national income accounting concepts. This chapter does so with
reference to India.
Value-added
National income accountants measure the value-added at each stage of production so as to avoid double counting. For example, if a farmer sells coconuts to an
oil mill for `100, and if the mill processes the coconuts and sells coconut oil to a
hotelier for `125, then the value-added by the oil mill is `125 100 = `25. If the
hotelier cooks food with coconut oil and sells it for `200, then the value-added
by the hotelier is `200 125 = `75. The value-added in production is equal to
`100+25+75 = `200. In the absence of indirect taxes, subsidies, etc., this figure
is the same as the final value of the food at market price.
Source: Authors.
Macroeconomics Simplified
Expenditure
Output
Households
Factor Services
Firms
Income
The income method sums up all payments to the factors of production, that is, the
remuneration of workers, and the gross operating surplus (i.e., profits) of firms. This
gives an estimate of GDP at factor incomes. This should, in principle, be equal to
value-added.
The expenditure method sums up the components of final demand (expenditure on
final goods) in estimating expenditure on the GDP.
As outlined in the circular flow of income model, expenditure on goods and services
provides firms with the income used to pay the factor services (labour and capital) used
in producing value-added. Thus, as long as all incomes are channelled back into the
economy in the form of consumption and investment expenditure, income will equal
the expenditure and output.
Note that the national accounts only include flows that are evident in the monetary
economy. Thus, if a man paints his house by himself, then this work is not counted.
But if the man pays a firm to do it for him, then the work will be counted as additional
income and economic activity. Similarly, the work of cleaners and house-keepers only
appears in the national accounts if this is paid work; women who work for no pay are
not counted. This is a problem for India because many economic activities in villages are
non-monetary and hence do not appear in the national accounts.
Macroeconomics Simplified
Table 2.1 Indias GDP at Market Prices (`, Billion) in Current Prices
201112
` Billion
50,562.19
56.4
10,426.77
11.6
27,490.72
30.6
1,893.84
2.1
2,429.68
2.7
Plus exports
21,436.47
Less imports
27,224.54
6.5
Discrepancies
2,734.33
3.0
89,749.47
100
Source: Reserve Bank of India: Components of GDP (at market prices). Available on: http://
dbie.rbi.org.in/DBIE/dbie.rbi?site=statistics (downloaded 29 March 2013).
70,000
60,000
50,000
` Billion
32%
Gross Fixed Capital
Formation
40,000
30,000
57%
20,000
10,000
Government
Consumption
11%
90
19 91
92
19 93
94
19 95
96
19 97
98
20 99
00
20 01
02
20 03
04
20 05
06
20 07
08
20 09
10
20 11
12
1
3
Private Consumption
19
pensions and disability grants). There is thus a difference between government spending
on goods and services and the size of the government budget, which includes a significant amount of transfers.
Gross capital formation consists of investment spending (i.e., gross fixed capital formation) plus changes in stocks (i.e., inventories) and valuables. Investment is the total
amount spent by producers in the private and public sectors on capital goods during
the period concerned before provision is made for depreciation. It is made up of fixed
investment in machinery, construction and the change in inventories.
Note that investment refers to additions to the physical stock of capital (hence the
term fixed capital formation). It does not include financial investments such as the
purchase of bonds or shares. Such purchases would be regarded as savings rather than
as investments in the national accounts.
Investment has planned and unplanned components. Spending on plant and
machinery (fixed capital expenditure) is obviously planned. Some inventory accumulations are also probably planned, but most inventory accumulations are unplannedthat
is, the accumulation of unsold inventories which firms would have preferred to have
sold. Keynesian economic thinking draws an important conceptual distinction between
planned and unplanned inventory accumulationsbut the national accounts simply
estimate the total value of inventories (planned and unplanned). When this is added to
investment, the result is an estimate of gross capital formation.
Investment growth is typically the most volatile of all the components of GDE.
Figure 2.3 tracks the growth rate of GDE in India from 199091 to 201213. It shows
25%
Gross Fixed Capital
Formation
20%
15%
Private Consumption
10%
Government
Consumption
5%
201213
201011
200809
200607
200405
200203
200001
199899
199697
199495
199293
5%
199091
0%
Gross Domestic
Expenditure (C + I + G)
10%
10
Macroeconomics Simplified
that investment is indeed the most volatile component, whereas consumption is relatively stable and government spending becomes more important in periods of economic
downturn, notably in 2008. Variation in consumption in India is often driven by weather
conditions that affect agricultural output.
83,535
8,767
74,768
6,214
89,749
80,982
768
82,767
74,000
88,981
80,214
60,158
31,415
22,647
26,519
68,858
61,564
post-tax. As will become clear later on in the text, macroeconomic models often use
these formulations.
A Few Identities
Expenditure on GDP = C + GCF + G + X Z;
GDP at factor incomes = C + GDS + T;
where GCF = gross capital formation and GDS = gross domestic savings.
11
12
Macroeconomics Simplified
Government
budget
deficit/surplus
Balance of
payments current
account surplus/
deficit
National accounting conventions ensure that GDS and GCF are by definition equal.
In a closed economy (i.e., assuming no international trade or capital flows), GDS and
GCF will always be equal because any difference between savings and investment will be
reflected in inventory accumulation (and in India, also by accumulation of valuables),
which, when added to investment, becomes GCF. In an open economy, any difference
between savings and investment will be reflected in net capital inflows from abroad.
In order to ensure that savings equal investment, national income accountants add the
change in gold and other foreign reserves to GDS. This means that GDSGCF will necessarily (by definition) be equal to zero.
National accounting identities are ex post (i.e. after the fact) identities. These differ
from the planned (or ex ante) relationships used in Keynesian macroeconomic models. Whereas GCF and GDS are necessarily equal, planned investment (I) and planned savings (S) are not necessarily equal. This is because I does not include unplanned inventory
accumulation and S does not include changes in foreign reserves. GCF = GDS is a national
accounting identity, whereas I = S is an equilibrium condition in a Keynesian model. Do
not confuse the ex post national accounting identities with the ex ante equilibrium
condition.
Table 2.3 Inflation and Real GDP Growth in India (`, Billion)
GDP at Factor
Cost Current
Prices
Growth
(%)
200506
33,905.03
14.1
32,530.73
9.5
1.042
4.2
200607
39,532.76
16.6
35,643.64
9.6
1.109
6.4
200708
45,820.86
15.9
38,966.36
9.3
1.176
6.0
200809
53,035.67
15.7
41,586.76
6.7
1.275
8.5
200910
61,089.00
15.2
45,161.00
8.6
1.353
6.1
201011
72,670.00
19.0
49,370.00
9.3
1.472
8.8
201112
83,535.00
15.0
52,436.00
6.2
1.593
8.2
201213
94,619.79
13.3
55,034.76
5.0
1.719
7.9
GDP
Growth
Deflator
(%)
When national accountants formulate the GDP deflator, the first step they need to
take is to select a base year. The base year should ideally be a normal year (i.e,. it should
not be a drought year). Next, a representative basket of goods (for the economy as a
whole) needs to be selected, and its price changes recorded over time.
The most recent base year used by the Reserve Bank of India is 200405. Table 2.3
shows that in this year, the GDP at factor cost was the same for constant and current
prices estimates. Table 2.3 records the growth rates for the current price estimates, the
constant price estimates and the GDP deflator. The growth in the GDP deflator gives us
an estimate of inflation for the economy as a whole (i.e., for producers and consumers).
The difference between the growth rates for the constant and current price GDP estimates is approximately equal to growth in current price GDP minus inflation.
The data in Table 2.3 show that the Indian economy grew strongly between 2004
05 and 201011, but slowed down significantly over the following 2 years. The global
financial crisis slowed real income growth in 200809, but unlike most of the rest of the
world, Indian growth recovered quickly. Inflation, however, continued to be a problem,
indicating that growth was beginning to push up against structural and other constraints.
It was thus to be expected that as growth slowed subsequently in 201213, so did inflation.
13
14
Macroeconomics Simplified
Industry (%)
Services (%)
195152
50.7
12.1
36.5
196162
42.6
16.0
42.3
197172
40.2
16.4
43.1
198182
34.1
20.5
45.0
199192
29.4
20.0
50.4
200102
23.0
19.2
57.8
201112
17.4
18.5
63.9
Year
quarrying constitute the primary sector of the Indian economy. The share of this sector
has fallen from about half of value-added in the 1950s to 18 per cent today. The secondary
sector (industry, manufacturing, electricity, gas and water supply and construction) rose
over the same period (now amounting to over 20 per cent of the economy)but it was
the tertiary sector (especially services) which has seen the fastest growthaccounting for
almost two-thirds of the economy today.2
Some commentators now worry that the growth has been unbalanced, that it has not
provided an adequate boost to industry or jobsand that neoliberal economic policies
exacerbated the relative decline of manufacturing whilst privileging large corporations.3
Others are concerned about the impact of this structural change on inequality as the
spread of earnings is wider in industry and services than in agriculture.4
Structural change away from agriculture and towards industry and services is typical of the process of economic development. Table 2.5 shows that India shares many
characteristics of low income countries, including a relatively large agricultural sector.
However, India has made great progress, especially during the 2000s, in terms of growth
and development and poverty rates have fallen in urban and rural areas.5 However, many
challenges remain both in terms of maintaining growth and ensuring that it enhances
2
For an accessible introduction to economic sectors in India, see, http://www.youtube.com/watch?v=
VdiT7qlEhFs
3
See for example, Alternative Survey Group. 2012. Alternative Economic Survey, India 200708: Decline
of the Developmental State. p. 113. Delhi: Daanish Books.
4
Pieters, J. 2010. Growth and Inequality in India: Analysis of an Extended Social Accounting Matrix,
World Development 38 (3): 27081.
5
See for example, Binswanger-Mkize, H. 2012. India 19602010: Structural Change, the Rural Nonfarm
Sector, and the Prospects for Agriculture. Berkeley: Department of Agricultural and Resource Economics,
University of California. http://areweb.berkeley.edu/documents/seminar/Binswanger.pdf
Middle-income
Countries
26,945
2,081
291
589
98.4
82.9
61.4
62.8
79.6
68.8
58.4
64.8
Agriculture as % of GDP
1.5
9.7
25.7
17.8
Low-income
Countries
India
India was ranked 134 out of 187 countries by the UNDP. The
India Human Development Report gave regional scores: Kerala
had the highest HDI score and Gujarat the lowest. This reflects
Keralas relatively progressive social policies.
welfare, notably by ensuring that the economy delivers on-job creation for relatively
unskilled people whilst also allowing high-wage, high productivity jobs to expand.
Indian policy makers also have to contend with great regional variations across the
country (Table 2.5).6
15
16
Macroeconomics Simplified
Politicians and mass media often give the impression that a rising GNP means
that the welfare of citizens is also rising. Although economic growth makes it
easier to reduce poverty and improve social services, there is no one-to-one
correlation between growth and welfare as economists James Tobin and William
Nordhaus pointed out over 40 years ago in their book Is Growth Obsolete.8
Remember, GNP measures only production, trade and consumption and it does
not include welfare measures such as happiness, inequality or environmental
sustainability and does not take into account externalities such as congestion
and pollution.
Source: Authors.
Green accounting and critiques of the GNP as an indicator of welfare draw attention
to the fact that growth should not be the sole objective of macroeconomic policy, but
that policy makers should also worry about the sustainability of that growth.
A noteworthy attempt to broaden the focus of measuring welfare beyond economic
indicators is the project of measuring gross national happiness. This idea, pioneered in
Bhutan (and praised recently in a lecture by Joseph Stiglitz at the Indian Statistical Institute), holds that human happiness is multi-dimensional and that social, spiritual and
environmental factors matter as well as material.9
The idea of measuring human happiness originates with the utilitarian philosopher Jeremy Bentham, but has largely been conflated with income by economists. This
appears to be changing. For example, French President Sarkozy asked renowned economists Joseph Stiglitz, Amartya Sen and Jean-Paul Fitoussi to assemble a Commission
on the Measurement of Economic and Social Progress to point out the limitations of
the GDP as an indicator of economic performance and social progress and to search for
better indices of human welfare and development. In another development, the World
Happiness Report was commissioned for the 2 April 2012 United Nations conference.10 It
argues that happiness can be reliably measured, that it should be taken seriously and that
we need to include other dimensions of human well-being (community trust, environment, governance, etc.).
Welfare economists typically assume that income and well-being move together: as
individuals get richer, they become happier, and as economies grow, the average level of
happiness rises too. This notion was challenged fundamentally by Richard Easterlin in a
Tobin, J. and W. Nordhaus. 1972. Is Growth Obsolete? New York: Columbia University Press.
Stiglitz, J. 2012. Indian Economy Doing Well: Joseph Stiglitz. Indian Express.com, 14 February 2012.
http://www.indianexpress.com/news/indian-eco-doing-well-joseph-stiglitz/898557/2).
10
Helliwell, J., R. Layard and J. Sachs. 2012. World Happiness Report. Produced for the UN by the Earth
Institute, Columbia University. http://issuu.com/earthinstitute/docs/world-happiness-report
8
9
1974 paper Does Economic Growth Improve the Human Lot? Some Empirical Evidence.11
In this classic article, which ironically was rejected by the American Economic Review,12
Easterlin argued that even though richer people were happier than poor people, the
average level of measured well-being for a country changed little over time, even when
there was substantial growth in average incomes. This suggests that people compare
themselves to those around them, and that growth may well have other costs to human
well-being even as living standards rise. His paper sparked significant work on the economics of happiness and to cross border researches in economics and psychology to find
out the determinants of human well-being/happiness.
Thirty-six years later, Easterlin and others published a paper using data from 37
countries (both developed and underdeveloped) showing that this happinessincome
paradox (sometimes called the Easterlin paradox) still holds, notable: at a point in
time both among and within nations, happiness varies directly with income, but over
time, happiness does not increase when a countrys income increases.13 They note that
in the short run, happiness does fluctuate with income because happiness falls during
Easterlin, R. 1974. Does Economic Growth Improve the Human Lot? Some Empirical Evidence,
Nations and Households in Economic Growth: Essays in Honor of Moses Abramovitz, (eds) David, P. and M.
Reder, 89125. New York: Academic Press.
12
Powdthavee, N. 2010. The Happiness Equation: The Surprising Economics of Our Most Valuable Asset.
London: Icon Books.
13
Easterlin, R. A., L. A. McVey, M. Switek, O. Sawangfa and J. S. Zweig. 2010. The Happiness-Income
Paradox Revisited. Proceedings of the National Academy of Sciences 107 (52): 2246368.
11
17
18
Macroeconomics Simplified
recessions and rises during booms. But in the longer run, that is, over periods of 10 years
or more, there is no correlation between economic growth and happiness. But as the
World Happiness Report notes, this finding depends on the data sources used for measuring happiness, so one needs to be cautious about drawing firm conclusions. Even so,
the report strongly endorses the need for policy makers to target many other aspects of
human welfare besides income (health, education, work satisfaction, psychological wellbeing, social solidarity, etc.) in order to promote happiness and human development.14
14
CHAPTER 3
Source: Authors.
20
Macroeconomics Simplified
well and otherwise to interfere as little as possible and only when market failures are
obvious.
This chapter outlines the basic stylised neoclassical general equilibrium macroeconomic model. Markets are assumed to be perfectly competitive, that is, characterised by
many buyers and sellers, homogeneous products, free entry and exit, perfect knowledge
and perfect mobility of capital and labour (factor mobility). Following Lon Walras, the
first general equilibrium theorist, economies are modelled as a continuous competitive
exchange equilibrium in which a mythical auctioneer is assumed to manage an instantaneous bidding process between buyers and sellers in all markets, thereby ensuring that
market-clearing is reached. This process, known as ttonnement, that is, making offers
to buy and sell which are carried out when an economy-wide set of market-clearing
prices is found, is central to neoclassical dynamics. The model assumes instantaneous
adjustment in all markets (i.e., the role of an auctioneer is over in a flash and functions
merely as a metaphor to justify the market-clearing assumption).
exactly in the same proportion. In mathematical terms, this means that all supply and
demand functions, with prices taken as independent variables and quantity as a dependent
one, are homogeneous of the zero degree.2
As discussed in more detail later, neoclassical models assume that the equilibrium
values of real variables (i.e., employment and output) are determined in the labour and
goods markets, whereas nominal variables (in current prices) are determined in the
money market (banking and financial sector). Given that real variables are not affected
by changes in money supply in this model (because all markets are assumed to be clear),
money is said to be neutralthat is, it facilitates exchange, but does not affect the overall income or output. The stylised neoclassical model accordingly has two parts in it: a
theory of how equilibrium output and employment are determined, and a theory of how
the aggregate price level is determined.
21
22
Macroeconomics Simplified
y=f (N)
y4
y3
y2
y1
N
N1
N2
N3
N4
Short-term models of the business cycle and income determination thus regard
capital and technology as constants. With technology and capital as constants, output
becomes a function of employment (as shown in Figure 3.1).
Given that any additional employment has to be combined with a fixed capital stock
and technology, the model assumes that each additional worker contributes less to output than is contributed by their already-employed fellow workers. This can be seen in
Figure 3.1. As employment increases from N1 to N4, output rises from y1 to y4 but by everdecreasing amounts. The marginal product of labour (MPN) thus decreases as employment increases. Although each additional worker contributes positively to output, the
ever-flattening shape of the curve indicates diminishing marginal returns to labour.
In agriculture, of course, output fluctuates as a result of external shocks such as rainfall, locust infestations. Thus, it is possible for employment and technology to remain
constant, but for output to fluctuate for these reasons. Thus, in case of a bumper crop,
every level of employment would be associated with a higher output, that is, the production function would shift up (i.e., N1 might now be associated with y2 and N2 with y3 in
Figure 3.1).
The demand for labour curve can be derived from the production function as follows:
Assuming perfect competition, firms will hire workers until the (average) money wage
(W) is equal to the general price level (P), multiplied by the marginal product of labour
(MPN).
W = P MPN.
(3.2)
The marginal product of labour (MPN) is the change in output per unit change in the
quantity of labour employed. MPN is thus the slope of the production function. The additional revenue associated with the employment of an additional worker is given by PMPN.
As long as the cost of hiring a worker (i.e., the wage) is less than the revenue gained, firms
will continue to hire workers. As more workers are hired, PMPN declines. Employment
will thus increase until W = PMPN. Firms have no incentives to increase employment
beyond this point as marginal cost will exceed marginal revenue from such employment.
W = P MPN can also be expressed in terms of the real wage:
W/P = MPN,
(3.3)
W/P = the real wage (i.e., money wage deflated by the price index).
In short, assuming a fixed capital stock and perfect competition, the real wage (W/P)
will equal the marginal product of labour (MPN). Once this relationship is established,
the demand for labour function can be derived from the production function.
Consider employment level N1 in Figure 3.2. As can be seen from the production
function (the left-hand diagram), N1 is a relatively low level of employment and hence is
associated with a relatively high marginal product. The slope of the production function
(which is equal to the MPN) is relatively high. Let us assume that MPN at N1 = `800. Now
consider employment level N2, which is a relatively higher level of employment associated with a correspondingly lower MPN. Assume that MPN at N2 = `600.
Given the assumption of perfect competition, this model depicts firms as employing
workers up to the point where the real wage equals the marginal product of labour. This can
be expressed in terms of demand for labour function (DN). As is evident from the demand
for labour function, when the real wage is `800, then only N1 amount of employment will
be demanded. When the real wage is `500, then more workers (N2) will be employed. The
demand for labour function is thus a negative function of the real wage.
y
W/P
y=f (N)
`800
`600
DN
0
N1
N2
N1
N2
23
24
Macroeconomics Simplified
W/P
(W/P)2
y2
y1
(W/P)1
DN2
DN1
N1
N1
Note that the demand for labour function is drawn for a given production function.
If, for example, we move to a longer-term scenario and allow for an increase in capital
stock, then the production function will shift up, and the demand for labour function
will shift outwards accordingly. This is shown in Figure 3.3. Technical change (which
increases productivity) has the same effect.
The supply of labour (SN) is a positive function of the real wage (Figure 3.4.). The
higher the real wage, the greater will be the number of workers offering their services to
employers. By making the supply of labour a function of the real wage (rather than the
nominal, i.e., money, wage), the model assumes that workers value their wages in terms
W/P
SN
of purchasing power. Thus, for example, if the money wage was `600 and the price index
was 2, then the real wage would be equal to `600/2 = `300. If, however, the economy
experienced some inflation and the price index rose to 3, then the real wage would fall
to `600/3 = `200. Although the money wage (`600) remains the same, the number of
workers prepared to offer their services will decline owing to the fall in the real wage.
Note that the supply of labour function is drawn for a constant labour force. If the
labour force were to expand (as happens when soldiers re-enter the labour market after
a war, or as it becomes more socially acceptable for women to work in an industry), then
the labour supply curve would shift outwards.
For the labour market to be in equilibrium, the amount of labour supplied must equal
the amount of labour demanded: SN = DN. The equilibrium point is shown in Figure 3.5
by (Nf,(W/P)f). At that point, all the workers who are prepared to offer their services to
employers at the going wage ((W/P)f) will have jobs. Those who still want jobs, but are
not prepared to work at that equilibrium wage are seen as choosing not to work, that is,
withdrawing from the labour force. Thus, there is no involuntary unemployment and
full employment is said to prevail.
If, however, the real wage happened to be (W/P)1, then the labour market would be
characterised by excess supply. At (W/P)1, only N1 amount of workers will be demanded,
yet N1 amount of labour will be supplied. In other words, N1N1 workers will be unable
to find jobs at the going wage (W/P)1. They will fall into the category of involuntarily
unemployed.
W/P
SN
(W/P)1
(W/P)f
(W/P)2
DN
N2
N1
Nf
N1
N2
25
26
Macroeconomics Simplified
Armed with the above understanding of the production function and the labour market, we can now proceed to derive the neoclassical aggregate supply (AS) curve, which
depicts the relationship between output and the price level.
The AS curve can be derived by means of a four-quadrant diagram (Figure 3.7). The
production function is reproduced in the southeast quadrant of the diagram. Although
the production function is drawn at an unusual angle, you ought to be able to see that
output increases as employment increases (from the origin down towards the bottom of
the page), albeit at a decreasing rate.
The labour market is reproduced in the south-west quadrant. Again, it should not
take you too long to get used to looking at the labour market from this angle. As the real
wage increases (from the origin out towards the left), the supply of labour will increase
and the demand for labour will decrease.
The north-west quadrant depicts the average money wage (W) in the form of a rectangular hyperbola. As indicated in Figure 3.8, a rectangular hyperbola with W/P on the
horizontal axis, and P on the vertical axis, will have the value of W for all points along it.
This is because the rectangular hyperbola has the form: xy = constant. In the north-west
quadrant of the diagram above, this translates into W/PP = W.
y
y = f(N)
yf
Ne
W/P
NS
(W/P)e
ND
Ne
27
28
Macroeconomics Simplified
Rectangular
hyperbola
W
W/P
SN
Production
function
Labour
market
DN
y = f (N)
N
(1,10)
10
(2,5)
(5,2)
(10,1)
xy = 10
1
10
P2
P1
W2
P3
W1
0
W/P
(W/P)3
(W/P)1
(W/P)2
29
30
Macroeconomics Simplified
AS
P3
P1
W3
W/P
P2
W1
W2
(W/P)2
(W/P)1
(W/P)3
y1
N2
N3
N1
N3
N2
SN
DN
y = f (N)
The AS curve depicts the relationship between the aggregate supply of real output
(y) and the price level (P). When plotting the AS curve, it is useful to begin by finding
the combination of y and P associated with equilibrium in the labour market. As is clear
from the south-west quadrant, a real wage of (W/P)1 will clear the labour market and
determine the full-employment level of N1. Given N1 amount of employment, the production function in the south-east quadrant tells us that the full-employment level of
output (y1) will be produced. To find the price level associated with y1, we need to track
the real wage (W/P)1 up into the north-west quadrant. Assuming that W1 is the ruling
money wage, the price level associated with the labour market-clearing real wage is P1.
The co-ordinate (y1,P1) is our first point on the AS curve.
To obtain a second point on the AS curve, pick a lower price level, P2. A lower price
level, in combination with the same money wage (W1), will result in a higher real wage
(W/P)2 (note that (W/P)1 = W1/P1 and that (W/P)2 = W1/P2). The higher real wage will
result in disequilibrium in the labour market: at real wage (W/P)2, N2 labour will be
demanded, yet N2 will be supplied. In other words, N2N2 people will be unemployed.
Assuming that wages are flexible, the labour market will respond to this situation of
excess supply of labour by bidding down the money wage until the equilibrium real wage
is restored. The diagram shows that money wages will fall from W1 to W2, at which point
the equilibrium real wage will be restored (note that (W/P)1 = W1/P1 = W2/P2). As a result
of this adjustment in the labour market, the price level P2 will also be associated with the
full-employment level of output y1. The second point on the AS curve is thus y1, P2.
To obtain a third point on the AS curve, pick a higher price level, P3. Assuming the
original money wage of W1, the higher price level, P3, will result in a lower real wage of
(W/P)3. At this lower real wage, only N3 of labour will be supplied, yet N3 of labour will
be demanded. Given our assumption of flexible wages, this situation of excess demand
in the labour market will result in bidding up of the money wage so as to restore the
equilibrium real wage. As shown in the diagram, money wages will rise to W3, at which
point the market-clearing real wage will have been restored (note that (W/P)1 = W1/P1
= W2/P2 = W3/W3).
By joining the three points together, we get an AS curve which is perfectly inelastic at
the full-employment level of output (y1). Other price levels can be considered. However,
given the assumption of flexible money wages, the labour market will always adjust to
full employment through changes in money wage. This means that N1 of employment
will always be created, which in turn implies (via the production function) that the fullemployment level of output, y1, will always be produced. Thus, the AS curve will always
be vertical at the full-employment level of output. Changes in the price level have no
effect on real output. The classical dichotomy thus prevails.
Says law
31
32
Macroeconomics Simplified
of savings and the demand for investment determines the equilibrium rate of interest. This
is shown in Figure 3.11. Put differently, fluctuations in the interest rate ensure that savings and planned investment will always be equal. Assuming a closed-economy model
with no government sector, income is divided between consumption (C) and savings (S)
as follows:
y = C(i) + S(i). (3.4)
Expenditure will consist of:
E = C(i) + I(i). (3.5)
Galbraith, J. 1994. The World Economy Since the War: A Personal View. pp. 8081. London: Macmillan.
Lange, O. 1942. Says Law: A Restatement and Criticism. (eds.) Lange, O. F. McIntyre and T. Yntema,
Studies in Mathematical Economics and Econometrics. pp. 4968. Chicago, IL: University of Chicago Press.
5
Becker, G. and W. Baumol. 1952. The Classical Monetary Theory: The Outcome of the Discussion.
Economica XIX, November: 276355.
3
4
Supply of loanable
funds (i.e., savings)
ie
Demand for
loanable funds
(i.e., investment )
Se, Ie
S,I
S(i) = I(i).
Note that in national accounts, savings are by definition equal to gross domestic capital formation. In other words, S = I + change in inventories. The national accounting
identity should not be confused with the equilibrium condition S = I. According to the
Keynesian model, there is no inherent reason why S should equal planned investment
(I). In a neoclassical model, by contrast, savings and investment will always be brought
into line by fluctuations in the interest rate. Hence, S(i) will always equal I(i) and therefore the economy will always be in equilibrium. Assuming wage flexibility, such equilibrium will be at a full-employment level.
If interest rate fluctuations ensure the equivalence of savings and investment, then
there will never be deficient demand. For example, consider what happens in the model
if households decide to save more and spend less. The increase in savings (represented
in Figure 3.12 as an outward shift in the savings function from S1 to S2) would result in
an equivalent cut in consumption (C). However, because the interest rate has adjusted
downwards from i1 to i2 as a result of the outward shift in S, investment increases by
exactly the same amount as the drop in C.
In other words, changing the level of savings does not affect total aggregate
demandit merely affects its composition (i.e., less consumption, more investment).
The implication here is that there will be no income which is neither invested nor
consumed, and hence there will be neither shortages or gluts nor unexpected changes
in inventories.
33
34
Macroeconomics Simplified
S1
S2
i1
i2
C
I1
S,I
(3.7)
For example, if the stock of money in an economy was `10 million, and each bank
note exchanged hands 40 times during the year in income-generating transactions, then
the nominal value of income that year must be `10 million 40 = `400 million. To isolate the effect of price changes, the following manipulations are useful:
MV = Y,
y = Y/P,
Y = Py,
MV = Y can therefore be rewritten as:
MV = Py,
MV/P = y. (3.8)
Given that V is assumed to be constant and that y is assumed to come to rest at full
employment, any change in M will automatically translate into a change in P. This can
easily be seen by continuing the above example where M = `10 million; V = 40; Y = MV
= `400 million.
If P = 1, then y = `400 million/1 = `400 million. If the money supply is doubled, and
V stays constant (by assumption), then nominal income (Y) rises to `800 million (i.e.,
`20 million multiplied by 40). However, given that real output is assumed to be at a fullemployment level, this neoclassical model assumes that prices will simply be driven up
until the real value of national income is restored at `400 million. The equations imply
that the price level would rise until it is doubled. In other words, if money supply doubles, so will the price level. Increasing the money supply will have no lasting effect on
real income. It will simply fuel inflation.
An alternative cash-balance version of the Quantity Theory of Money was developed
by Alfred Marshall,6 A. C. Pigou,7 John Maynard Keynes in his youthful period8 and
D.H. Robertson.9 This Marshallian version can be summarised in terms of the following
equation:
M = kPY,
(3.9)
where (M) is the exogenously determined money supply, (k) is that fraction of money
income (PY) which people want to hold in the form of cash or cash balance, (P) is the
general price level and (Y) is total output. The above equation states the equilibrium
condition in the money market.
According to the cash-balance version, the value of money is a function of the demand
for money which is influenced by the demand for cash balances. This version of the
Quantity Theory assumes that when people demand more cash balances, it will reduce
their demand for commodities and as a consequence the price level will fall. Conversely,
as people demand less cash balances in their hands, this will lead to higher demand for
goods and services and as a result, the price level will increase.
Both versions of the Quantity Theory of Money consider only the transaction or
active demand for money and neglect the speculative demand for money or the demand
for idle cash balances. The demand for money is discussed in more detail in Chapter 6.
Marshall, A. 1960. Money, Credit and Commerce. New Jersey: M Kelley Publishers.
Pigou, A. C. 1949. The Veil of Money. London: Macmillan.
8
Keynes, J. M. 1923. A Tract on Monetary Reform. London: Macmillan and Co.
9
Robertson, D. H. 1966. Essays in Money and Interest. London: Fontana Library.
6
7
35
36
Macroeconomics Simplified
supply at a stable and predictable rate in order to accommodate the expansion of output
whilst avoiding inflation. Under these conditions, as the claim goes, business cycles will
be avoided.
The Lucas argument is flawed theoretically and empirically. Firstly, the theoretical
assumption that fluctuations in output are caused by people, confusing relative with
general price changes lacks credibility. It is an unconvincing microfoundation. People
certainly cannot predict the future, but by reading newspapers and listening to radio,
they ought to be able to pick up information very quickly about changes in the general
price level.
Secondly, the model fails to fit reality, as recessions tend to be deeper and longer than
predicted by such models. There is also something unsatisfactory about the assumption
that peoples choices concerning how much to work and produce are simply based on
wages and prices determined competitively in the market. Unlike the real-world, New
Classical economic agents are never constrained by a shortage of jobs or customers.
Assumptions about perfectly competitive labour markets are the hallmark of New
Classical thinking. They are central to the modern variant of New Classical explanations of the business cycle. The recent real business cycle theory developed by Kydland
and Prescott (which has largely superseded the Lucas explanation in New Classical circles) attributes declines in output to productivity shocks resulting from changes in technological capacity.13 The argument goes something like this: a change in technological
capacity may cause the marginal productivity of labour to fall, which in turn reduces
the amount of labour demanded. As the wage falls, people increase leisure and reduce
consumption.
Like the Lucas explanation, real business cycle theories do not stand up well to
empirical testing. Most observers remain sceptical about unexplained changes in technological capacity being at the root of recessions. Even more controversial, however,
is the real business cycle theorys claim that given individual tastes and available technology, levels of output and employment are efficient, even in recessions. The classical dichotomy is reinstated with a vengeance.14 Attempts by government to boost the
economy are regarded as being at best ineffectual, and at worst harmful. As Mankiw
puts it:
Of all the implications of real business cycle theory, the optimality of economic fluctuations is perhaps the most shocking. It seems undeniable that the level of welfare is lower in
a recession than in the boom that preceded it.15
13
See, Kydland, F. and E. Prescott. 1982. Time to Build and Aggregate Fluctuations. Econometrica 50 (6)
and Snowdown, B., H. Vane and P. Wynarczyk. 1994. A Modern Guide to Macroeconomics: An Introduction to
Competing Schools of Thought. Chapter 6. Aldershot: Edward Elgar.
14
See, Plosser, C. 1989. Understanding Real Business Cycles. Journal of Economic Perspectives 3 (3):
5177.
15
Mankiw, Real Business Cycles, 83.
37
38
Macroeconomics Simplified
Keynesians agree. They start out with the assumption that recessions and their
accompanying human misery are the result of massive market failure that needs correction through judicious state intervention.
AS1 AS2
P1
W/P
W1
W2
(W2/P)1
(W2/P)2
y1
y2
N1
SN1
N2
SN2
y = f (N)
DN
Using a four-quadrant diagram, show what happens to the neoclassical supply curve
when there is a drought which reduces agricultural output?
P
AS2 AS1
P1
W/P
W1
y
y2
SN1
y1
N1
y = f (N)
DN N
If we assume that drought has no impact on the economy other than to reduce agricultural output, then the AS curve will shift backwards as agricultural output falls at all
levels of employment (i.e., the production function shifts inwards).
39
CHAPTER 4
In contrast to neoclassical thinkers, Keynesians assume that markets function imperfectly, and that individual maximising behaviour in the presence of uncertainty can
lead to socially irrational outcomes. As Mankiw puts it, the Keynesian school believes
that understanding economic fluctuations requires not just studying the intricacies
of general equilibrium, but also appreciating the possibility of market failure on a
grand scale.1
Keynesians believe that there are no inherent or inevitable reasons why savings
should equal investment, or why market forces should result in full employment. Inthe
Keynesian world view, prices do not adjust quickly and economic adjustment takes
place primarily through changes in output and employment. For most Keynesians,
the role of an economist is to develop policies which nudge the economy towards full
employmentand not to waste time developing Walrasian general equilibrium models
which are relevant only in the long run (if at all).
John Maynard Keynes on the Long Run
In the long run we are all dead. Economists set themselves too easy, too useless
a task if in tempestuous seasons they can only tell us that when the storm is
long past the sea is flat again.2
Unlike his Treatise on Money (1930), which explored the problem of economic fluctuations rather than extended depressions, the General Theory was a specific response to
the dire economic situation of the time. As Skidelsky notes:
41
42
Macroeconomics Simplified
To read the book in an empty landscape is no longer possible, indeed it was not possible
then. It appeared at the tail end of the greatest depression in modern history, and straightaway became a central part of the argument about what caused it and what could be done
to cure it, and prevent further depressions in the future.4
The Great Depression was a serious, unparalleled economic disaster. The immediate
cause was the US stock market crash on Black Thursday, 24 October 1929, but this was
preceded by another acknowledged cause: overheated asset prices and tight monetary
policy (which was an attempt to stem the tide of stock market speculation). When the
stock market bubble burst in the context of this constrained demand, the financial crisis
amplified its way through the real economy. The USA economy sunk into a protracted
depression, experiencing massive retrenchments and bankruptcies on a wide scale. The
shock reverberated across international boundaries, especially in Europe, but also in
many developing countries including India. It took the Second World War to rid the
world finally of the recessionary fall-out.
Writing in the 1930s, it is hardly surprising that Keynes rejected the neoclassical
market-clearing model. He argued that it was largely irrelevant to the economic society
in which we actually live, with the result that its teaching is misleading and disastrous
if we attempt to apply it to the facts of experience.6 Keynes believed that the neoclassical
Skidelsky, John Maynard Keynes, p. 538.
See, Manikumar, K. 2003. A Colonial Economy in the Great Depression, Madras (19291937). Hyderabad:
Orient Blackswan.
6
Keynes, J. M. 1936. (Reprinted in 1964). The General Theory of Employment, Interest and Money. p. 3.
London: Macmillan.
4
5
vision of perfectly functioning markets leading to full employment was a special case
only, whereas his unemployment equilibrium was the more general casehence
Keyness General Theory.
Like the orthodox economists of the time,7 Keynes believed that individuals were capable of rational economic thinking. However, in contrast to standard perfect-competition
analysis, he argued that economic decision-making, particularly as it pertained to
investment, was profoundly structured by imperfect information, business confidence,
risk and, most importantly, uncertainty. Business people, in his opinion, operated in an
uncertain murky world where information was often unavailable or too expensive (in
terms of time and effort) to obtainand where decisions were often made on the basis
of emotions or gut instincts. Under these conditions, Keynes insisted that it was incorrect to depict the macroeconomy as being driven towards some socially optimal fullemployment position by Adam Smiths invisible hand.
Todays New Keynesians have explored further the issue of decision-making under
conditions of uncertainty and imperfect information. There is work showing that if you
remove the assumption of perfect rationality, then less than perfect economic outcomes
will result. Akerlof, for example, argues that it is often rational for economic agents to be
near-rational, that is, to operate according to good rules of thumb rather than invest
extra time and money in getting better information. An economy full of near-rational
individuals working in a context of less than perfect competition could very well end up
in a Keynesian world of strong fluctuations in output and employment.8
Skidelsky on Keynes
Marx accused economists of his day of abstracting from the class struggle; Keynes
accused economists of his day of abstracting from the existence of uncertainty.
For Keynes, most of the things which go wrongand rightin decentralised
market economies stem from the central fact that human beings take decisions
in ignorance of the future. Ignorance enters into all the motives for forwardlooking action, investing them, at the limit, with the character of dreams and
nightmares.9
Keynes started out with the assumption that the circular flow of income had many
leakages and that especially in times of insecurity, income can be hoarded, either in cash
7
The orthodox economists of the time were the neoclassical economists (Pigou, Marshall, etc.). Keynes
is responsible for much semantic confusion by labelling these economists as classical economists, when in
fact they were neoclassical. The label classical is normally reserved for the likes of Smith, Ricardo and Marx.
8
See, Akerlof, G. A., and J. L. Yellen. 1985. A New Rational Model of the Business Cycle with Wage and
Price Inertia. Quarterly Journal of Economics, 100 (supplement): 82338.
9
Skidelsky, R. John Maynard Keynes, p. 539.
43
44
Macroeconomics Simplified
or by banks fearing to lend to risky borrowers. In such conditions, savings will exceed
investment and aggregate demand (i.e., total expenditure) will fall short of aggregate
supply (AS)and thus Says law would no longer hold.
Keynes believed it was the duty of government to counter the failings of market processes through judicious application of economic policy. This, of course, implies a faith
in the ability of civil servants to understand the macro picture, and act reasonably and
appropriatelyan idea which has earned Keynesianism much criticism from both the
left and the right. Whether justified or not, Keyness General Theory provided a theoretical rationale for the interventionist demand-management economic policies that have
become associated with Keynesianism.
The General Theory is the most influential economics book ever written, with the
possible exception of Karl Marxs Capital. Yet many of Keyness ideas were poorly developed, obscurely expressed, and often consisted of little more than throw-away lines.
As a result, economic theorists have been arguing ever since about what Keynes really
meant, and even what Keynes should have meant.10 (They have been doing the same
with Marx, but that is another story.)
Because The General Theory is long, complex and difficult to understand, Keyness
theories have entered into standard economic texts via interpreters. The earliest of these
was Hicks, who in 1937 published his very famous summary of Keyness ideas in terms
of an ISLM framework.11 Reading this has subsequently become a rite of passage for all
aspirant economists.12 Indeed, you too are about to experience this necessary trauma in
later chapters.
Textbook interpretations are inevitably simplifications that give priority to some
ideas over others.13 Nevertheless, Keyness most powerful and clear ideanamely that
an economy can come to rest at a less than full-employment equilibrium owing to a deficiency in aggregate demandcan be represented in terms of a simple model of income
determination (the Keynesian Cross discussed later), or in terms of a simple model of AS
and aggregate demand (and its more complex version presented in Chapter 8).
I shall argue that the postulates of the classical theory are applicable to a special case only
and not to the general case, the situation which it assumes being a limiting point of the
possible positions of equilibrium. Moreover, the characteristics of the special case assumed
by the classical theory happen not to be those of the economic society in which we actually
live, with the result that its teaching is misleading and disastrous if we attempt to apply it
to the facts of experience.14
It is thus ironic that the neoclassical synthesis model (discussed in Chapter 8) ends
up treating the Keynesian case as a special short-run situation explainable within the
confines of classical theory as caused by a malfunctioning labour market. For Keynes,
the classical approach was wrong headed, and he devotes his second chapter to explaining why. Then, in Chapter 3, he introduces his centrally important principle of effective
demand. As Alvin Hansen notes, this chapter is a highly important part of Keyness
epoch-making book. This chapter is of special significance because here, after repeated
failures, an impressive attack was at long last made upon Says law.15
Keynes argued that supply does not create its own demand (as Say argued) but rather
that the level of employment (and hence output) depended on effective demandthat
is, actual spending and the expectations that firms make about it. As Keynes put it:
[I]n a given situation of technique, resources and factor cost per unit of employment, the
amount of employment, both in each individual firm and industry and in the aggregate,
depends on the amount of the proceeds which the entrepreneurs expect to receive from
the corresponding output.16
Keynes goes on to describe a simple model which captures the substance of the General Theory:
Let Z be the AS price of the output from employing N men, the relationship between Z and N
being written Z= (N), which can be called the AS Function. Similarly, let D be the proceeds
which entrepreneurs expect to receive from employment of N men, the relationship between
D and N being written D = f(N), which can be called the Aggregate Demand Function.
Now if for a given value of N, the expected proceeds are greater than Z, there will be an
incentive to entrepreneurs to increase employment beyond N and, if necessary, to raise
costs by competing with one another for the factors of production, up to the value of N for
which Z has become equal to D. Thus, the volume of employment is given by the point of
intersection between the aggregate demand function and the AS function; for it is at this
point that the entrepreneurs expectation of profit will be maximised. The value of D at
the point of aggregate demand function, where it is intersected by the AS function, will be
called the effective demand.17
Keynes, The General Theory, p. 3.
Hansen, A. 1953. A Guide to Keynes. New York: McGraw-Hill.
16
Keynes, The General Theory, p. 24.
17
Ibid., p. 25.
14
15
45
46
Macroeconomics Simplified
There are many complex ideas buried in these two paragraphs. Indeed as Keynes
remarks, the rest of The General Theory is largely occupied with examining the various
factors upon which these two functions depend.18 But the Keyness central insightthat
expectations matter in determining the level of employmentis evident in the above
quote. In essence, employment is ultimately determined by effective demand which is
that level of spending in the economy which equals what firms expected to be the case.
Put differently, if actual spending was higher than what firms expected would be the
case, then they would hire more workers and produce more output until actual spending
in the economy = expected spending = effective demand. And, if actual spending was
less than expected spending, firms would find themselves with unsold stock, and so they
would adjust their expectations downwards and reduce employment and output.
The representation of AS and aggregate demand in relation to employment is an
essential characteristic of Keynes approach. His argument can be expressed in graphical form in Figure 4.1. The aggregate demand price function slopes upwards because as
total employment rises, there is more spending in the economy, and hence firms expect
sales to sell more output. At full employment, Lfull, employers expect no further increase
in the demand for their output and so the curve becomes infinitely elastic at that point.
The AS price function also increases with employment. We depict it as rising relatively
slowly at first (as labour is abundant and so wages are initially subdued and the costs of
production rise slowly), but then as employment increases, the AS cost function rises
faster as costs are driven upwards as full employment approaches, at which point the
curve becomes perfectly inelastic.
In Figure 4.1, D is the point of intersection between the AS price and aggregate
demand price functions. At levels of employment below L1, aggregate demand is greater
than AS which means that firms had underestimated demand and that they could make
money by hiring more workers and increasing output for the booming market. Employment and output will thus rise. At levels of employment above L1, it makes sense for
firms to reduce their employment and output as the revenues they get from sales is less
than their production costs. Employment and output will thus fall.
What this model shows, in essence, is that an economy can come to rest at a less
than full-employment position (i.e., there is involuntary unemployment equal to the
difference between Lfull and L1) because of a shortfall in aggregate demand. The centrally
important policy implication is that if a government can boost aggregate demand,
then it can push an economy out of recession and towards full employment. Expansionary fiscal and monetary policies can help boost demand, potentially from D1 to
D2 in Figure4.1 (and we use the ISLM model to explore this further in Chapters 6
and 7). The ASAD neoclassical synthesis model in Chapter 8 introduces the role of
prices. The model assumes that expansionary economic policy boosts employment by
sending a signal to firms to increase employment (as sales increase), but it also makes
it economically feasible for firms to do so by inflating the price level and eroding real
wages (which lowers to cost to firms of hiring workers).
18
Ibid.
Z = (N)
Aggregate supply
function
Aggregate
supply
price,
Aggregate
demand
price
D2 = f(N)
D
D1 = f(N)
Aggregate demand
function
L1
Lfull
Level of employment
Tarshis, L. 1947. The Elements of Economics: An Introduction to the Theory of Price and Employment.
Boston, MA: Houghton Mifflin.
20
Harcourt, G. 2010. The Crisis in Mainstream Economics. Real World Economics Review 53: 4751.
19
47
48
Macroeconomics Simplified
(Box Continued)
Classics introduced the ISLL analysis which was changed to ISLM by Alvin
Hansen.) Oscar Lange21 also produced an interpretation of Keynes which was to
become the foundation for the neoclassical synthesis.22 Later in the 1948 edition of Paul Samuelsons classic introductory text book Economics, he introduced
the Keynesian cross analysis. Both these fix-price frameworks were followed by
the ADAS framework with flexible prices by Modigliani and others. Only postKeynesians such as Sidney Weintraub, Hyman Minsky and others tried to extend
the original ADAS theory of employment of Keynes further in to the analytics of
income distribution.
Source: Authors.
The various Keynesian models discussed later in the book are all interpretations and
adaptations of Keyness economics and many scholars believe that they are poor reflections of the richness of Keyness economic thought. Indeed, Axel Leijonhufvud argues
that it is necessary to distinguish clearly between Keynesian economics (such as the
ISLM and ASAD interpretations presented in this text book) and the economics of
Keynes, discussed above.23 At best, the mainstream Keynesian models capture key elements of his work, though often at the cost of ignoring many of his richer insights.
In any event, what is central to all Keynesian models is the idea that output adjusts in
response to aggregate demand, and that it is the expectations of firms about the level of
spending that is crucial. The simple one-sector Keynesian model captures this by being
entirely driven by output-adjustment dynamics.
Quantity Adjustment
In contrast to the neoclassical model, in which flexible prices ensure that the economy
only comes to rest at full employment, the simple Keynesian model assumes that equilibrium output and employment are ultimately determined by the point of intersection
between (planned) output and aggregate demand (i.e., actual spending). The key idea is
that entrepreneurs plan their output in accordance with expectations about demand. If the
business community expects the economy to be characterised by strong consumption and
Lange, O. 1938. The Rate of Interest and the Optimum Propensity to Consume. Economica V (new
series) No. 17, February: 1232.
22
Toporowski, J. 2012. Lange and Keynes, SOAS Department of Economics, working paper no. 170.
http://www.soas.ac.uk/economics/research/workingpapers/file74887.pdf. Accessed on 28 March.
23
For example, see, Leijonhufvud, A. 1968. On Keynesian Economics and the Economics of Keynes: A
Study in Monetary Theory. New York: Oxford University Press.
21
investment demand, then more workers are likely to be employed, and more output will
be produced. If, however, they decide that it will be weak, they will employ fewer workers and produce less output.
According to the Keynesian framework, the economy comes to restthat is, reaches
an equilibrium positionwhen aggregate demand (i.e., actual spending) is equal to
the level of demand expected by firms. In other words, when the amount of output that
firms produce (given their expectations about demand) is exactly sufficient to meet what
consumers are prepared to spend, then the economy will be in equilibrium. If aggregate
demand is greater than the expected (and hence is greater than the output supplied to
the market by firms) demand, then buoyant business conditions will be reflected in falling inventories. Firms will respond to their fall in inventories by increasing production
and hence also employment. If, however, aggregate demand is less than expected, then
firms will experience rising inventories and falling profits, and reduce employment and
output accordingly.
Note that aggregate demand (E) consists of the total amount of actual planned spending in the economy. In an open economy, E = C + I + G + X Z. Unlike the estimate of
Expenditure on the GDP (which appears in the national accounts), aggregate demand
in Keynesian models only includes planned investment in capital goods (i.e., excludes
inventory accumulation).
The distinction between planned investment, that is, investment demand (I), and
unplanned investment in inventories is central to the Keynesian process of output
adjustment. Investment in inventories is assumed to be mostly unplanned, in the sense,
that it fluctuates owing to unexpected changes in demand. When actual demand is
greater than expected, then stocks will disappear off shop shelves faster than expected.
In other words, inventory investment falls. This sends a signal to producers to increase
output until actual demand is equal to expected demand. When actual demand is less
than expected, shops will experience an unwanted accumulation in stocks and they will
reduce their orders to firms. This sends a signal to firms to reduce output until actual
and expected demand is equal.
Because the output adjustment story is so central to Keynesian thinking, simple
Keynesian models tend to assume that prices are fixed. This happens in the Keynesian
cross model (discussed later) and in the ISLM framework. However, this over-simplifies
the Keyness thought. Keynes argued that prices and output were likely to rise as demand
rose, especially as the economy moved towards full employment. The three-sector ASAD
model presented in Chapter 8 captures this dynamic more explicitly by including price
adjustment.
49
50
Macroeconomics Simplified
adjust their expectations about what the economy can absorb in terms of output, and
the economy will come to rest in a state characterised by unemployment and surplus
capacity. The model is thus demand-driven. Output adjusts accordingly. Pretty much
the same result will be obtained whether prices are assumed to be completely fixed, or
simply very slowly adjusting.
These ideas can be illustrated by means of a simple Keynesian model of income determination, the so-called Keynesian Cross. Assuming a closed economy (i.e., no foreign
trade), and no government sector, the model can be depicted as follows:
E = C + I (4.1)
C = a + bya>0 1>b>0. (4.2)
y = E (4.3)
where E = aggregate demand, C = consumption demand and I = investment demand.
Recall that as prices are assumed constant, the above variables are real and nominal. Equation 4.1 is a definitional relationship. It defines aggregate demand (E) as being
equal to the sum of consumption demand (C) and investment demand (I). It is equal to
the amount that is actually spent in an economy on consumption and investment.
Investment is exogenous (i.e., determined outside the model). In this simple model,
it is assumed that real income has no effect on investment. In the two-sector ISLM
Keynesian model, I is an endogenous function of the interest rate. As there is no financial sector in the simple Keynesian model of income determination discussed here
(and hence no interest rate in the model), investment appears as an exogenous variable.
Investment is thus represented as a horizontal line in Figure 4.2. Because exogenous
variables are determined by forces which have not been modelled by equations within
the model, any change in an exogenous variable is regarded as an autonomous change.
Consumption, however, is endogenous because it is determined within the model.
Equation 4.2 is a behavioural relationship which states that C is a linear function of real
C = a + by
a
I
0
income (y). Note that this is a radical departure from the neoclassical orthodoxy which
regards consumption as being a function of the interest rate (Chapter 3).
C is the dependent variable, and y is the independent variable in the Keynesian consumption function. The positive intercept (a) is exogenous and represents autonomous
consumption spending, that is, spending occurs even when real income is zero (note
that for an economy to do this, it would have to be living off savings).
The parameter (b) is the behavioural coefficient for real income and gives the slope
of the line. It tells us the rate at which consumption rises as income rises. It is, in other
words, the marginal propensity to consume out of income. It gives the proportion of
any increment in income which will be spent on consumption. As specified above, b is
greater than zero, but less than 1.
For example, if C = 15 + 0.8y, this tells us that consumption would be 15 even if
income was zero, and that for every increase in real income, 80 per cent will go towards
consumption. It is assumed that even if income rises or falls sharply, this fixed proportion will not change.
51
52
Macroeconomics Simplified
(Box Continued)
Neoclassical economists disagree with the Keynesian consumption function,
arguing that consumption is a function of expected income over the life cycle.
Modigliani, Ando and Brumbergs life-cycle hypothesis points out that people
maximise utility over their lifetimes, with young people being net borrowers,
middle-aged people net savers and retirees live on their savings. Milton Friedman
argued that people base their consumption on their permanent income (the
level of income they expect to persist) and save their transitory income (windfall gains) which they draw during times of unexpected shortfall. Such theories
imply that consumption is subject to strong smoothing forces and will not vary
sharply with changes in current income (as assumed by Keynesians).24
Source: Authors.
Equation 4.3, namely y = E, is the equilibrium condition. It states that for the model
to come to rest in a state of equilibrium, real output or income (y) must equal aggregate demand. In other words, actual spending (E) equals the amount producedwhich
implies that actual aggregate demand equals the level expected by producers. The economy will be in a state of quantity adjustment until y = E.
Equations 4.14.3 are the structural equations of the model. By manipulating them to
yield solutions for the endogenous variables (C and y) in terms of exogenous variables
and parameters only, we can create reduced form equations.
For example, to find the reduced form for y, substitute Equation 4.2 in Equation 4.1,
to get Equation 4.4. Then substitute Equation 4.4 in Equation 4.3 and re-arrange terms
to get Equation 4.5.
E = C + I, (4.1)
C = a + by, (4.2)
y = E, (4.3)
E = a + by + I. (4.4)
y = a + by + I,
y by = a + I,
y(1 b) = a + I,
y = (a + I). (1/1 b). (4.5)
See, Modigliani, F. and R. Brumberg. 1954. Utility Analysis and the Consumption Function: An
Interpretation of Cross-Section Data in Post-Keynesian Economics, ed. Kurihara K. New York: Rutgers
University Press.
Ando, A. and F. Modigliani.1963. The Life Cycle Hypothesis of Saving: Aggregate Implications and
Tests. American Economic Review 53 (1): 55-84.
Friedman, M. 1957. A Theory of the Consumption Function. New Jersey: Princeton University Press.
See also, Brown, W. 1988. Macroeconomics. New Jersey: Prentice-Hall, for a user-friendly summary of
the debates on the consumption function.
24
1/(1 b) is known as the multiplier. Equation 4.5 shows that any change in the exogenous variables (a and I) will result in a change in y equal to a change in the exogenous
variable multiplied by 1/(1 b). The multiplier effect shows that an autonomous increase
in demand will initially boost income by an equivalent amount, but that this increase
gives rise to further rounds of demand stimulus and output growth.
If we introduce the government sector into this simple model, then Equation 4.1 will
be altered as follows:
E = C + I + G. (4.6)
As can be seen in Equation 4.7, the government can influence y by adjusting its
expenditure (G). More specifically, if G goes up by an amount DG, then y will go up by
an amount DG/(1 b). Because b is less than 1 but greater than 0, the term (1 b) will
always be positive, but less than 1. This means that the increase in y (i.e., Dy) will always
be greater than DG. For example, assume that b (i.e., the slope of the consumption function) is 0.8. If G goes up by `2,000 billion, then y will go up by `2,000 billion divided
by 1 0.8, that is, by `10,000 billion. Dy = DG/1 b = `2,000 billion/1 0.8 = `2,000
billions/0.2 = `10,000 billion.
In other words, any increase in government spending has a much greater (or multiplied) effect on y. In the above example, the effect on y was 5 times greater than the
original increase in G. This is the result of a multiplier, that is, 1/(1 b), which in the
above example was 1/0.2 = 5.
Note that the multiplier is driven by the slope of consumption function, that is, by
the marginal propensity to consume. In the above example, the reason why an injection
of `2,000 billion of government spending results in a 5 times greater increase in real
income, is simply because the model assumes that people will spend 80 per cent of any
income they receive.
For example, the Government of Indias Mahatma Gandhi National Rural Employment Guarantee Act provided 2.55 billion person days of work to poor people in rural
areas in the financial year 201011, resulting in significant multiplier effects in terms of
higher incomes, improved economic outcomes and reduced distressed migration and
greater investment on small farms.25 According to recent estimates, the multiplier effect
of an expansion in agricultural output is high, in the region of about 13.4.26 This implies
See for example: Shar, M. 2009. Multiplier Accelerator Synergy in NREGA. The Hindu. http://www.
thehindu.com/2009/04/30/stories/2009043055630800.htm. Accessed on 28 March.
Binswanger-Mkize, H. 2012. India 19602010: Structural Change, the Rural Nonfarm Sector, and the
Prospects for Agriculture. Department of Agricultural and Resource Economics. Berkeley: University of
California. http://areweb.berkeley.edu/documents/seminar/Binswanger.pdf
26
Pal, B., S. Pohit and J. Roy. 2012. Social Accounting Matrix for India. Economic Systems Research 24
(1): 7799.
25
53
54
Macroeconomics Simplified
that for every `1,000 allocated to this programme, national Indian output could rise
by as much as `13,400 as this increased rural income is spent on other products (e.g.,
food, clothes, agricultural equipment, services), thereby boosting the incomes of people
selling those products, which in turn boosts their spending and the incomes of other
suppliers and so on.
This is why government spending on the poor, if appropriately targeted, can benefit
not only the poor but also the entire economy, resulting eventually in higher incomes
and tax revenues. It suggests that by manipulating the level of demand, the government
couldat least theoreticallyensure that all available resources were drawn into productive use.
The multiplier theory shows how relatively small injections (or withdrawals) can
have a multiplied effect on output. As Keynes explained, it is to the general principle
of the multiplier to which we have to look for an explanation of how fluctuation in the
amount of investment, which are a comparably small proportion of the national income,
are capable of generating fluctuations in aggregate employment and income so much
greater in amplitude than themselves.27
The idea that people will spend a constant proportion of current income gives the
multiplier its power. This assumption has been contested, the key idea being that consumption is smoothed over peoples lifetimes and hence is not subject to significant
fluctuation with income (see box above). Keynes himself acknowledged that there were
factors besides current income which affected consumptionincluding expectations of
future income, windfall changes in asset prices, large alterations in government policy
but that when it came to analysing fluctuations in economic activity (short-run cycles),
he concluded that the aggregate income measured in terms of the wage-unit is, as a rule,
the principal variable upon which the consumption constituent of the aggregate demand
27
28
function will depend.29 In other words, he believed that there was a sufficiently strong
relationship between current income and consumption that it significantly affected fluctuations in aggregate demandand hence also income.
Keynes simple model of income determination and the consumption-driven multiplier can be presented diagrammatically in terms of the famous Keynesian cross diagram
(Figure 4.3). The 45 line represents the equilibrium condition, y = E (i.e., Equation 4.3).
A 45 line converts any distance along the vertical axis (the ordinate) into an equal distance on the horizontal axis (the abscissa). At all points along the line, aggregate demand
is equal to output. For there to be exactly enough output to meet aggregate demand, it
must be the case that actual demand is the same as the level expected by firms when
making their output decisions. In other words, at all points along the 45 line, aggregate
demand = expected demand = output.
Keynes, The General Theory, p. 96.
Rao, V. K. R. V. 1952. Investment, Income and the Multiplier in an Underdeveloped Economy. Indian
Economic Review, February. (Eds.) A. N. Agarwala and S. P. Singh. Reprinted in 1958. The Economics of
Underdevelopment. London: Oxford University Press. pp. 20518.
29
30
55
56
Macroeconomics Simplified
E=y
E
E
Z
E1
X
45
y1
In Figure 4.3, the economy is in equilibrium at the point (y1, E1) where aggregate
demand (i.e., the E = C + I + G1 line) cuts the 45 line. Note that because G and I are
exogenous (and thus would be represented by horizontal lines) the slope of the C + I +
G1 line is b (i.e., given by the consumption function). It cuts the vertical axis at a + I +
G1 when y = 0.
In order to clarify exactly what is meant by the equilibrium position, consider point
X. At point X, aggregate demand exceeds output. In other words, actual expenditure
exceeds the expectations of firms. This is not a position of equilibrium because it means
that inventories will start dropping in response to the greater-than-expected demand. As
inventories start to drop, this sends a signal to firms to revise their expectations upwards
and produce more output. As they do this, output rises to y1, that is, until it is sufficient
to meet available demand. At (y1, E1), aggregate demand = expected demand.
By contrast, aggregate demand will be less than the output at point Z. In other words,
actual expenditure will be less than expected by firms. Inventories will start to accumulate in the form of unsold stocks. This will send a signal to producers to revise their
expectations downwards and cut back on production. As they do this, output falls to y1,
that is, until it is sufficient to match the reduced level of demand.
Given that all it takes to bring an economy to a point of equilibrium is equality
between actual and expected levels of expenditure, Keynes argued that equilibrium positions such as y1 could easily be positions of less than full employment. Because government spending is a large component of aggregate demand, and can be manipulated as
a policy variable, he argued that the state could help move the economy towards full
employment by increasing spending.
If government spending was to increase from G1 to G2, this would, through the effect
of the multiplier, increase y by substantially more than the increase in government
spending. As shown in Figure 4.4, output would rise from y1 to y2, which is more than
the increase in G.
However, there are many informational problems facing a government when it
attempts to boost an economy to full employment. Firstly, the government has to have
a reasonable estimate of what the full-employment level of output actually is. Secondly,
the governments economists must have a pretty good estimate of the value of the multiplier, and of the time it is likely to take entrepreneurs to revise their expectations sufficiently quickly (and appropriately) to create the additional desired output.
These are all fairly bold assumptions, and none of them appear (at least not in such
unqualified terms) in the General Theory. Nothing, in fact, even approximating the
Keynesian cross diagram appears in Keyness book.31 It is highly unlikely that Keynes
would have subscribed to the notion that expectations could be manipulated through
fine-tuned demand adjustment on the part of government to bring the economy to a
state of full-employment equilibrium. Keynes was certainly of the view that boosting
demand could help in times of recession, but he would have balked at the idea that
demand-led government spending could target full-employment output in such a neat
and effective manner.
E=y
C + I + G2
E2
G
C + I + G1
E1
y
45
y1
y2
57
58
Macroeconomics Simplified
As explained in detail in the introductory chapter, after the publication of the General Theory, Keyness followers divided into two camps: the hydraulic Keynesians (who
favoured policy models such as the Keynesian cross analysis); and the fundamentalist
Keynesians like Joan Robinson who saw unstable expectations and uncertainty as forming the core of Keyness contribution (later disequilibrium and new Keynesian groups
were added to the Keynesian camp). The fundamentalists viewed the hydraulic interpretation as a bastardisation of Keyness thought. Modern-day fundamentalists are called
Post-Keynesians.32
Questions of legitimacy aside, the simple model of income determination does provide a framework for illustrating how an economy can come to rest at a position of less
than full employment. It is a one-sector model because it only models the real sector of
the economy, otherwise known as the goods market. This focus on the goods market
is pertinent as it highlights the reaction of producers to changes in demanda central
Keynesian dynamic. However, to do greater justice to Keyness ideas, it is necessary to
include the financial sector in the analysis. This requires developing a model with two
sectors, namely a goods market and a money market. This is done in Chapter 6, which
introduces the ISLM model.
demand to price cuts, the entrepreneur may rather choose to keep selling the inventories
at the original price, and instead slowly start cutting back on production and allow the
workforce to decline through retrenchment packages or natural attrition.
New Keynesian attempts to build more acceptable microfoundations concentrate
on different aspects of disequilibrium stories, such as the above. One strand of New
Keynesian research concentrates on the labour market, and specifically on why wages
tend to be above market-clearing levels as a result of turnover costs and efficiency wages
(i.e., higher than market-clearing wages being paid in order to generate higher productivity).33 The other strand concentrates on explaining why prices are sticky and adjust
slowly to changes in demand.
Let us consider first some of the New Keynesian arguments as to why wages are
sticky and persist at higher than market-clearing levels, thus resulting in unemployment. Possible reasons for such stickiness include efficiency wages, and insideroutsider
theories of wage determination.34
Efficiency wage theories argue that it does not pay firms to reduce wages, as this is
likely to have an adverse effect on morale and labour productivity. However, if workers
are paid above market-clearing levels, then they will have less incentive to shirk. They
are also likely to be more co-operative.
Lowering real wages during recessions is likely to harm relations on the shopfloor,
and undermine future attempts on part of the employers to gain the co-operation of the
workforce. Furthermore, some workers may choose to leave as a result of the wage cut,
and if firms have to replace them, then they will have to carry out the associated hiring
and training costs.
Insideroutsider theories start with the existence of turnover costs, and then add the
proposition that trained insiders are not perfect substitutes for untrained outsiders.
This implies that employed insiders can, to some extent, protect their wages against
undercutting by unemployed outsiders. Furthermore, given that employed insiders often
have to be relied upon to train newly employed workers, employers have an incentive to
keep on the good side of the already-employed workforce. Employing new workers at
lower wages could lead to harassment of new recruits by insiders, or to future demands
by such recruits for higher wages.
As regards New Keynesian arguments about sticky prices, or nominal rigidities in
their jargon, the basic contention is that firms are reluctant to change prices regularly,
and that small nominal rigidities at firm level add up to large macroeconomic effects.
This was first proposed in the menu costs literature which focuses on the technological
costs of changing nominal prices (literally, the costs of changing menus and price lists).35
33
Lindbeck, A. and D. Snower. 1988. Co-operation, Harassment and Involuntary Unemployment: An
Insider Outsider Approach. American Economic Review 78 (1): 16788.
34
This discussion is drawn from Greenwald, B. and J. Stiglitz. 1993. New and Old Keynesians. Journal
of Economic Perspectives 7 (1): 3334.
35
See Mankiw, N. G. 1985. Small Menu Costs and Large Business Cycles: A Macroeconomic Model of
Monopoly. Quarterly Journal of Economics 100: 52937.
59
60
Macroeconomics Simplified
Where firms are imperfectly competitive and face small barriers to price flexibility (such
as menu costs), then it can be shown that even small barriers can have large macroeconomic effects.36
Menu costs, however, tend to be small and there is little empirical evidence that they
are a significant source of even small nominal rigidities. Greenwald and Stiglitz have
provided a more convincing argument by adding the role of risk to the analysis of pricing decisions.
When a firm considers the various ways it might react, it perceives greater uncertainty about the consequences of price and wage adjustmentsbecause those consequences depend on the uncertain responses of rival firms, customers and workersthan
about the consequences of output adjustments. In fact, for those goods which can be put
into inventory, the only risk associated with producing too little is the risk associated
with higher production costs in the next period, when any inventory deficiency must
be made up.37
The issue of nominal rigidities is still an open question. As Romer reminds us,
economists do not have a good understanding of the price-adjustment policies of
firms, or even of the considerations that underlie their choices of policies.38 The task
of providing adequate microfoundations to Keynesian macroeconomics has a long
way to go.
Not everyone thinks this is a good research programme, however. There remains a
body of Keynesian and Post-Keynesian thinking which rejects the project of providing a
rationale (to neoclassical economists) for the alleged rigidity of money wages and prices.
According to James Tobin (Americas most prominent Keynesian), showing how nominal shocks can have real consequences was not Keyness idea:
I dont find it extremely convincingand Im sure Keynes would have notthat the whole
effective demand problem is that there are real costs of changing nominal prices on the
menu at the restaurant. I think Keynes would have laughed at the idea that menu costs
are a big enough resource-using problem to cause the Great Depression or any other
substantial losses of economic activity. Its not credible. If I had a copyright on who could
use the term Keynesian; I wouldnt allow them to use it.39
hegemony and most governments shifted towards more restrictive economic policies
which reduced the role of government.
However, the pendulum started swinging back during the 1990s. Slow growth in
many countries lead governments (and economists) to reconsider Keynesian principles
of demand management. In Japan, growth stagnated from 1992 onwards and was dealt
a particularly severe blow during the Asian crisis of 199798. In text book Keynesian
fashion, the Japanese government embarked on an expansionary fiscal policy to try and
boost spending. The government even went as far as to issue monetary coupons to poor
households (Keynes would have been impressed!). Paul Krugman, one of the leading
American economists, even wrote a book after the Asia crisis of 1998 called The Return
of Depression Economics.40
Krugman, however, was ahead of his time. The long boom of the 2000s rekindled
faith in the invisible hands of the market. It was only after the 2008 crisis that people
realised that the long consumption-fuelled boom rested on toxic assets (American
sub-prime mortgage bonds packaged into complex instruments which disguised their
risky nature) and that the regulatory authorities had failed to notice, anticipate or even
imagine that such a problem existed. The academic silver lining of this economic cloud
is that Keynesian insights have re-entered the mainstream. This new-found tolerance
for Keynesianism and suspicion of economic models based on perfect competition and
rational expectations has made macroeconomics a more open and exciting discipline
today.
Even some Marxists, who typically are suspicious of Keynes, given his role in articulating a theory which helps policy makers save capitalism, have surprising praises for
Keynes these days. For example, Samir Amin writes:
Keynes vision of the future of humanity was optimistic. He saw that the level of development
of the productive forces that had been acquired enabled humanity to emancipate itself from
the economic question (in his beautiful speeches to our great grandchildren). Asociety
that was freed from the chains of obligatory work was therefore possible: a society that
passed its time cultivating human relationships, a society that was truly emancipated and
cultivated. This objective, in its way, is none other than that of Marxs communism. Itis
the reason why capitalism is a system that is now obsolescent, whose time is now over.
The thinking of Keynes constitutes, Ibelieve, one of the examples that prove the rightness
of Marxs vision: humanity aspires to communism. Not only its popular classes (whom
Keynes distrusted), but even its greatest thinkers. Keynes was certainly not the first one
to have conceived this radiant future. Before him, the utopians had done so. ... However
the equally necessary reading of Keynes the economist is, in my view, disappointing. Of
course Keynes was far above the conventional vulgar economists of his day (and their
descendants, the pure economists of today). And his proposals constitute an approach to
reality infinitely more powerful than those of our miserable liberals.41
Krugman, P. 1999. The Return of Depression Economics. New York: Norton.
Amin, S. 2011. Ending the Crisis of Capitalism or Ending Capitalism? p. 190. Bangalore: Books for
Change.
40
41
61
62
Macroeconomics Simplified
CHAPTER 5
This chapter discusses the relationship between investment and interest rate as preparation for our subsequent discussion of the ISLM model. The ISLM model is a twosector model in that it includes both real (goods market) and financial (money market)
sectors. It is a simple, but useful, model because it allows us to explore the ways in which
financial variables affect real output, and vice versa.
The key connection between the goods market and the money market is the relationship between the interest rate (a financial variable) and investment spending (which,
being a component of aggregate demand, helps determine goods market equilibrium).
This chapter explores this relationship, pointing to the assumptions made about it in the
ISLM model. It should thus be regarded as background information to be absorbed
before moving on to the ISLM model discussed in Chapters 6 and 7.
64
Macroeconomics Simplified
government in 1995, do not have fixed coupon rates and the coupon rate is adjusted at
regular intervals (e.g., 6 months), typically according to yields on Treasury bills.1 For the
purposes of theoretical discussion in this book, we confine our attention to fixed coupon
rate bonds. During the life of a loan contract, the borrower makes regular fixed interest
payments to whoever holds the bond. At the end of the loan period, the principal (i.e.,
face value) is repaid to the bond-holder. A bond with no retirement date is known as
perpetuity. It only has a coupon rate and a face value, and the issuer is under no obligation to pay back the face value. The bond is thus a financial asset that pays a fixed
amount to the bond-holder indefinitely.
Bonds can be envisaged as tradeable I owe you (IOUs). The original bond purchaser (i.e., lender) is free to sell the financial asset on the secondary bond market
where prices are set by forces of supply and demand. The price of such second-hand
bonds could rise above or fall below their issue price (i.e., the face value). When that
happens, the effective interest rateor yieldfor todays buyer diverges from the coupon
rate of interest.
The fact that bonds can be traded in this way renders them risky investments as prices
can go up or down. For this reason, a range of derivative financial instruments have been
created. For example, collateralised debt obligations (CDOs), which pooled together US
mortgage bonds from different places and with different risk profiles, were created supposedly to reduce the exposure of buyers to particular mortgage bonds. Unfortunately,
the CDOs (and the rating agencies which approved them) did not take into account
the systemic risk that was building up, and which affected the entire mortgage market,
thereby eliminating the advantage of the CDO. The 2000s also saw the rise of credit
default swaps (CDS) which effectively allowed people to buy insurance on mortgage
bonds and CDOs. The models discussed in this text book do not take into account such
1
For more information on bonds in India, see the Reserve Bank of Indias Government Securities Market in IndiaA Primer. http://www.rbi.org.in/scripts/FAQView.aspx?Id=79
complex derivatives. While this is a limitation, the advantage is that it allows us to focus
on the key underlying dynamics which derivatives like CDOs and CDSs potentially render less risky in periods of economic calm, but amplify during times of crisis.
The yield would thus be equal to the coupon payment/bond price plus the change in the price of bonds.
65
66
Macroeconomics Simplified
were the same. In fact, the rate of return on all financial assets would rise. Bond yields
and other interest rates and rates of return thus tend to move in the same direction.
In the ISLM model, the only interest-bearing asset is a bond, and the only financial market is the secondary bond market. There are no banks or stock exchanges and
no internal sources of funding for firms. The only interest rate is the ruling bond yield.
Even so, the model still has explanatory power because financial rates of return tend
to move together.
I=I(i)
you save it and earn interest? If there was a stock market, and shares in your firm were
publicly traded, you might consider Tobins q, that is, the ratio of the market value of
your firm to the replacement cost of capital: if it was high, you might decide to expand
your business by purchasing more plant and equipment (i.e., investing). If it was low, you
might decide to purchase some shares (i.e., increase your stake in the companyan act
of saving). The ISLM model, however, does not have a stock market. It assumes that the
only way you can save is to purchase bonds in the bond market.
After investigating the bond market, you discover that your best bet is to buy a perpetuity with a face value of `100,000 and a coupon rate of 10 per cent. This means that
you will earn `10,000 a year from the bond. Under these circumstances, you would be
most unlikely to invest `100,000 in plant and machinery unless you expect to earn a
rate of return of 10 per cent or more. In short, investment in capital goods will only be
undertaken if the expected rate of return on that investment is equal to or greater than
the yield on bonds. If a very high return can be obtained from the bond market, then
only very profitable investment projects will be undertaken in the face of it. This implies
that at high rates of interest, investment will be low. Conversely, when returns are low,
more investment projects become viable, and hence investment spending will be high.
67
68
Macroeconomics Simplified
Obviously, the decision is much more complicated than this. When considering
whether to purchase a bond, you will also be worrying about its future market value.
Forming expectations about future prices is always a risky and fraught business.
Similarly, when trying to work out the rate of return on a proposed investment project
(by calculating the net present value [NPV]see box below), you will probably end up
making all sorts of guesses about future demand for the product, trends in cost prices,
potential competition, etc. You will also have to make a fairly long-term projection about
the path of costs and revenues over the life of the project.
Present Value
According to neoclassical theory, firms consider both the expected marginal
product of capital (discounted over time) and the real cost of capital, which
is affected both by interest rate and depreciation. The present value (PV) of a
capital asset that yields a stream of revenue (Y) over t years is:
PV = Y/(1 + i)1 + Y/(1 + i)2 + + Y(1 + i)t
where i = interest rate
The PV of an asset paying `5,000 per year for the next 3 years is (assuming an
interest rate of 10 per cent):
PV = 5,000/(1 + 0.1)1 + 5,000/(1 + 0.1)2 + 5,000/(1 + 0.1)3 = `4545.45 + `4123.23
+ `3759.40 = `12428.08.
Net Present Value
Rational firms use the NPV criterion when deciding whether to buy capital equipment or not. NPV is calculated by subtracting the supply price of capital (Sk),
that is, the price the firm must pay to acquire the machine, from the PV of the
capital asset:
NPV = PV Sk = Y/(1 + i)1 + Y/(1 + i)2 + + Y/(1 + i)t Sk.
The implication is that more capital investment will take place at lower interest rates than at high interest rates. The problem is that there is no way of knowing for sure what the future returns or future interest rates are going to be. As
Keynes pointed out many years ago, animal spirits and guesswork are important
determinants of investmenteven when the investment decision is made on a
purely rational basis.
Source: Authors.
Keynes believed that uncertainty and irrational fears and desires played a central role
in driving investment and accounted for its disruptive volatility. Such factors are clearly
exogenous to the relationship between interest rate and investment, and hence act to
I=I(i)
shift to position of the investment demand schedule as depicted in Figure 5.2. When
entrepreneurs are uncertain about future profits, they will discount their future earnings
at a high rate, thus reducing the expected return on all investments. This means that less
will be invested at all rates of interest, and the investment schedule will move closer to
the origin (as shown in Figure 5.2). If investors suddenly fear that disaster will strike and
hence cut back on their investment demand, this will have a similar impact on the position of the investment schedule. Of course, if expectations about the future economic
environment improve, then the rate at which entrepreneurs discount future earnings
will fall. This means that more investment will be undertaken at all rates of interest.
The investment schedule would thus shift out. The same occurs if investors confidence
increases for less rational reasons.
The accelerator theory of investment (see box below) tries to capture some of these
dynamics by explicitly modelling investment as a function of expected output. Such
dynamics can only be represented in the ISLM model as an exogenous shift of the
investment schedule in response to changed expectations about output.
69
70
Macroeconomics Simplified
(Box Continued)
In = v* (Qet Qet 1)
Qe is the estimate of expected sales and v* is a multiple of the expected sales
by assuming that the stock of capital, that is, the plant and the equipment that
a firm desires (K *) is a multiple of its expected sales:
K * = v* Qe.
Thus, the accelerator hypothesis says that the level of net investment (In)
depends on the change in expected output.
The flexible accelerator theory is a variant of the accelerator theory. It states
that the desired ratio of capital to expected output may be affected by the user
cost of capital. As per this theory, net investment does not always close the gap
between desired capital and the capital stock in the preceding period, but by
only a fraction of it.
Source: Authors.
Pa (5.1)
Thus, if the nominal interest rate was 3 per cent and inflation was 2 per cent, then
the real interest rate would be approximately 1 per cent. Given this potential for inflation to reduce the real rate of return to savers, we would expect savers to insist on higher
nominal interest rates to compensate them for expected inflation over the life of a loan.
The nominal (i.e., market) rate of interest can thus be conceptualised as approximately
the sum of the desired real rate of interest and expected inflation:
i = id
Pe (5.2)
nominal
= desired real + expected
interest rate interest rate inflation
The premium for expected inflation (Pe) is commonly referred to as the Fisher effect
(named after the economist Irving Fisher who identified it). The nominal rate of interest (i.e., that rate of interest ruling in the market) is thus modelled as consisting of the
markets consensus about the rental value of money (i.e., the desired real rate) plus a
premium determined by expected changes in the price level. The higher the expected
rate of inflation, the higher the nominal rate of interest.
In terms of the bond market, this means that expected inflation will result in a fall in
bond prices, and hence in a rise in (nominal) yieldsso as to ensure that the (inflationadjusted) desired rate of return is obtained.
Of course, lenders can only protect themselves adequately from inflation if they correctly anticipate price changes. When actual inflation exceeds expected inflation, then
borrowers benefit at the cost of lenders (and vice versa). This occurred in the mid- and
late-1970s when unexpected oil price increases boosted inflation above expected levels
and real interest rates turned negative in most countries. By contrast, real rates of return
were unexpectedly high in the early 1980s and the late 1990s because of unanticipated
recessionary conditions.
In the neoclassical version of the ISLM model, prices of all goods are assumed to
adjust quickly and instantaneously to fluctuations in demand and supply to restore equilibrium in all markets. The assumption of perfect competition means that actual and
expected price increases blur into the same thing. You will see in Chapter 7 (which
introduces the neoclassical version of the ISLM model) that as the general price level
rises, bond prices fall and the nominal interest rate (i) rises.
71
72
Macroeconomics Simplified
% Yield
Years to maturity
short-term (i.e., 1-year) bonds, wait for interest rates to rise, and then buy long-term
(i.e., 2-year) bonds and lock into the higher interest rate. In other words, instead of holding a 2-year bond for 2 years and obtaining 6 per cent a year, investors would prefer to
purchase two 1-year bonds: the first at the prevailing interest rate of 6 per cent; and the
second at the beginning of the following year when interest rates are expected to rise to
12 per cent. The average yield over the 2-year period would thus be ((1.06)(1.12))1/2 = 9
per cent (see the term structure formula outlined in the text box).
Investors will thus start selling any 2-year bonds they own, and buying 1-year bonds.
This will drive the price of 1-year bonds up (and hence the yield down), and the price of
2-year bonds down (and hence the yield up). The net effect of this portfolio adjustment
is to shift the prevailing yield curve from flat to ascending.
The process of buying and selling bonds continues until any differential in expected
returns over the 2-year investment period is eliminated. That condition could occur, for
example, when the 1-year forward rate is 12 per cent, and the yield on 1-year bonds equals
4 per cent and 2-year bonds yield 8 per cent. With this term structure, investors will be
indifferent between holding one 2-year bond yielding 8 per cent, or holding two 1-year
bonds yielding 4 per cent and then investing the proceeds the following year in bonds
with an expected 12 per cent yield. Such portfolio adjustment will result in the yield curve
shifting from a flat curve to one that is upward sloping (as shown in Figure 5.4).
This expectations-driven relationship between short- and long-term interest rates
determines the shape of the yield curve. If, for example, interest rates are expected to
increase, then the yield curve will be upward sloping; if they are expected to decrease, it
will be downward sloping; and if no change in interest rates is expected, the yield curve
will be flat.
73
74
Macroeconomics Simplified
Yield
8%
4%
1
Years to maturity
The expectations theory of the yield curve assumes that investors are indifferent
between long- and short-term securities. However, given that short-term securities tend
to be more liquid (i.e., the secondary market for them is more active) than longer-term
securities (which also display greater fluctuations in price), this assumption is usually
not true. Because of the problems associated with the liquidity of long bonds, borrowers seeking to fund capital projects with long-term bonds usually have to pay lenders a
liquidity premium in order to tempt them to purchase long-term bonds rather than the
more liquid short-term bonds. This liquidity premium implies that an upward-sloping
yield curve will be steeper than that predicted by the expectations theory.
If one assumes that the expectations theory has some validity, the yield curve provides
information about market expectations and future interest rates. It is thus observed with
great interest by many economic analysts.
If the yield curve is upward sloping, then one could (in terms of this theory) surmise
that the consensus in the market is that interest rates will rise in the future. Since interest rates tend to be procyclical (i.e., rise and fall with the business cycle), one might
conclude further that market participants expect an economic upswing in the future. By
the same logic, if the yield curve is downward sloping, then one could conclude that the
market expects interest rates to fall (which may imply that an economic downswing is
expected). The fact that descending yield curves are common near the final phase of a
period of economic expansion lends some credence to this theory.
Information can also be obtained by monitoring the spread between long-term and
short-term interest rates (such as the difference between the yield on a 9-year bond
and a 3-month bond). As the spread narrows (i.e., the yields move closer together), one
could surmise that the market consensus is that the rate of economic expansion will
slow down, and/or that future inflation is not expected to be so high. Alternatively, if the
spread widens (i.e., the yield curve becomes steeper), then this may indicate that inflation is expected to be higher in the future, and thus higher yields on longer-term bonds
are demanded.
When market analysts talk about the yield curve, they are often, in fact, referring
to the spread between short-term and long-term interest rates. For example, in some
countries, a rough yield curve can be calculated as a spread between the average long
and short bond rates. When plotted over time, this gives a yield curve of sorts. Normally, the yield curve rises ahead of booms and falls ahead of recessions.
Figure 5.5 plots the Indian yield curve for three periods: March 200102, March
200607 and March 201112. It shows that yields have risen over time (no doubt to keep
(5.3)
where R = actual interest rate; f = forward interest rate; t = time period for which
the rate is applicable; n = maturity of the bond. The postscript identifies the
maturity (n) of the bond, and the prescript represents the time period in which
the bond originates (t). Thus, tR1 is the actual market rate of interest on a 1-year
bond today (i.e., time t); similarly, tR10 is the current market rate of interest for
a 10-year bond. For the forward rates, the prescript identifies future rates. Thus
f refers to a 1-year interest rate, 2 years from now.
t+2 1
Suppose that the current 1-year rate is 6 per cent and that the market expects
the 1-year rate 1 year from now to be 8 per cent, and 2 years from now to be 10
per cent. Using the above notational form, this can be represented as follows: tR1
= 6%; t + 1 f1 = 8%; t + 2 f1 = 10%. Given such market expectations, we can calculate
the current 3-year rate of interest by applying the formula.
(1 + tR3) = [(1.06)(1.08)(1.10)]1/3
R = (1.259)1/3 1 = 8%.
t 3
This tells us that an investor with a 3-year time horizon will be indifferent
between buying a 3-year security yielding 8 per cent, or buying three successive
1-year bonds that will also yield, on average, 8 per cent.
Source: Authors.
75
Macroeconomics Simplified
10%
9%
8%
Yield (per annum)
76
7%
6%
5%
4%
3%
2%
1%
0%
1
10 11
abreast of inflation) but that the slope of the yield curve has changed too. In 200102,
it was clearly upward sloping, implying that bond-buyers expected inflation to rise over
time, and hence would only accept higher yields on long dated bonds. In March 200607,
however, the yield curve was flat, suggesting that bond-buyers did not expect inflation to
rise. As of March 201112, the yield curve remained flat, although the negative slope over
the first few years suggests that inflationary expectations were negative over the fairly
short term, but were flat over the longer term.
India has begun experimenting with IIBs.4 In 1997, the Reserve Bank of India
(RBI) issued a 6 per cent Capital-indexed Bond 2002, but the market response was
poor because only the principal repayment at the time of redemption was indexed to
inflation. A new version was designed in 2004 which linked both the principal repayments and coupon payments to the wholesale price index (WPI). However this also
was unsuccessful, perhaps because this benchmark was unconvincing and because of
the lack of depth in the bond market.5 When the Finance Minister P. Chidambaram
announced in his 2013 budget speech that he would, in consultation with the RBI, be
introducing IIBs or securities, the news was unsurprisingly greeted with both excitement
and scepticism. By early 2013, it was clear that there was a need for inflation-indexed
instruments which would grow the principal in line with inflation (and calculate coupon payments on this rising principal) to counterbalance the strong demand for gold
import. Negative real interest rates were discouraging people from buying bonds and
so gold imports were ballooning and undermining the balance of payments position.6
The challenge was to come up with an instrument that would appeal more to savers
than gold.
4
See A technical paper on Inflation Indexed Bonds. http://rbi.org.in/scripts/PublicationReportDetails.
aspx?ID=598. Accessed on 9 December 2010.
5
Ranu, A. 2013. Inflation Indexed Bonds, please. The Hindu Business Line. http://www.thehindubusinessline.com/features/investment-world/inflation-indexed-bonds-please/article2744860.ece
6
Adhikari, A. 2013. Will Inflation-indexed Bonds Check Rising Gold Demand? Business Today. http://
businesstoday.intoday.in/story/union-budget-2013-14-inflation-indexed-bond-gold-import/1/192935.
html. Accessed on 28 February 2013.
77
78
Macroeconomics Simplified
Years to maturity
According to the expectations theory of the yield curve, this suggests that market participants expect yields to stay the same for 5 years and then rise. A liquidity premium
could be flattening the curve which might otherwise be slightly downward sloping for
bonds with maturities of less than 5 years.
CHAPTER 6
We now turn to a standard rite of passage for economics students: the ISLM model.
This model has two sectors, namely the goods market and the money market. It shows
how dynamics in the financial sector affect the level of real economic activity.
As noted in Chapter 4, the ISLM framework was actually developed by Hicks in
19371 as an interpretation of Keynes in an article published several months after the
General Theory. Although many economists who considered themselves Keynesians were
uncomfortable with it, Keynes himself wrote and told Hicks that he had next to nothing
to say by the way of criticism.2 Modern-day new Keynesians and post Keynesians consider
Keyness judgement to have been rather rash in this instance. They argue that the ISLM
model served to obscure and trivialise many of Keyness revolutionary ideas and that
Keynes was too quick to welcome the ISLM model because it had Keynesian policy
implications.
Skidelsky on the ISLM Framework
The ISLM diagram, first drawn by John Hicks in 1936 is the General Theory as it
has been taught to economics students ever since: 384 pages of argument whittled down to four equations and two curves. Hicks, Harrod, Meade and Hansen in
America, the leading constructors of ISLM Keynesianism, had a clear motive:
to reconcile Keynesians and non-Keynesians so that the ground for policy could
be quickly cleared. These early theoretical models incorporated features which
were not at all evident in the magnum opus, but which conformed more closely to
orthodox theory. The constructors of the model also thought they were improving
the original building. Joan Robinson, no slouch with insults, would later label
the result bastard Keynesianism but Keynes was the bastards father.3
Source: Authors.
See, Hicks, J. 1937. Mr Keynes and the Classics. Econometrica (5) 2: 14759.
Hicks, J. 1981. ISLM: An Explanation. Journal of Post-Keynesian Economics 3 (2): 13954.
2
Quoted in Hillier, B. 1991. The Macroeconomic Debate. Models of the Closed and Open Economy. p. 42.
Oxford: Blackwell.
3
Skidelsky, R. 1992. John Maynard Keynes: The Economist as Saviour, 19201937. p. 538. London:
Macmillan.
1
80
Macroeconomics Simplified
Nevertheless, the ISLM model is widely interpreted as being Keynesian in character, particularly as regards the way in which consumption, savings, investment and the
demand for money are modelled.
I = j di, (6.1)
4
James Tobin, interviewed in Snowdon, B., H. Vane and P. Wynarczyk. 1994. A Modern Guide to Macroeconomics: An Introduction to Competing Schools of Thought. p. 129. Aldershot: Edward Elgar.
5
Stanley Fischer quoted in ibid., p. 35.
I = j di
0
I
The goods market equations start off with two definitional equations and an equilibrium condition:
E = C + I, (6.2)
y = C + S, (6.3)
y = E, (6.4)
6
In this regard, the investment schedule is like the demand curve that depicts the dependent variable
(quantity) on the horizontal axis and the independent variable (price) on the vertical axis. This is one of the
many oddities in the discipline of economics.
81
82
Macroeconomics Simplified
For example, if savers decide to keep some of their surplus cash rather than lend it
to firms to finance their planned investment, this implies that aggregate demand will be
insufficient to absorb available output. This can be represented as follows:
S > I,
C + S > I + C,
y > E.
This is clearly a disequilibrium position. If aggregate demand (i.e., spending) is too
low to absorb available output, then stocks will start to accumulate on shop shelves.
In other words, firms will experience unplanned additions to their inventories. This
indicates that producers expectations towards demand were too high. As was the
case in the simple model of income determination, this sends a signal to producers
to reduce their expectations and cut back on output and employment. Output (and
hence real income) will thus fall, bringing down C and S in the process, until it is
brought into line with aggregate demand. At that point, savings will have fallen to a
point where S = I.
If, however, S < I, then aggregate demand will be greater than the supply of available
output. Stocks will start falling at an unexpectedly rapid rate. This indicates to producers that their expectations about demand were too low. They adjust their expectations
accordingly and increase employment and output. As output (and income) rises, so too
will consumption and savings until S = I. At that point, the economy will come to rest,
that is, be in equilibrium.
Note once again the important role of expectations in the analysis. The economy
moves to a position of equilibrium through a process of output adjustment which is
driven by firms adjusting their expectations, and subsequently also their output decisions, in the light of unexpected changes in inventories.
So far, we have ascertained that for the goods market to be in equilibrium, the case must
be that planned investment equals savings. This is stated in Equation 6.5.
S = I. (6.5)
Investment is no longer seen as exogenous (as in the simple Keynesian model of
income determination) but is rather modelled as a negative linear function of the
interest rate:
I = j di. (6.1)
As regards the savings function, this can be derived from the familiar Keynesian consumption function through a process of substitution and re-organisation of terms.
83
84
Macroeconomics Simplified
C = a + by,
y = C + S,
y = a + by + S,
S = a + (1 b)y. (6.6)
Equation 6.6, the savings function, shows that savings is a positive function of income.
The slope of the savings function, that is, the marginal propensity to save, is equal to 1 minus
the marginal propensity to consume. For example, if the marginal propensity to consume
was 80 per cent, then the marginal propensity to save would be 20 per cent. The intercept
of savings function is simply the negative of autonomous consumption. In other words, if
an economy consumes amount a while experiencing zero income, then it must be the case
that the economy has run down on savings to do this. The intercept of the savings function
is thus a. The relationship between savings and consumption function is depicted in Figure 6.2. If the marginal propensity to consume (b) rises, then the slope of the consumption
function will become steeper, whereas the slope of the savings function will become flatter.
C
S = a + (1b)y
C = a + by
a
0
Self-test
What happens to the slope of the consumption function and the slope of the
savings function when the marginal propensity to consume increases?
Assume autonomous consumption spending increases. Draw the new consumption and savings functions.
Source: Authors.
The Keynesian idea that savings and consumption are positive functions of income
(rather than functions of the interest rate) is a radical departure from neoclassical economics. It turns the classical argument for thrift (i.e., economies need to save before
they can grow) on its head. In Keyness vision, it is consumption, not savings, which
drives output. Consumption boosts output, which in turn boosts savings. In terms of
this argument, cutting consumption in order to finance growth is wrong-headed, as
consumption is the mainspring of growth. As Keynes puts it:
It has been usual to think of the accumulated wealth of the world as having been painfully
built-up out of the voluntary abstinence of individuals from the immediate enjoyment of
consumption which we call thrift. But it should be obvious that mere abstinence is not
enough by itself to build cities or drain fens. ... It is enterprise which builds and improves
the worlds possessions. ... If enterprise is afoot, wealth accumulates whatever may be happening to thrift; and if enterprise is asleep, wealth decays whatever thrift may be doing.7
Substituting Equations 6.1 and 6.6 into Equation 6.5, we get the following equation:
S = I
a + (1 b)y = j di
Solving for y gives us
y = (a + j di)/1 b. (6.7)
This can be rewritten as follows:
y = (a + j di) (1/1 b).
Note the presence of the familiar Keynesian multiplier (1/1 b). Thus, a change in
a + j di will have a multiplied impact on y.
Solving for i gives us
i = (j + a (1 b)y)/d. (6.8)
Equations 6.7 and 6.8 are equations for the IS curve, that is, investment equals the
savings curve. The IS curve provides us with various combinations of y and i which will
make investment and savings equal and hence will ensure that the goods market is in
equilibrium. Given that investment is a negative function of the interest rate, whereas
savings is a positive function of output, we would expect y and i to have a negative relationship in equilibrium. If interest rates were low, and hence investment was high, the IS
curve tells us that y would have to be high so as to generate sufficient savings to match
the high levels of investment. If, on the other hand, interest rates were high (and hence
investment was low), y would have to be low so as to ensure equilibrium between low
7
85
86
Macroeconomics Simplified
IS
y
savings and low investment. This can be seen more easily in the diagrammatic derivation of the IS curve.
Equation 6.8 can be rewritten as:
i = (a + j)/d ((1 b)y)/d.
The IS curve has an intercept of (a + j)/d and a slope of (1 b)/d as shown in
Figure 6.3. The term for the slope tells us that the IS curve will be more elastic; the
smaller the marginal propensity to save (1 b), the larger the interest sensitivity of
investment (d). In other words, the flatter the slopes of the investment schedule8 and the
savings function, the more elastic will be the IS curve.
S
S=I
S2
S1
S1
0
45
I1
I2
0
a
y2
y1
i
i1
i1
I = j di
i2
0
S= a+(1b)y
S2
I1
I2
i2
0
IS Curve
y1
y2
In order to derive the IS curve, we need to ascertain two points along it.9 Given that
the IS curve represents combinations of i and y for which the goods market is in equilibrium, we need to generate two combinations of i and y for which I = S (remember that
the equilibrium condition for the goods market is such that I = S).
Start off by picking a low level of savings, S1. To see what level of output would generate S1, trace the corresponding value of y from the savings function. This tells us that
for S1 to be possible, it must be the case that y is y1. Given that we are searching for the
combinations of i and y for which the goods market is in equilibrium, it is now necessary to ascertain the corresponding equilibrium investment level for S1. The equilibrium
condition in the north-west quadrant shows that this will be I1. To find out what rate of
interest is associated with I1, turn to the investment schedule which indicates that the
interest rate would have to be i1 in order to produce an investment level of I1. This provides the first point on the IS curve, namely y1,i1.
To get a second point on the IS curve, pick a higher level of savings, S2. The savings
function indicates that the associated level of output is y2. To ascertain the corresponding
Note that we only need to find two points because we have assumed that the investment function is,
like the savings function, linear in character.
9
87
88
Macroeconomics Simplified
equilibrium investment level for S2, read this off from the equilibrium condition which
gives I2. The corresponding rate of interest (read off from the investment schedule) is i2.
This provides the second point on the IS curve, namely y2,i2.
What we have done here diagrammatically is the equivalent of substituting the investment function and the savings function into the equilibrium condition in order to derive
the IS curve in the south-east quadrant.
The IS curve shifts (i.e., a new IS curve will be derived) when constants and exogenous variables change in the model. In the simple IS curve (i.e., excluding government),
these are a (the negative of autonomous consumption); autonomous investment (j) the
marginal propensity to consume (b); the sensitivity of investment to the interest rate
(d) and finally, exogenous factors such as business confidence and expectations (which
affect the position of the investment schedule).
For example, if producers suddenly became more confident about earning future profits,
this would result in an outward shift of the investment schedule.
As shown in Figure 6.5, this results in a similar outward shift of IS from IS1 to IS2. The
exogenous increase in investment implies that for the goods market to be in equilibrium,
more output is required (at all rates of interest) to bring S in line with I. The outward
S
S
S3
S=I
S = a+(1b)y
S2
45
I1
I2
I3
S1
y1
y2
i
i1
i1
IS1
i2
i2
I1
I2
I
y3
IS2
y
shift of the investment schedule indicates that more investment will be forthcoming at
all rates of interest. Interest rate i1, which was associated with a level of investment I1, is
now associated with I2. For there to be sufficient savings to match that level of investment, output must be y2 in order to generate S2 of savings. The first point on the new IS
curve is thus y2,i1. Interest rate i2 is now associated with a higher level of investment, I3.
For there to be goods market equilibrium, output must be y3 and savings S3. The second
point on the IS2 curve is thus y3,i2.
With the aid of Figure 6.6, consider what happens to the IS curve if the marginal
propensity to save increases. The effect of an increase in the marginal propensity to
save will be manifested in an increase in the slope of the savings function. This indicates that more is saved at every level of output. Note that as autonomous consumption has not changed, the intercept (a) stays the same. The function simply swings
up to the left.
S
S2
S2
S=I
S1
S1
45
I1
y3
I2
y1
y4
y2
a
i
i1
IS2
i2
I1
I2
IS1
y3
y1
y4
y2
89
90
Macroeconomics Simplified
Thus, for S1 of savings to be generated in order to equate I1, only y3 of output is now
required rather than y1. The first point on the new IS curve is y3,i1. Likewise, only y4 of
output is now required to ensure sufficient savings to equate with I2. The second point
on the new IS curve is thus y4,i2. As can be seen from the diagram, the IS curve has
shifted to the left. It has also, like the savings function, become less elastic (i.e., steeper).
In order to use the ISLM model as a means of analysing fiscal and monetary policies, it
is necessary to derive the IS curve for a model that includes a government sector.
Aggregate demand becomes
E = C + I + G, (6.9)
where G = government spending (assumed to be exogenous).
A model that brings in government spending, usually also brings in taxation. We
bring in taxation here by modelling consumption and saving as functions of disposable
income, that is, income after tax has been deducted.
yd = y T, (6.10)
y = yd + T, (6.11)
T = ty, (6.12)
where yd = disposable income; T = tax revenues; t = the (proportional) tax rate, 1 > t > 0.
Disposable income is now divided up between C and S:
yd = y T = C + S. (6.13)
Therefore,
y = C + S + T. (6.14)
Goods market equilibrium still requires that aggregate demand equals output:
y = E.
(6.4)
Substituting Equations 6.9 and 6.14 into Equation 6.4, and removing C from both
sides of the equation gives us
S + T = I + G. (6.15)
Note that because S and C are now functions of disposable income, the savings function is slightly different from the earlier version (Equation 6.6).
S = a + (1 b)yd. (6.16)
Substituting Equations 6.10 and 6.12 into Equation 6.16, we get
S = a + (1 b)(y ty),
S = a + (1 b)(1 t)y. (6.17)
The savings plus tax function is thus
S + T = a + (1 b)(1 t)y + ty,
S + T = a + y ty by + bty + ty,
S + T = a + (1 b + bt)y. (6.18)
Substituting Equations 6.18 (the savings plus tax function) and 6.1 (the investment
function) into Equation 6.15 (the equilibrium condition), we arrive at the following
equations:
a + (1 b + bt)y = j di + G,
y(1 b + bt) = j di + G + a.
Solving for i
di = j + G + a y(1 b + bt),
i = (j + G + a)/d ((1 b + bt)y)/d. (6.19)
This tells us that the IS curve will be steeper: the smaller the interest sensitivity of
investment to the interest rate (d), the higher the tax rate (t), the smaller the marginal
propensity to consume (b), and the higher the marginal propensity to save (1 b).
Solving for y
y = (j di + G + a)/(1 b + bt). (6.20)
The new multiplier is
1/(1 b + bt).
Equations 6.19 and 6.20 are the new equations for the IS curve when the government sector is included. Note that the multiplier with taxes is smaller than the multiplier
derived from the model excluding a government sector.
91
92
Macroeconomics Simplified
For example, if b = 0.2 and t = 0.3, then the multiplier in the model excluding government is
that parents react to bond-financed expansion by saving now in order to help their (born
and unborn) children pay future tax increases. While one should not underestimate
parental love, this assumption probably pushes parental responsibility a little too far.11
S+T
S+T= a+(1b+bt)y
S+T=I+G
(S+T)2
(S+T)1
(I+G )
i
y2
y1
i
i1
i1
IS
I(i)+G
i2
i2
(I+G)1
(I+G)2
(I+G)
y1
y2
93
94
Macroeconomics Simplified
quadrant, that is, the equilibrium condition, now has S + T and I + G as the ordinate and
abscissa, respectively. This reflects the new equilibrium condition of S + T = I + G for the
goods market in a model including the government sector.
The north-east quadrant consists of new savings plus tax function. Note that the
intercept remains at a (i.e., the negative of the intercept of the consumption function).
The slope of the S + T line is determined by the marginal propensity to save (1 b) and
the tax rate (t) multiplied by the marginal propensity to consume (b). The slope of
the S + T function will be steeper than the original S function by the tax rate times the
marginal propensity to consume. A rise in the tax rate will thus cause the S + T function to swing up to the left. Figure 6.8 shows how the S + T function swings up from the
intercept as a result of the tax rate rising from t1 to t2.
The south-west quadrant consists of the I + G schedule (reproduced in Figure 6.9).
Given that G is exogenous, that is, determined outside the model, the downward slope
of the I + G function is determined entirely by the negative relationship between investment and the interest rate. The amount of government spending is simply added to
investment.
As shown in Figure 6.9, when government spending rises, this will be manifested in
an outward (parallel) shift of the I + G function (the other factor which would cause an
outward shift in the I + G function is of course a positive change in expectations and
business confidence).
S+T
S+T= a+(1b+bt2)y
S+T= a+(1b+bt1)y
y
a
I(i) + G1
I(i) + G2
I+G
Shifts in the IS curve will be caused by changes in constants and exogenous variables.
These are as follows: a (the counterpart in the savings function of autonomous consumption); autonomous investment (j); the marginal propensity to consume (b); government
spending (G); the tax rate (t); the interest sensitivity of investment (d) and exogenous
factors (such as changes in investor confidence).
Self-test
Make sure you can shift the IS curve in response to any positive or negative
change in the above-mentioned constants and exogenous variables.
Source: Authors.
By definition, all points along the IS curve are combinations of i and y for which the
goods market is in equilibrium. In order to grasp this point completely, it is useful to
consider points on and off the IS curve.
Consider point X in the south-east quadrant of Figure 6.10. Point X consists of coordinates (y2, i1). This is not a point of equilibrium for the goods market because the relatively high output level of y2 will produce a relatively high level of savings plus tax, i.e.,
(S + T)2, whereas the relatively high interest rate of i1 will induce only (I + G)1 of investment plus government spending. In other words, the components of aggregate demand
95
96
Macroeconomics Simplified
S+T
S+T
S+T= a+(1b+bt)y
S+T=I+G
(S+T)2
(S+T)2
(S+T)1
(S+T)1
45
(I+G)1 (I+G)2
I+G
a
i
y1
y2
i1
i1
i2
i2
I(i)+G
(I+G)1
(I+G)2
I+ G
y1
IS Curve
y2 y
(i.e., government spending plus planned investment spending) will be less than available
savings plus taxation. Put differently, injections of demand will be less than withdrawals.
This implies that aggregate demand will be insufficient to absorb the amount of output produced (y2). The goods market is thus characterised by a state of excess supply.
Unsold stocks will start accumulating. This, in turn, will send a signal to producers to
cut back on output. Assuming that the interest rate remains unchanged at i1, output will
have to fall to y1 before equilibrium is established at (y1,i1), which is a point on the IS
curve. All points to the right of the IS curve are points of excess supply.
Now consider point Z, which consists of co-ordinates (y1,i2). This is not a point of
equilibrium for the goods market because a relatively low output level of y1 will produce
a low savings plus tax level of (S + T)1, whereas a relatively low interest rate like i2 will
induce a relatively high level (I + G)2 of planned investment and government spending.
In other words, injections into the system will be greater than withdrawals.
This indicates that producers underestimated the demand for their products and produced too little. The goods market is characterised by a state of excess demand. As stocks
start to run down in response to the stronger than expected aggregate demand, producers
react by generating more output. Assuming that interest rates remain unchanged, output
will rise to y2, at which point equilibrium will be established at (y2,i2). All points to the left
of the IS curve are positions of excess demand.
The IS curve slopes downwards because as interest rates rise (e.g., from i2 to i1 in Figure 6.10), investment will fall. The fall in investment reduces demand. As demand falls,
stocks start to rise, and this sends a signal to producers to cut output. Output will fall
until goods market equilibrium is restored at y1, i1.
97
98
Macroeconomics Simplified
Money demanded for transaction purposes is held in the form of active balances.
Money kept for speculative or precautionary motives is said to sit in idle balances.
The focus on liquidity is central to Keyness thinking. Because much of the motivation for holding idle balances has to do with uncertainty and imperfect information,
Keyness liquidity preference theory of money brings these aspects into the heart of his
theory of employment and income determination. The very fact that people hold idle
balances implies that resources are not being fully utilised (and implies that Says law
does not hold). This means that the economy is operating at a less than full-employment
level of output.
A pure neoclassical model would, of course, reject such a theory. Given the assumptions about perfect information, there would be no need for any precautionary or speculative cash balances, and no reason for output to stagnate at a less than full-employment
level. The fact that the ISLM model is strongly influenced by Keyness liquidity preference theory, reflects its essentially Keynesian character.
Keyness vision, which one can trace back to his youth, has to do with the logic
of choice, not under scarcity, but under uncertainty; with its daring corollary
that the desire for goods is more easily satisfied than the desire for money, or
liquidity.12
L1
depicted as having the dependent variable (L2) on the horizontal axis and the independent variable (i) on the vertical axis.
We model the demand for idle balances as a negative linear function of the interest
rate:
L2 = f hi, (6.22)
where L2 = idle balances; f = precautionary holdings of cash (unrelated to the interest
rate); h = interest elasticity.
Idle balances comprise precautionary holdings of cash, which are not related to
changes in the interest rate, and speculative holdings that are sensitive to the interest
rate. The speculative demand for money is a negative function of the interest rate (as
shown in Figure 6.12) because at high rates of interest, the opportunity cost of holding
i
L2
99
100
Macroeconomics Simplified
cash (i.e., the yield) is high. People would be better off holding bonds. At high interest
rates, bond prices are low, and there is thus a good chance that bond prices may rise (and
thus afford an opportunity for capital gain). Conversely, at low interest rates, the opportunity cost of keeping money in idle balances is low. When people expect interest rates
to rise (i.e., the price of bonds to fall), it makes better sense (from a speculative point of
view) to maintain a high level of liquidity and purchase bonds later when the prices fall.
One could hypothesise that at a certain low interest rate, people will move all their
assets into cash; that is, the demand for idle balances will be perfectly elastic at that
interest rate (as shown in Figure 6.13). People expect the price of bonds to fall (and the
interest rate to rise) and hence will not wish to enter the bond market at all. If such a
liquidity trap prevailed, then the L2 curve would become perfectly elastic at some low
interest rate. A liquidity trap occurs when the nominal interest obtainable on shortterm assets other than money is close to zero, and people are indifferent as to whether
they hold money or short-term assets. As we will see later, this renders monetary policy
ineffective because the reserve bank becomes powerless to reduce interest rates further.
Economists have argued that this was the case in Japan when short-term interest rates
averaged 0.5 per cent between 1996 and 1999.13 A similar situation arose in the USA and
Europe after 2008 after interest rates fell close to zero.
Total money demand is the sum of desired (or planned) active balances and desired
(or planned) idle balances.
L = L1 + L2,
L = ky + f hi. (6.23)
(In case, you were wondering why the letter L is being used to denote money demand,
it denotes liquidity preference.)
i
L2
The money supply (M), which is given exogenously in this model, is absorbed either in
the form of active or idle balances. The money supply has a planned (desired) and an
unplanned component. It consists of active balances (to satisfy the transactions demand
for money) and planned and unplanned idle balances.
M = L + unplanned (and hence undesired) money balances.
L1
L1 = M
45
L1a
90
L2b
45
L2 = M
L2
As can be seen in Figure 6.14, the money supply (M) can be absorbed entirely in idle
balancesin which case it will be represented by position L2 = M on the horizontal axis
or absorbed entirely in active balances, in which case it will be represented by position
L1 = M on the vertical axis. Alternatively, the money supply can be absorbed in various
combinations of active and idle balances, as represented by all points along the budget
line. At all points along the line, different combinations of L1 and L2 will equal M. Given
that the distance from the origin to M on each axis is the same (because the money supply
will be the same in each case), the budget line will form a 45 angle with each axis.
Assume that the economy has L1a of active balances and L2b of planned idle balances.
The money supply thus consists of L1a + L2b + amount of (unplanned and undesired)
idle balances.
The money market is in equilibrium only when money demand (i.e., desired active
and idle balances) equals the money supply (i.e., actual active and idle balances).
Just as the IS curve provides combinations of y and i for which the goods market is in
equilibrium, the LM curve plots combinations of y and i for which the money market is
in equilibrium. The money market is in equilibrium when the supply and demand for
money are equal, that is, L = M.
L = M. (6.24)
101
102
Macroeconomics Simplified
i = (y/v + f M)/h,
i = (f M)/h + (y/v)/h. (6.26)
Equations 6.25 and 6.26 are the equations for the LM curve. They tell us that there
are various combinations of i and y for which the money market is in equilibrium. The
slope of the LM curve in Equation 6.26 depends on the velocity of circulation (vk) and
the interest elasticity (h) of the demand for money. In case of a liquidity trap, the LM
curve will be perfectly elastic at the same interest rate at which the demand for idle
balances becomes perfectly elastic.
L1
L1
L1 = M
L12
L12
L11
L11
90
y1
y2
L22
y
LM
i2
L21
L2 = M
L2
i
i2
i1
i1
y1
y2
L22
L21
L2
In order to illustrate the economic forces at work moving the money market towards
equilibrium, consider point X (i.e., y1,i2) in Figure 6.16. At point X, L11 of active balances are demanded in order to finance the transactions demand associated with y1.
For the money market to be in equilibrium, L21 worth of idle balances would have to be
demanded. However, as can be seen in the diagram, the interest rate i2 associated with
point X is too high to generate demand for idle balances equal to L21. Instead, only L22
worth of idle balances are demanded. This is because at high interest rate of i2, people
would prefer to hold more bonds than idle balances.
Under these conditions, people find themselves with excess idle balances equal to
L21 L22 (in case you ask yourself why these are excess money balances, remember that
103
104
Macroeconomics Simplified
the money supply is fixed and must be absorbed either in the form of active or idle balances. Given that active balances are determined by income, any discrepancy between
money demand and money supply will be accounted for in [unplanned] idle balances).
Given that people only want to hold L22 worth of idle balances when the interest
rate is i2, they will respond by altering their portfolios by purchasing more bonds. This
will drive the price of bonds up and the yield (i.e., the interest rate) down. Point X is
thus an unstable position. Bond prices will rise and interest rates will fall until they
reach the level i1, at which point the money market will be in equilibrium at point y1,i1
(which is on the LM curve). All points to the left of LM indicate an excess supply of
money (ESM).
Now consider point Z in Figure 6.16. Point Z is associated with level of output y2.
This means that L12 worth of active balances will be demanded in order to finance
transactions associated with output level y2. For the money market to be in equilibrium, the north-east quadrant shows that L22 worth of idle balances would have to be
L1
L1
L =M
1
L12
L11
y1
i
i2
y2
L22
i
i1
y1
y2
L21 L2 = M L2
L22
L21
L2
demanded. But for people to want to hold only L22 idle balances, the interest rate must
be i2 rather than i1.
In case of point Z, the interest rate is not high enough to reduce the desired holdings
of idle balances to the level required for equilibrium to exist in the money market. Given
the low ruling interest rate of i1, a higher level of idle balances (L21) than is consistent
with money market equilibrium will be demanded. Under these conditions, people find
themselves with excess holdings of bonds (or, put differently, insufficient idle balances).
They thus start selling bonds. This drives the price of bonds down and the interest rate
(or yield) up. The money market will eventually come to rest at point y2,i2 (which of
course is on the LM curve). All points to the right of the LM curve are positions indicating excess demand for money (EDM).
In sum, all combinations of interest rate and output which lie above the LM curve
are characterised by excess idle balances (or excess money supply). People will thus start
adjusting their portfolios by buying bonds, thus driving the price of bonds up and the
interest rate down until the LM curve is reached. All combinations of interest rate and
output that lie below the LM curve are characterised by EDM. People will start adjusting
their portfolios by selling bonds, thus driving the price of bonds down and the interest
rate up until the LM curve is reached.
So far, the bond market dynamics that drive the changes in the interest rate (i.e., the
bond yield) have been hidden from view in the LM geometric derivation. Figure 6.17
brings them in explicitly. The diagram shows that if the economy was at y1,i1, then people would have enough transaction balances to cover the transactions associated with
Excess demand
for bonds
D (y1)
i2
LM
i1
S (y1)
i3
y1
105
106
Macroeconomics Simplified
income level y1, and they would be comfortable with their holdings of idle balances,
given the interest rate of i1. If the economy was, however, at y1,i2, then there would be an
ESM (and hence an excess demand for bonds in the bond market). Bond prices would
rise as people demand more bonds (to reduce their stock of idle balances). Yields will fall
until equilibrium is restored at y1,i1.
If the economy was at y1,i3, then there would be an EDM, and hence also an excess
supply of bonds (as people adjust their portfolios so that they comprise a higher proportion of cash balances). Bond prices would fall (and yields would rise) until equilibrium
is restored at y1,i1.
The LM curve slopes upwards because, as real income rises, the demand for active balances increases, and (assuming an unchanged money supply) people will have to adjust
their portfolios and sell bonds, thus driving the interest rate upwards. Referring to Figure 6.16, the rise in income from y1 to y2 results in a movement along the L1 curve from
L11 to L12. As the interest rate rises, the demand for idle balances falls. This can be seen
as a movement back along the L2 curve from L21 to L22. Both these movements are represented as a movement along the LM curve from y1,i1 to y2,i2.
Now consider what happens to the LM curve if the money supply rises. This will happen if the reserve bank prints money and gives it to the treasury to spend on goods and
services, or if the reserve bank engages in open market operations and buys government
bonds from the public. If the government monetises the debt by issuing government
bonds and selling them to the reserve bank, then the money supply will rise as government expenditure takes place.14 Such financing strategy amounts to printing money to
finance government expenditure.
Taking the output level y1 in Figure 6.18, let us trace through the effect of increase in
money supply. Output level y1 still generates the same transactions demand and hence
the same demand (L11) for active balances. However, with an increase in money supply,
the north-east quadrant shows that excess idle balances now exist at the prevailing level
of output and interest rate. People adjust their portfolios by buying bonds, thus driving
the price of bonds up and the interest rate down. As the interest rate falls, the demand
for idle balances rises (i.e., there is a shift down along the L2 curve) until money market
equilibrium is restored at i3.
Figure 6.19 tells the same story, but this time includes the bond market in picture. As
people find themselves with excess holdings of money (as a result of increased money
14
Note that this expenditure also pushes out the IS curve (as demand and output rises in response). Here,
we are only considering the monetary impact of the expenditure when it is financed in this way.
L1
M'
L1
M
L12
L11
y1
i
i2
i4
y2
L22
y
LM1
L21
L24 L23
M'
L2
LM2
i1
i3
L2
supply), they demand more bonds, and hence the demand for bonds shifts out, thus
driving down the yield. The LM curve shifts out.
Now consider the impact of an increase in money supply caused by the reserve bank
engaging in open market operations and buying bonds from public. This time, the impact
of monetary policy is initially in the bond market, as the reserve bank increases the demand
for bonds through its open market operations. The demand for bonds function shifts out,
and yields fall. However, as people sell bonds to the reserve bank, the money supply rises
(as people end up with more cash, and the reserve bank ends up with more bonds).
Finally, let us see what happens to the LM curve if (for whatever exogenous reason) people decide to hold more idle balances, that is, to increase their preference
for liquidity at all interest rates. This could be caused by a general lack of confidence.
As people demand more idle balances at all interest rates (i.e., L2 shifts out), they
107
108
Macroeconomics Simplified
L 1 = M2
L1
L1
L 1 = M1
L2 = M1
L2 = M2
S (y1)
D1 (y1)
D2 (y1)
i
i1
LM1
L2
LM2
i2
y1
L2
Figure 6.19 An Increase in the Money Supply (Showing the Bond Market)
Source: Authors.
sell bonds, thus pushing out the supply curve of bonds (Figure 6.20). The LM curve
shifts backwards from LM1 to LM2 (if this decline in confidence is also reflected in an
exogenous decline in investment demand, then the IS curve will also shift back, thus
exacerbating the recession).
L1
L =M
1
L1
L12
L11
L22
y
i
L21 L2 = M
L2
i1
LM2
LM1
y1
L22
L21
L2
Figure 6.21 summarises the goods and money market conditions in the ISLM
framework. Any point which falls to the right of the IS curve will be characterised by
excess supply in the goods (ESG) market. Stocks will accumulate and output will start to
fall until goods market equilibrium is restored. Any point to the left of the IS curve is
characterised by excess demand in the goods (EDG) market. Stocks will start falling,
Example
Let j = 600; d = 100; G = 450; a = 60; b = 0.5; t = 0.1; f = 10; h = 20; M = 408
and v = 5.
What are the equilibrium values for i and y?
Answer: i = 0.1 and y = 2,000.
Source: Authors.
109
110
Macroeconomics Simplified
ESG
ESM
LM
ie
ESG
EDM
EDG
ESM
EDG
EDM
IS
0
ye
this will send a signal to producers to increase output, and hence, output will rise until
goods market equilibrium is restored.
Any point to the right of (i.e., below) the LM curve represents EDM. As people sell off
bonds (so as to satisfy their demand for money), the price of bonds will fall and interest
rates will rise until money market equilibrium is restored. Any point to the left of (i.e.,
above) the LM curve represents a point of ESM. People will thus demand more bonds,
the price of bonds will rise, and the interest rate will fall, until equilibrium is restored in
the money market.
Consider point Z in Figure 6.22. Point Z (y1,i1) happens to be a point of equilibrium
in the money market (it is on the LM curve), but not a point of equilibrium in the
goods market (it is off the IS curve). Under these circumstances, the goods market is
characterised by excess demand. Stocks start falling, and hence output starts to rise.
Assuming no rise in interest rates, output would rise until goods market equilibrium
is restored at y2,i1.
However, in the ISLM model, which brings together the goods and money markets,
the rise in output will have an effect on the money market, which in turn will affect the
eventual level of output. As soon as output rises, this has an effect on the money market.
As output rises, so does the transactions demand for money. People respond by selling
bonds (in order to satisfy their increased demand for money). Bond prices start to fall,
and interest rates start to rise. As interest rates rise, this dampens down the demand for
investment, hence also the total demand in the economy.
i
LM
ie
i1
X
IS
y1
ye
y2
In the ISLM model, the dampening effect of interest rates on investment, and
hence also on output, is captured in the negative slope of the IS curve. In other words,
if output jumped from y1 to y2, then the interest rate rises, which occuring shortly
thereafter, would result in a movement back along the IS curve from y2 to ye. In practice, of course, the economy would probably not overshoot as far as y2 because interest rates would start rising soon after output started to rise. In any event, economic
dynamics in both the goods and money markets will eventually come to rest when ye,ie
is obtained.
Self-test
Explain the goods and money market dynamics which will move the economy
from point X to ye,ie.
Source: Authors.
111
112
Macroeconomics Simplified
increase in output. Interest rates do not enter the model, and investment is regarded
as exogenous.
In the ISLM model, however, the picture is rather different. If demand increases,
output will rise (as in the simple model of income determination). However, as output
rises, so too will the interest rate. Higher interest rates imply lower investment. Lower
investment will lead to lower demand and hence a smaller eventual increase in output.
In other words, by building a more sophisticated model in order to take into account
dynamics in both the goods and the money market, we observe that an increase in output following an increase in demand is moderated by a consequent rise in interest rate
and decrease in demand for investment.
Figure 6.23 compares the two Keynesian models. The top diagram depicts the
Keynesian cross framework, which captures only trends in the goods market. If aggregate
demand rises (e.g., as a result of an increase in government spending from G1 to G2),
then E rises from E1 to E2. Stocks fall and output rises from y1 to y2.
y=E
E2 = C1 + I1 + G2
E1 = C1 + I1 + G1
y2
y1
LM1
LM2
i2
i1
IS2
IS1
y1
y3
y2
Figure 6.23 Comparing the Keynesian Cross and the ISLM Models
Source: Authors.
The ISLM model is depicted below the Keynesian cross in Figure 6.23. If expenditure rises (e.g., as a result of an increase in government spending from G1 to G2), then
the IS curve shifts out from IS1 to IS2. Equilibrium output rises from y1 to y3. Note
that y3 is less than the simple models equilibrium output level of y2. This is because of
the negative relationship between interest rates and investment (as captured in the
slope of the IS curve) and the positive relationship between interest rates and output
(as captured in the LM curve). The net result of dynamics in the goods and money
markets shows that an increase in output (resulting from an increase in expenditure)
is less in the ISLM model than in the Keynesian cross framework. In other words, the
multiplier effect with respect to an increase in demand on output is less in the former
than in the latter.
For an increase in output in the ISLM model to be as high as that in the Keynesian
cross model, the dampening effect (on investment and hence also on output) of the
interest rate rise must be addressed. If the interest rate could be made to stay the same
(at i1), then the rise in output would be the same in the two models.
This could be brought about by an increase in money supply. If money supply grew in
line with income, then people would have sufficient cash balances. They would thus not
have to sell bonds in order to finance a higher level of transaction. The LM curve would
shift out to LM2 and the increase in output would then be the same in the Keynesian
cross model as in the ISLM model.
According to the ISLM model, this would happen if government financed the
increased expenditure by having the reserve bank print moneythus expanding the
money supply in line with the increased expenditure. If, however, the government
financed the increased spending by borrowing money from the public, then the money
supply (and hence the LM curve) would remain unchanged.
This analysis suggests that governments are best off printing money to finance
expenditure if their goal is to have the maximum impact in terms of stimulating output. The obvious problem with this policy prescription is the danger of inflation. The
Keynesian version of the ISLM model discussed so far assumes that prices are constant.
Hence, inflation never rears its ugly head.
However, in the neoclassical version of the model, price adjustment comes back
into play, and the danger of inflation is very real indeed. Chapter 7 discusses monetary
and fiscal policies in more depth using both Keynesian and neoclassical versions of
the ISLM framework.
113
114
Macroeconomics Simplified
S+T
(S+T)2
S+T
(S+T)1
S+T=I+G
(I+G)
i
a2
a1
i
i1
LM
IS2
i2
IS1
(I+G)1 (I+G)2
(I+G)
y2 y1
115
116
Macroeconomics Simplified
If the money supply is 800 and the demand for money is 500 400i + 0.5y, which if the
following represents a point on the LM curve?
a) y = 660, i = 10%
b) y = 640, i = 10%
c) y = 680, i = 10%
d) y = 650, i = 10%
e) None of the above
Answer: c
If the IS curve is y = 1000 5000i and the LM curve is y = 200 + 3000i, then the equilibrium combination of y and i is:
a) y = 1250, i = 5%
b) y = 1000, i = 10%
c) y = 1000, i = 20%
d) y = 500, i = 10%
e) y = 250, i = 15%
Answer: d
117
CHAPTER 7
This chapter explores the economic effects of fiscal and monetary policies within
an ISLM framework. The first section builds on the basic Keynesian interpretation
developed in Chapter 6. The second section shows how the ISLM model behaves
when neoclassical assumptions are adopted.
Monetary policy in the ISLM model refers to the alteration of the (nominal) money
supply by the reserve bank. The money supply is regarded as exogenous and hence as
directly controllable by the monetary authorities. Although the discussion of monetary
policy is limited in the ISLM framework, it nevertheless allows some basic exploration
of the economic impact of an expanding or contracting money supply.
Fiscal policy is the management of governments budget (which consists of inflows of
tax revenue and outflows of expenditure) in order to affect the level and composition of
aggregate demand in the economy.
When expenditure exceeds tax revenues, the government runs a primary deficit.1
Such a deficit can be financed by the sale of government bonds to the public, or by
monetising the deficit, that is, when the government issues bonds and the reserve bank
buys them. Selling government debt to public in order to finance expenditure has no
net monetary impact (as the money spent by the government has been obtained from
public via bond sales). Monetising the deficit, however, expands the money supply as it
basically amounts to the reserve bank printing money and giving it to the government to
spend. The money supply thus expands in line with government spending, resulting in
an equal outward shift in IS and LM (see panel A in Figure 7.1). The stimulus packages
after 2008 in the USA involved such an approach (sometimes called Obamanomics),
with the expansionary monetary stance referred somewhat euphemistically as quantitative easing.
When the government runs a primary budget surplus (tax revenues exceed expenditure), the monetary impact depends on what the government does with the surplus.
If it buys back government bonds from the public, then the net monetary impact is
1
The conventional deficit/surplus includes interest payments on the government debt as part of expenditure. The primary deficit/surplus excludes interest payments to the holders of government debt.
LM1
LM1
LM2
LM2
i2
i1
i1
IS2
IS2
IS1
y1
y2
IS1
y
y1
y2
zero (as the excess of taxation over spending is returned to the public through such
redeeming of the debt). If it buys back government bonds from the reserve bank, then
the money supply will contract (as the excess of taxation over spending ends up in the
reserve banks vaults).
In other words, the impact of fiscal policy on money supply depends on the way that
fiscal policy is financed. Pure fiscal policy is defined as a government budgetary change
that leaves the money supply unaltered. This occurs when government spending is taxfinanced (as in panel A of Figure 7.2) or financed by selling bonds to (i.e., borrowing
from) the public (as in panel B of Figure 7.2), or when the government uses a primary
surplus to redeem government debt to the public. Note that the multiplier effect is larger
in the bond-financed expansion than in the tax-financed expansion. This is because
the increase in taxation reduces disposable income (and makes IS steeper) and hence
reduces consumption spending. Because of the balanced budget multiplier, the overall
impact on demand of an equal increase in taxation and government spending is still
positivebut not as large as in the bond-financed expansion.
In practice, governments rarely follow pure fiscal policies. They may, for example,
finance the increase in government spending in part through selling bonds to the reserve
bank (i.e., printing money), and in part by raising taxation. This is illustrated in panel B
of Figure 7.1.
119
120
Macroeconomics Simplified
LM1
LM1
i2
i2
i1
i1
IS1
y1
y2
IS2
IS2
IS1
y
y1
y2
Fiscal policy
When the economy is stuck in a deep recession, Keynesians typically argue that it is the
duty of government to increase spending. The extreme version of this is captured most
graphically, perhaps the most notorious passage in Keyness General Theory:
If the [British] Treasury were to fill old bottles with bank notes, bury them in suitable
depths in disused coal mines which are then filled up to the surface with town rubbish,
and leave it to private enterprise on well tried principles of laissez-faire to dig up the notes
again ... there would be no more unemployment.2
2
Keynes, J. M. 1936. (Reprinted in 1964). The General Theory of Employment, Interest and Money. p. 129.
London: Macmillan.
Although Keynes was no doubt exaggerating for effect, this passage highlights an
important early Keynesian belief, namely that the very act of injecting demand is more
important than the nature of that spending. Even if government spending was unproductive in and of itself, it could still have a beneficial effect through the multiplier process. Note that in the case of burying bank notes, this would result in an outward shift in
IS (as a result of the government employing people to do the burying) and in an increase
in the money supply (once people had dug up the bottles and started spending the notes)
and hence in an outward shift in LM.
As discussed earlier, the monetary impact of fiscal policy depends on how government spending is financed. Assuming an expansionary pure fiscal policy, the impact
of government spending is merely to shift out the IS curve, leaving the money supply
unaltered (as in Figure7.2). Output rises (in response to increased demand) from y1 to y2
as the IS curve shifts out. Interest rates rise as people sell bonds to finance the increased
transactions demand associated with the expansion in output and income. This is represented as a movement along the LM curve from i1 to i2.
Fiscal policy will be most effective where the LM curve is relatively elastic, and the
IS curve relatively inelastic. Under these conditions, an increase in demand will result
in a relatively limited increase in interest rates; and, given the relative insensitivity of
investment to interest rates, this will have only a minor dampening impact on the level
of output.
This story gets told in reverse when it comes to a restrictive fiscal policy. Assume
that the economy is initially in equilibrium at the intersection of IS2 and LM1 (at y2,i2)
in Figure 7.2. If the government pursues a (pure) restrictive fiscal policy, then the I + G
curve will shift backwards, as will the IS curve (from IS2 to IS1). Output will fall to y1 as a
result of the decrease in demand, and interest rates will fall to i1 as people demand lower
transaction balances and therefore buy bonds, thus increasing the price of bonds and
lowering the yield.
Because output falls in response to the decrease in demand, the Keynesian model
implies that the economy can get trapped in a low-level equilibrium. There are no forces
other than an increase in demand to pull the economy out of this low-level equilibrium. The policy implication is that governments can cause recessions if they implement
restrictive fiscal policies.
Monetary policy
Monetary policy in an ISLM framework entails the reserve bank affecting the money
supply through open market operationsthat is, the sale and purchase of bonds to and
from the public. If the reserve bank buys bonds from the public, then the money supply
increases. If it sells bonds to the public, then the money supply decreases.
An increase in the money supply results in an outward shift in the LM curve. This was
discussed in detail in Chapter 6 with reference to Figures 6.18 and 6.19. As the money
supply increases, the demand for bonds increases, bond prices rise and the yield falls
(from i1 to i2 in Figure 7.3). This decline in the interest rate impacts the real sector (the
121
122
Macroeconomics Simplified
i
LM1
LM2
i1
i2
IS1
y1
y2
goods market) via a negative relationship between investment and interest rate. As the
interest rate falls, investment demand increases and hence demand increases. Output
then increases in line with the increased demand from y1 to y2. This is represented as a
movement along the IS curve in Figure 7.3.
The increase in output as a result of an increase in money supply is particularly a
Keynesian story. The Keynesian transmission mechanism can be explained as follows:
the rise in money supply leads to a fall in the interest rate, which leads to an increase in
investment and hence an increase in demand and eventually output. Neoclassical economists, by contrast, believe that a rise in money supply has no lasting impact on output.
Although the Keynesian transmission mechanism indicates that monetary policy can
have a positive impact on output, the outcome depends on the elasticities of key functions. If investment is relatively insensitive to changes in the interest rate, and if the
demand for idle balances is very elastic, monetary policy will be less effective.
The simple Keynesian ISLM model is a comparative static model where one equilibrium point is simply compared with another. The dynamic path that the economy traces
in getting from one equilibrium to another is not modelled explicitly. As real-world
economies are rarely in equilibrium, policy responses are never as clear as suggested by
the ISLM framework.
For example, assume the economy is on an adjustment path described in Figure 7.4.3
At stage 1, the economy is experiencing falling output and interest rates. This corresponds
3
LM
1
4
3
2
IS
y
y
Time
to the downward phase of the business cycle sketched below the ISLM diagram. At some
point, the impact of lower interest rates acts to stimulate demand, and hence output
rises, thus halting the downward movement in interest rates (stage 2). This corresponds with the early stage of the economic upswing. As output continues to increase,
interest rates rise, thus causing the economy to cool off at the top of the upswing (stage
3). The economy then dips back towards recession as output falls (stage 4). Such an
adjustment path is characterised by real-world tendencies for the goods market and the
money market to overshoot their optimal positions (as predicted by IS and LM) with
the result that the economy is characterised by a never-ending cycle of over-shooting
correctionover-shootingcorrection.
What is an appropriate policy response in such scenario? Should the government
pursue expansionary fiscal and monetary policies in stage 1, or should it wait for the
economy to correct itself? If the government decides to intervene (perhaps on the basis
that self-correction will take too long and will entail too much of a loss in terms of
123
124
Macroeconomics Simplified
output), then at what point does it halt such policies? Clearly, expansionary economic
policies are ill-advised at stage 3, but at what point, exactly, does stage 2 turn into stage
3? And what indicators can the governments use to make an informed guess about the
optimal point for policy changes?
During the 1950s and 1960s, when Keynesian demand-management was widely
accepted and practised, most governments were prepared to prime the pump of
demand. However, as advanced capitalist economies moved towards full employment in
the mid- to late-1960s, inflation began to characterise stages 2 and 3. Boosting demand
under such conditions was thus costly in terms of macroeconomic stability. This experience resulted in a conservative backlash against expansionary economic policies in most
of the advanced capitalist countries.
A similar policy trajectory has become evident in the period from 2008. The global
financial crisis of 200708 resulted in major economic stimulus packages in advanced
capitalist countries, especially the USA and the United Kingdom. When the US economy started slowing in 2008, President Bush approved a $268 billion stimulus. But as the
economy worsened, the new President Obama obtained a $787 billion from Congress
in January 2009. This was widely believed to have staved off a major depression. Even
so, the cost was high in terms of rising government debtespecially in the USA, where
federal debt rose from about 30 per cent of GDP in 2000 to over 100 per cent of GDP
today. Commentators like John Taylor, whose Taylor rule advises governments to adopt
neutral monetary and fiscal policies, now argues that excessively low interest rates in the
2000s were a major cause of the financial crisis and that expansionary fiscal policies subsequent to the crash were unhelpful and should be rolled back.4 More Keynesian commentators like Paul Krugman, however, argue that fiscal cut-backs are prematureand
will extend the recession (as they did in the 1930s in the USA, and in Japan in the late
1990s).5 He argues that austerity programmes for heavily indebted countries like Greece
are counterproductive as they will depress demand further, thereby raising rather than
reducing debt to GDP ratios.
In Krugmans view, the root of the economic policy errors leading up to the 2008 economic crisis was that economists, as a group, mistook beauty, clad in impressive looking mathematics, for truth and that as memories of the Great Depression faded, they
fell back in love with the idealized vision of the economy in which rational individuals
interact in perfect markets.6 He argues that the freshwater economists in Chicago who
argued against the US fiscal stimulus are neoclassical purists and that they had forgotten the lessons of Keynesif they ever knew them. Keynes biographer, Robert Skidelsky
See, Taylor, J. 2010. Getting Back on Track: Macroeconomic Policy Lessons of the Financial Crisis.
Federal Reserve Bank of St Louis Review, May 92 (3): 16576.
5
Krugman, P. 2010. British Fashion Victims. New York Times, 22 October.
6
Krugman, P. 2009. How Did Economists get it so Wrong? http://www.ie.ufrj.br/hpp/intranet/pdfs/krugman_september_6_2009_howdideconomistsdidsowrong.pdf. Accessed on 31 March.
4
agrees. He published a book in 2009 with the provocative title: The Return of the Master:
Why Sixty Years after His Death, John Maynard Keynes is the Most Important Economic
Thinker for America (New York: Basic Books).
The debate between Keynesian and Neoclassical economic models is thus alive and
well today.7 We now turn to an analysis of fiscal and monetary policy in the neoclassical
version of the ISLM model.
125
126
Macroeconomics Simplified
L1
M/P1
L1
M/P2
y
i
M/P2 M/P1
LM2
L2
LM1
L2
Having shown how price changes can enter the ISLM model, let us now consider
the economic effects of expansionary fiscal and monetary policies when a neoclassical
world is assumed.
Fiscal policy
Assume the economy is in equilibrium at yf , i1in Figure 7.6. Note that as this is a neoclassical world, the equilibrium level of output is assumed to be at full employment (hence
yf ). As explained in Chapter 3, the neoclassical model assumes that all markets function
perfectly and hence that supply and demand will be brought into line in all markets by
changes in price. This means that the economy will come to rest at the full-employment
level of output.
L1
L1
M/P1
M/P2
M/P2
y
LM2
M/P1
LM1
i2
i1
IS1
yf
IS2
0
y
L2
L2
127
128
Macroeconomics Simplified
Now let us assume that the government attempts to boost the level of output by pursuing an expansionary pure fiscal policy. The IS curve shifts from IS1 to IS2 as a result.
This increase in demand will, as in the Keynesian model, result in the running down
of stocks. However, unlike in the Keynesian model, this will generate an increase in the
price level rather than an increase in output because the economy is already operating
at full capacity. Instead of responding by increasing output, neoclassical entrepreneurs
simply raise prices in the face of increased demand.
The rise in price level from P1 to P2 is shown in the north-east quadrant by an inward
shift in the real money supply from M/P1 to M/P2. This contraction in the real money
supply results in a backward shift in the LM curve from LM1 to LM2. (People find themselves with insufficient holdings of real balances to finance transactions, so they sell
bonds, bond prices fall and yields rise.) Prices rise (and hence the LM curve will shift
backwards) until the full-employment equilibrium (yf) is restored. In the diagram, this
point is yf,i2 (as determined by the intersection of LM2 and IS2).
In other words, there is no increase in output in the final result of an expansionary
fiscal policy. All that has happened is a rise in the interest rate and in the price level.
Government spending has increased, yet total demand in the economy is the same as
before the increase took place.
Can you work out why this is the case? The answer has to do with the rise in interest
rates: higher interest rates mean lower levels of investment. The rise in government
spending thus simply crowds out an equivalent amount of investment through negative effect on investment of the higher interest rate. Full crowding out is said to have
taken place.
In order to demonstrate further the processes at work in the neoclassical model, let us
examine the effect on output and the interest rate of a restrictive fiscal policy.
As shown in Figure 7.7, the economy is initially at full-employment equilibrium (yf ,i1).
When the government pursues a restrictive fiscal policy, the IS curve will shift inwards
from IS1 to IS2. This decreases demand in the economy below the full-employment level
of output. If output were to fall (as in the Keynesian model) in response to the drop in
demand, then the economy would come to rest at y1,i2.
However, as this is a neoclassical model, the economic forces at work are rather different. Instead of output falling in response to lower level of demand, price adjustment
will take place. As demand falls, the goods market will be characterised by excess supply.
This impacts directly on prices. Prices will fall in order to re-equate supply and demand
in the goods market.
As prices fall (from P1 to P2), the real money supply will rise from M1/P1 to M1/P2, thus
shifting the LM curve outwards in the process, until the full-employment equilibrium
level of output is restored once again. The economy will thus come to rest at the intersection of IS2 and LM2. The full-employment level of output will be restored but at a lower
level of interest rate (i3). The net result of a restrictive fiscal policy is that full employment is restored, but this time as a result of greater investment demand (in response to
the lower interest rate) and less government spending.
L1
M/P2
L1
M/P1
M/P1
LM1
M/P2
L2
i
LM2
i1
i2
i3
IS2
y1 yf
IS1
L2
Self-test
Using a four-quadrant diagram, show what happens to the equilibrium level of
output and the interest rate when government lowers taxation. Assume a neoclassical model.
Source: Authors.
Monetary Policy
In a neoclassical model, any attempt on part of the authorities to increase the level of
output through an expansionary monetary policy is ultimately self-defeating. As was the
129
130
Macroeconomics Simplified
case with expansionary fiscal policy, this result is a function of the neoclassical assumption
that the economy is already at a full-employment level of output. Any attempt to boost
demand above this level simply results in a rise in the price level.
When the reserve bank pursues an expansionary monetary policy, the real money
supply initially increases as a result of the increase in the nominal money supply. This
is shown in Figure 7.8 by the outward shift in M/P from M1/P1 to M2/P1. The LM curve
shifts out from LM1 to LM2 accordingly (as people get rid of excess idle balances by purchasing bonds).
For the new level of demand to be satisfied, output would have to rise to y1. However,
given that there is no spare capacity in the economy (because the neoclassical model
assumes all factors of production are fully employed), it is not possible for the increase in
demand to be satisfied by an increase in output. All that happens is due to rise in prices.
L1
M2/P1
L1
M1/P1
M2/P2
y
LM1
M1/P1 M2/P1
M2/P2
i
LM2
IS
yf y1
L2
L2
As prices rise, the real money supply contracts, thus pulling the LM curve back towards
the left (as people sell bonds in order to restore their real transactions balances). This
process will continue until the full-employment level of output is once again restored.
This will occur when the original value of the real money supply is restored, that is,
when M2/P2 = M1/P1. At this point, the LM curve will have returned to its original position, LM1.
In short, an expansionary monetary policy has no impact on the real level of output.
A rise in the money supply simply generates a rise in the price level. This result should
be familiar to you: it is simply the quantity theory of money coming through in the guise
of an ISLM framework.
Self-test
Using a four-quadrant diagram, show what happens when government pursues a
restrictive monetary policy. Assume a neoclassical model.
Source: Authors.
131
132
Macroeconomics Simplified
Alternatively, one could argue that the appropriate policy response is rather to
improve the institutional workings of the financial system (e.g., reforming banks and
lending practices) and make the labour market more flexible. The more neoclassicallyminded economists tend to opt for such measures rather than go along with Keynesian
prescriptions (and we pick up this theme again in Chapter 8).
Which model is appropriate for India in the post-2008 crisis world? As shown in
Table 2.3, Indias economic growth rate dipped in 200809, picked up over the next 2
years but then slowed once more. In contrast to the rest of the world, Indian growth
was not constrained by inadequate aggregate demand. Rather inflation and capacity
constraints appear to be more pressing problems, with some commentators worrying
also about relatively poor employment growth. There are also growing concerns about
Indias neoliberal approach to economic growth and development, with some calling
for greater government involvement in the economy.10
Rather than choosing between Keynesian and neoclassical models to suit different
economic environments, one could construct a hybrid model allowing for an increase
both in output and in prices when demand increases. In such a model, the expansionary impact of an increase in demand would be dampened (but not eliminated) by an
increase in prices. This is illustrated in Figure 7.9 which shows an increase in output and
i
LM2
LM1
i2
i1
IS2
IS1
y1
y2
prices following an expansionary (pure) fiscal policy. The boost in demand pushes IS
outwards, and the rise in price level pulls LM slightly backwards.
According to Tobin, this kind of pragmatic flexibility about output and price adjustment is central to Keynesianism. When asked What are the fundamental propositions
which Keynesians adhere to?, he replied:
One way to put it is to say that there is a two-regime model of the economy. Sometimes
the economy is in a classical situation where markets are clearing (demand equals supply)
and the economys ability to produce output is supply-constrained. At other times the
economy is in a Keynesian situation in which the constraint on actual output is demand
aggregate spending. That I think is the distinction. Whereas for the real business cycle
theorists (like Ed Prescott) and new classical guys (like Robert Barro) you are always
supply-constrained. There is just one regime, and the observed cyclical fluctuations are
fluctuations in the voluntary willingness to be employed.11
Keynes himself would of course have agreed that Keynesian prescriptions become
increasingly unsuitable as the economy moved towards full employment. He was deeply
aware of the dangers of inflation and was fond of quoting Lenins dictum that there is no
subtler, surer means of overturning [capitalism] than to debauch the currency. In 1937,
when unemployment was still at 12 per cent, Keynes worried that the British economy
was in danger of over-heating. He wrote that we are in more need today of a rightly
distributed demand than of a greater aggregate demand.12
133
134
Macroeconomics Simplified
paid professionals and capitalists) store some of it in idle balances (and thus cause leakages in demand), then it follows that redistributing income to workers will be good in
terms of promoting full employment.
Three policy options thus face governments when attempting to boost economies
out of a less than full-employment equilibrium: increase government spending and/or
reduce taxation; expand the money supply or redistribute income in favour of those
classes with the highest propensity to consume.
According to Skidelsky, this is the hidden radicalism in Keyness theory of the consumption function:
For what Keynes has implicitly done is to make workers (those who normally save only a
tiny fraction of their incomes) the saviours of the capitalist system, businessmen its wreckers. While capitalists are febrile creatures, whose nerves can cause economies to go wild,
the workers go on stolidly consuming according to slowly changing habit, their demand
giving output such stability as it has. They lack the margin, perhaps the imagination, to
be nervous; the deadly thought is that the capitalist system will become more unstable as
workers get richer and start to behave more like capitalists.13
As Skidelsky notes, warming to his theme, the General Theory can be interpreted as a
morality play in which the social landscape:
[E]xhibits a life-denying rentier class which practises non-consumption in order to postpone the day of enjoyment; a business class driven by fantasies of triumph and disaster; a
working class victimised not by calculated oppression but by the unsteady commitments
of their controllers; and, on the other side of the divide, a radiant vision of cities beautified
and marshes drained, and the good life brought within reach of all under the benevolent
guidance of a Platonic state.14
Keynes believed that the solution to the crisis of capitalism (as manifested by the
Great Depression) was the wise economic management by an elite of dedicated and
skilled civil servants. (This lead his biographer to observe drily that senior economists
of the time clung to the notion of the self-adjusting market because they mistrusted
Keyness theory of the self-adjusting politician.)15
The question of state efficiency is very important. If governments were indeed comprised of skilled and efficient civil servants who were able to implement wise policies that
encouraged rather than undermined private investment, then the nature of the debate
over fiscal policy would be very different. It would be far more focused on questions like
the optimal allocation of spending between education and infrastructurerather than
on large ideological battles over the optimal level of the deficit.
Skidelsky, John Maynard Keynes, pp. 49899.
Ibid., p. 543.
15
Ibid., p. 593.
13
14
Ibid., p. 235.
Blyumin, quoted in ibid., p. 582.
135
136
Macroeconomics Simplified
LM1
LM1
IS3
IS2
IS1
IS2
IS3
IS1
y
A: A
n increase in debt-financed government
spending boosts private investment
(as investment responds positively to
demand)
y
B: A
n increase in debt-financed government
spending reduces private investment (as
investment reacts negatively to the rise
in government debt)
25%
Real investment
growth
20%
Inflation (GDP
deflator growth)
10%
Total government
deficit as % of GDP
0%
Short-term interest
rate
201112
201011
200910
200809
200708
200607
200506
200405
5%
200304
0%
Figure 7.11 Trends in Real Interest, Investment and the Deficit in India 200304
to 201112
Source: Reserve Bank of India: http://dbie.rbi.org.in/DBIE/dbie.rbi?site=statistics (downloaded 30 March 2013).
(i.e., balanced) fiscal policies. In 2010, Indias government pledged to shrink its budget
deficit by more than one percentage point of GDP in order to send the right signal to
investors about the governments seriousness to cut debt.19
This ideathat contractionary fiscal policies may have expansionary effects
assumed the status of a conventional wisdom worldwide during the 2000s, despite limited evidence for it and its dubious theoretical underpinnings.20 However, with the 2008
crisis, most countries changed tack in practice and reverted to Keynesian strategies of
employing fiscal stimulus to keep their economies afloat.
India, which weathered the global financial storm relatively easily, continued to act
on the assumption that prudent monetary policy was necessary for attracting investment
(and the authorities were probably justified in doing so given the jittery financial markets
and inflationary pressures). In terms of the Fiscal Responsibility and Budget Management
Actwhich constrains the fiscal deficit within defined targets21there is relatively little
19
Mukherjee Pledges to Narrow Indias Budget Deficit as Economy Accelerates. Industry News, 26
February 2010.
20
See, Blinder, A. 1997. A Core of Economic Beliefs? Challenge 40 (4) July/August: 4143.
21
See, Centre for Budget and Governance Accountability, FRBM-A Review, 2007, New Delhi for a discussion of the act and related economic policies.
137
138
Macroeconomics Simplified
room for manoeuvre on the fiscal policy side. However, in the post-2008 period, India
has allowed the deficit to exceed these targets, a kind of accidental Keynesianism which
some commentators believe might threaten fiscal stability in the future.22 Indeed, by
early 2013, the influential Economist magazine was describing Indian public finances as
a walk on the wild side and worrying about the level of public debt.23
Whether one worries more about the deficit and its impact on interest rates and
hence on private investment, or about the possibility that cutting government spending (to reduce the deficit) will have an even worse effect on growth and investment,
is a matter of judgement and theoretical presuppositions. Figure 7.11 shows that real
investment in India grew strongly between 200304 and 200405 as real interest rates
fell and economic growth picked up pace. This suggests that both the cost of borrowing
and the accelerator effect of growth on investment may have been important. But even
though real interest rates declined between 200708 and 200809, investment plummeted sharply as growth slowed. This suggests that growth was an important driver of
investment. According to estimates by the International Monetary Fund (IMF), changes
in real interest rates between 2006 and 2012 explained about a quarter of the change in
investment and that supply bottlenecks and slowing growth were also important.24
Note that the government deficit started rising sharply as Keynesian expansionary
fiscal policy characterised post-2008 crisis policy. To some neoclassically- minded economists, this might suggest that the government deficit crowded out private investment
to some extent (although the cost of borrowing was negative in real terms during this
period so this is unlikely). For more Keynesian- minded economists, the rise in the deficit was necessary to prevent an even greater collapse in growth and investment.
But the rise in short-term interest rates and in inflation from 2009 to 2010, however, were probably signalling that such expansionary policies were not sustainable. By
February 2013, the IMF was emphasising the need for decisive fiscal consolidation,
notably for a reduction in untargeted subsidies (like fuel subsidies) and for an increase
in tax revenues, including a general sales tax and greater efficiency in revenue collection.
The IMF argued that a restructuring of government expenditure towards greater reliance on cash transfers to the poor and better targeted investment expenditure on infrastructure would help limit the impact of a lower budget deficit on demand and hence
growthand would assist in unlocking the supply-side constraints that were slowing
investment and growth.25
Chhibber, A. and T. Palanivel. 2009. India Manages Global Crisis but Still Needs Serious Reforms for
Sustained Inclusive Growth. Stanford Center for International Development Annual conference on Indian
Economic Policy Reform.
23
A Walk on the Wild Side. The Economist, 23 February 2013. p. 42.
24
IMF Country Report No. 13/37. India: 2013 Article IV Consultation, February 2013. http://www.imf.
org/external/pubs/ft/scr/2013/cr1337.pdf, p. 21.
25
See, IMF Country Report No. 13/37. India: 2013 Article IV Consultation, February 2013: http://www.
imf.org/external/pubs/ft/scr/2013/cr1337.pdf
22
139
CHAPTER 8
A closed-economy three-sector aggregate supply (AS) aggregate demand (AD) model can
be generated by adding the labour market to the ISLM model (which is summarised in
the AD curve). It is a synthesis model in that it incorporates both price and output adjustment processes. Like the ISLM model, it comes in Keynesian and neoclassical versions.
The Keynesian less than full-employment equilibrium is modelled in the ASAD
framework in terms of labour market failure. Keynesian economics is boiled down in
this model to the special case scenario associated with the failure of wages to adjust
to market-clearing levels. This three-sector treatment of Keynesianism has come to be
known as the neoclassical synthesis in part because it explains Keynesian outcomes primarily in terms of the neoclassical narrative of imperfectly functioning labour markets.
The three-sector ASAD model shows that incorporating the labour market and
shifting from quantity adjustment to a combination of quantity and price adjustment
results in a different understanding of economic processesand in different policy
prescriptions. It also allows us to discuss the relationship between unemployment and
inflation.
The AS Curve
As shown in Chapter 3, the neoclassical AS curve depicts the relationship between output and price level as determined by the labour market and the production function.
In a neoclassical model, nominal wages are assumed to adjust quickly and efficiently
to ensure that a market-clearing real wage always prevails (see Chapter 3). Under these
conditions, full employment always exists and the AS curve is thus vertical at the fullemployment level of output. This is shown in Figure 8.1.
P
AS
P3
P1
W3
W/P
W1
P2
W2
(W/P)2
(W/P)1
(W/P)3
y1
N2
N3
N1
N'3
SN
N'2
DN
N
y = f (N)
141
142
Macroeconomics Simplified
Rather than assuming that money wages will decline during times of excess supply in
the labour market, the Keynesian version of the model assumes that money wages are
inflexible (or sticky) in a downward direction. This idea reflects Keyness insight that
workers will resist attempts to reduce their money wages but are less likely to protest
declines in real wages because it affects everyone equally:
Every trade union will put up some resistance to a cut in money-wages, however small.
But since no trade union would dream of striking on every occasion of a rise in the cost
of living, they do not raise the obstacle to any increase in aggregate employment which is
attributed to them by the classical school.5
Let us now derive this Keynesian AS curve, that is, assuming that money wages are
rigid downwards. Consider Figure 8.2. The combination of price level P1 and W1 gives
an equilibrium real wage that clears the labour market. The economy is thus operating
at the full-employment level of output, y1.
Now, consider what happens if the aggregate price level drops from P1 to P2. The
lower price level in combination with the same money wage (W1) results in a higher real
wage (W/P)2. At this higher real wage, only N2 amount of labour will be demanded. And,
as indicated by the production function, N2 of labour will produce only y2 of output. The
P
AS
P1
P2
W1
W/P
(W/P)2
(W/P)1
y2
y1
N1
N2
SN
DN
y = f (N)
N
new point on the AS curve is P2,y2. The Keynesian AS curve thus slopes upwards to the
point of full employment.
Unemployment in this model is thus assumed to be the result of failure of wages to
adjust downwards. It implies that the problem lies with wage-setting institutions and
labour laws that enable workers to protect their (nominal) wage settlements at the cost
of employment.
At the point of full employment, Keynesian dynamics give way to neoclassical dynamics, and the economics of full employment (rather than recession) applies. As Keynes
puts it: When a point has been reached at which the whole of the labour and capital
equipment of the community are employed, further increases in effective demand would
have no effect whatever except to raise prices without limit.6 In other words, as soon as
the economy reaches full employment, then the neoclassical model applies.
So, if the price level rises above P1 in Figure 8.2, then workers will immediately resist any
fall in real wages by bidding up the money wage. As in the neoclassical AS curve derivation, the money wage will rise to restore the full-employment equilibrium wage (W/P)1.
Consider what happens to the AS curve in a Keynesian model if the money wage
rises exogenously from W1 to W2. As can be seen in Figure 8.3, AS1 is derived assuming
a money wage of W1. If money wages rise from W1 to W2, then every level of output
up until the point of full employment will be associated with a higher price level. The
proportion of the AS curve below full employment thus shifts upwards from AS1 to AS2.
P
P3
P1
P2
W2
W1
W/P
AS2
AS1
y
N1
SN
DN
y = f (N)
N
143
144
Macroeconomics Simplified
LM3 (P3)
LM1 (P1)
i3
LM2 (P2)
i1
i2
IS
y
P
P3
P1
P2
AD
y3
y1
y2 y
145
146
Macroeconomics Simplified
Note that the AD curve consists of combinations of output and price level for which
the goods and money markets are in equilibrium. It thus serves as a summary curve of
dynamics in the goods and money market. All points to the left of the AD curve represent positions below IS and all points to the right of AD represent positions above IS.
It is important to distinguish between a movement along the AD curve and a shift of the
curve. The above derivation of the AD curve involved movements along the curve: as the
price level was altered, a corresponding level of output was determined. In contrast, shifts
of the curve occur when exogenous variables change. Shifts of the IS or LM curves, when
caused by anything other than a change in the price level, will result in a shift in AD.
Consider an expansionary monetary policy. An outward shift in the (real) money
supply will cause the LM curve to shift out from LM1 to LM2 (Figure 8.5). Output rises
from y1 to y2 as a result. As the price level has not changed, the new combination of output and price level becomes P1,y2. The AD curve thus shifts to the right. If an expansionary fiscal policy is pursued, then both the IS and AD curves will shift out (Figure 8.6).
Self-test
13
14
Pigou, A. 1943. The Classical Stationary State. Economic Journal 53 (212): 34351.
Patinkin, D. 1956. Money, Interest and Prices. Evanston: Row, Peterson and Company.
LM1 (P1)
LM2 (P1)
i1
i2
IS
y1
y2
P
P1
AD2
AD1
LM
i2
i1
IS1
y1
y2
IS2
y
P
P1
AD2
AD1
y
147
148
Macroeconomics Simplified
AS
P1
P2
P3
W1
AD
W/P
y3 y2 y1 y4
N2
N1
SN
y = f (N)
DN
N
supplied (y1) is thus inconsistent with equilibrium in the goods and money markets at
that price level. Given that y1 is greater than y3, this situation can be characterised as one
of excess supply.
In the two-sector Keynesian ISLM model, excess supply results in a build- up of
stocks which sends a signal to producers to cut back on output. In the three-sector
Keynesian ASAD model, excess supply also causes prices to fall from P1 to P2, thus
raising real wages. This results in a fall in employment and output until equilibrium is
restored at P2,y2.
Consider another point on the AD curve, P3,y4. At that low price level, real wages will
be relatively high, and hence, employment and output will be low. From the AS curve, it
was observed that the output will only be y3, when prices are at level P3. In other words,
we have a situation of excess demand. Prices will thus be bid up. As prices rise, real wages
fall. Employment and output will rise until equilibrium is restored.
LM1
i3
i2
i1
IS2
IS1
y
AS
P2
P1
AD2
AD1
y1
y3 y2
149
150
Macroeconomics Simplified
In terms of the ASAD framework, the expansionary fiscal policy will shift the AD
curve from AD1 to AD2. This will result in a price rise (from P1 to P2), and the economy
will come to rest at y3,P2. If the effect of the price rise was tracked back into the ISLM
model, this would be expressed in terms of a decrease in the real money supply and
hence in a backward shift in the LM curve from LM1 to LM2.
In other words, the three-sector model, by building in the effect on prices of an
increase in demand, shows that an increase in demand has a smaller impact on output
than the ISLM model. The multiplier is thus smaller in the three-sector Keynesian
model than in the two-sector model.
In this three-sector neoclassical synthesis version of Keynesianism, the less than fullemployment equilibrium is explained in terms of labour market failure.
Consider an economy at the full-employment equilibrium position of P1, y1 in Figure
8.9. Now let us assume that for some exogenous reason (e.g., an inexplicable loss of business confidence), the AD curve shifts to the left from AD1 to AD2. The economy plunges
into a recession, bringing down prices from P1 to P2. The lower price level causes real
wages to rise from (W/P)1 to (W/P)2. This causes employment to fall from N1 to N2, thus
resulting in a drop in output from y1 to y2. What allows the economy to stagnate at y2
is simply the fact that workers refuse to allow their money wage (W1) to fall. In other
words, their real wages are too high to clear the labour market. The rise in real wages
P
AS
AD2
P1
AD1
P2
W1
W/P (W/P)
2
(W/P)1
y2
y1
N2
N1
SN
y = f (N)
DN
N
(as a result of the fall in the price level) is thus responsible for the fall-off in employment
and the decline in output.
When the Keynesian less than full-employment equilibrium is explained in these
terms, the way in which expansionary demand policies impact the economy alters
accordingly. As you will recall from the two-sector ISLM model, an expansion in
demand leads to a decrease in inventories, which serves as a signal to producers to
increase output. In this three-sector model, the increase in demand also results in an
increase in output, but this time mediated also through the price mechanism and its
impact on real wages.
In this model, the less than full-employment equilibrium is the result of the price
level sinking (due to the impact of recessionary forces) and, in doing so, raising real
wages above market-clearing level. In Keyness view, the solution to the problem was
for the government to pursue expansionary fiscal and monetary policies in order to
boost the level of demand back from AD2 to its original full-employment level of AD1
in Figure 8.9.
Keynes was, in fact, willing to try any means of raising demand. In a radio broadcast
in January 1931, he made the following impassioned appeal:
The best guess I can make is that whenever you save five shillings, you put a man out of
work for a day. ... On the other hand, whenever you buy goods you increase employment
though they must be British, home-produced goods if you are to increase employment in
this country. ... Therefore, O patriotic housewives, sally out tomorrow early into the streets
and go to the wonderful sales which are everywhere advertised. You will do yourselves
goodfor never were things so cheap, cheap beyond your dreams. Lay in a stock of household linen, of sheets and blankets to satisfy all your needs. And have the added joy that you
are increasing employment, adding to the wealth of the country because you are setting
on foot useful activities, bringing a chance and a hope to Lancaster, Yorkshire and Belfast.
... For what we need now is not to button up our waistcoats tight, but to be in a mood of
expansion, of activityto do things, to buy things, to make things.15
Whatever the cause of increased demand, the effect is the same: prices rise and inventories fall and producers get a clear signal to increase output. But note that the rise in
price is more than a mere signal, as it facilitates a drop in real wages, which in turn
allows firms to increase employment and output from y2 to y1.
For Keynes, bringing about such an adjustment to real wages by increasing the price
level was far more sensible a policy than attempting to drive down nominal wages:
The business of forcing down certain levels of wages, and so forth, into equilibrium is
almost hopeless, or it will take a long time. The continuance of unemployment is to an
important extent due to the fact that we have got the level of wages ... out of gear with
everything else. The only way in which they will get into gear will be by an increase in the
level of prices.16
15
16
151
152
Macroeconomics Simplified
Those thinkers who are more neoclassically minded, however, prefer to act directly
on nominal wages (typically by eroding the power of labour to drive up nominal wages)
in order to bring about an adjustment to real wages. Indeed, breaking trade-union power
was an explicit and important aspect of Thatcherist economics in Britain under Prime
Minister Margaret Thatcher. Thatcher also promoted the privatisation of state assets and
lower taxation on the rich.
AS
P1
AD1
P2
W1
W/P
AD2
W2
(W/P)2
(W/P)1
y1
N1
SN
DN
y = f (N)
N
153
154
Macroeconomics Simplified
C+I+G2
C+I+G1
45
y1
y2
i
IS2
i2
LM1
IS1
i1
y
y1
y3
y
AS
AD1
AD2
P2
P1
y
y1 y4 y3 y2
Figure 8.11 The Effect of an Expansionary Fiscal Policy in the One-sector, Twosector and Three-sector Keynesian Models
Source: Authors.
more muted). However, the danger with using more complex models is that if one
set of relationships is wrong, the entire modelling exercise can be thrown off course.
And the more complex the model, the greater the need for reliable and accurate data.
When the data are poor and when relationships are unclear, simple models are sometimes more appropriate.
155
156
Macroeconomics Simplified
The challenge facing policy makers is to balance short-run supply-side and demandside policies in a way that does not undermine sustainable growth. If demand rises in
excess of existing surplus capacity, then the danger of inflation arises. This outcome has
to be avoided by increasing productive capacity (by the public or private sectors) at the
same time.
Note that investment works on the demand side (as modelled in earlier chapters)
by increasing spending on capital goods. But if one takes a longer term perspective, it
also works on the supply-side by increasing the capital stockthereby shifting out the
production function in long-run growth models. For this reason, policy makers keep
a careful eye on trends in fixed capital formation and try to find ways to boost private
investment by carefully-targeted public investment, for example in economic infrastructure such as roads and ports.
Rate
of change
in wages
The
Phillips
curve
Unemployment rate
periods of low unemployment and (2) if prices are determined in part as a mark-up on
costs, wage inflation translates into price inflation.
157
158
Macroeconomics Simplified
and demand in perfectly competitive markets), the Phillips curve can be drawn as a
relationship between prices and unemployment rather than as a relationship between
wages and unemployment. The priceunemployment version appears in the Keynesian
dynamic model discussed below.
But let us pause for a moment and consider the difference between the above Keynesian theory of inflation and the standard neoclassical explanation. As shown most baldly
in the Quantity Theory of Money, a fundamental tenet of neoclassical thinking is that
the money supply determines the price level. This result derives from the assumption
that the economy is at (or gravitating rapidly towards) full employment. The argument is taken slightly further in the ISLM neoclassical model, which suggests that any
expansionary government policies (be it monetary or fiscal) will ultimately generate
only inflation. The bottom line of this theoretical approach is thus that governments
cause inflation through their ham-handed attempts to fine-tune the economy. According to neoclassical theory, governments should facilitate the smooth operation of market
forces, and not attempt to regulate the level of employment by influencing the level of
demand.
By constructing an explanation of inflation in terms of the Phillips curve, Keynesians
were able to justify continued government intervention. In this model, prices rise as the
economy moves towards full employment. In other words, inflation becomes the necessary price for obtaining the social goal of minimal unemployment.
In short, the Phillips curve relationship suggests that there is a tradeoff between
the rate of unemployment and the rate of inflation. Policy makers must simply make
a judgement as to what combination is the most desirable and then design policies
accordingly.
In his influential presidential address to the American Economic Society,19 Milton
Friedman argued that the Phillips curve was mis-specified because it was related to
the rate of change of money wages rather than real wages. He pointed out that workers
are interested in real wages, not money wages, and that they bargain wages in terms of
some expectation about inflation. He therefore augmented the Phillips curve with the
anticipated rate of inflation as an additional variable determining the rate of change of
money wages. It is to a discussion of the role of expectations formation in bargaining
that we now turn.
See Friedman, M. 1968. The Role of Monetary Policy. American Economic Review 153 (1): 117.
Assuming naive expectations on the part of workers, the model can show the possibility of a constant tradeoff between the rate of unemployment and the rate of inflation.
As will become clearer later on, it is only under the assumption of naive expectations
that such a tradeoff is stable over time.
In order to show a stable relationship between the rate of inflation and the rate of
unemployment, it is necessary to trace the bargaining and price-setting process over
time. The model below traces such a path over three distinct time periods (in this case,
years). For the sake of simplicity, we will assume that workers and employers bargain
over money wages at the end of each year. They sign a contract that remains in force
until the end of the following year, at which point it is renegotiated and enforced over
the next year.
Because a rise in the price level erodes the real value of money wages, workers have
to make a judgement about the likely change in prices over the contract period and take
this into account when bargaining over their money wages. The kind of information
which workers use in forming these judgements, and the way in which such information is processed into expectations about the future, are captured in economic models
through an expectations function.
The most simple model of expectations formation is the naive expectations function, which assumes that prices ruling in the current period will continue to rule in the
following period. The naive expectations function can be expressed as follows:
Et(Pt+1) = Pt, (8.1)
where t = current period; t + 1 = following period.
Equation 8.1 says that current expectations (Et) about the price level in the following
period (Pt + 1) are given by the price level ruling in the current period (Pt). In other words,
workers are exhibiting naive expectations by assuming that the current price level will
continue unchanged into the future. Of course, if inflation is negligible and if workers
expect the government to pursue anti-inflationary policies, then holding such expectations is not as silly as it may first appear. However, as shown below, naive expectations in
the presence of inflationary forces will result in a fall in real wages.
Assume that the economy is initially at P1,y1 in Figure 8.13. Assume also that workers
wish to obtain a real wage of W1/P1 (i.e., (W/P)1). On the assumption that the existing
price level (P1) will hold in the future, they bargain a money wage of W1, which binds
them into a contract for the following period (the next year).
Now let us assume that the government wishes to boost employment and output.
Expansionary fiscal (and/or monetary) policies are pursued which push the AD curve
out from AD1 to AD2. This results in a rise in the price level to P2, a fall in the real wage
(to (W/P)2), an increase in employment and hence an increase in output to y2.
Because workers are bound into a money wage contract, they cannot defend this erosion of their real wage. However, at the end of the year, they attempt to restore their real
wage by bargaining up the money wage. What money wage will they bargain? Assuming
159
160
Macroeconomics Simplified
P3
W2
P2
W1
W/P
AS2
P1
(W/P)1
AS1
AD1
AD2
y1 y3 y2
(W/P)2
Nd
NS
SN
DN
y = f (N)
N
naive expectations on the part of workers, the model assumes that workers will act on
the expectation that the current price level (P2) will continue into the next period. They
thus attempt to restore their original real wage, (W/P)1, by bargaining a money wage of
W2. This results in an increase in the real wage and in an upward shift in AS from AS1 to
AS2. The new equilibrium is at P3,y3.
The unexpected increase in the price level (from P2 to P3) erodes the real wage of workers. Consequently, at the end of the year, they will bargain an increase in the money wage
on the assumption that P3 will continue into the next year. However, for the same reasons outlined for the previous period, the resulting upward shift in the AS curve ensures
that the price level will rise further than expected, and hence that the bargained real
wage is eroded once again.
The message of the story is that the attempt to maintain a rate of unemployment
lower than that associated with the desired real wage of workers, results in a constant
rate of inflation. The story is consistent with the Phillips curve, and implies that government can trade off inflation and unemployment. Figure 8.14 draws the Phillips curve
associated with the model in Figure 8.13.
The rate of inflation over two periods = (P3 P1)/P1 = P*
The rate of unemployment = (Ns Nd)/Ns= U*
(8.2)
(8.3)
Rate of
inflation
P
Unemployment rate
161
162
Macroeconomics Simplified
Rate of
inflation
P
P
P
Unemployment rate
inflation rate might be P*; however, in year 2, it is likely to be P**; and similarly, in year
3, it could be P***. This is shown in Figure 8.15. In other words, the Phillips curve could
shift upwards over time. The tradeoff between the unemployment rate and that of inflation thus becomes a short-run phenomenon only, with the long-run Philips curve being
vertical. Monetarists, such as Milton Friedman, used this kind of argument to attack
Keynesian economics during the 1970s.
Rational Expectations
Some economists, however, believe that even the idea of a short-run tradeoff between
inflation and unemployment is mistaken. By adopting a rational expectations framework,
they argue that economic agents are much better informed and wiser than suggested by
the adaptive expectations framework. They assume that people understand economic
theory and practice so well that they can accurately predict the path of inflation in the
future. The inflation forecast equation thus becomes
Et(Pt + 1) = Pt + 1. (8.5)
Assuming such perfect information and also flexible wages and prices, the rational
expectations model removes even the possibility of a short-run tradeoff between unemployment and inflation.
Many fancy mathematical economic models have been constructed on the basis of
rational expectations. The real business cycle models referred to in Chapter 3 fall under
this category. The problem, however, is the extreme implausibility of the assumption.
Information and understanding are costly to acquire. And, considering that the predictions of sophisticated economic forecasting models are regularly proved wrong, what is
the point of assuming that economic agents can do any better, even if they had all the
available information and understanding?
Nevertheless, the rational expectations school correctly places the spotlight on the
ways in which expectations, on part of economic agents, can potentially undermine government policies. Rational expectations enthusiasts may have been carried away with
the beauty of their models, but they have nevertheless shown how inappropriate it is to
assume naive expectations on part of economic agents.
The NAIRU
The possibility of a relationship between unemployment and accelerating inflation poses
problems for policies designed to reduce unemployment. The bouts of inflation that
beset the advanced capitalist countries from the late 1960s have generally made policy
makers more cautious about stimulating demand. Some economists warn that attempts
to lower unemployment below the natural rate of unemployment (or the non-accelerating inflation rate of unemployment, the NAIRU), could be counterproductive. In terms
of Figure 8.13, the NAIRU would be that level of unemployment associated with the real
wage that workers insist on defending (W/P)1. Any level of unemployment below that is
assumed to generate accelerating inflation through wageprice spirals.
163
164
Macroeconomics Simplified
(Box Continued)
the existence of adaptive expectations. Dholakia and Sapre take this work forward and find that there is a regular tradeoff between inflation and output (or
unemployment) with inflationary expectations based on the experience of the
past 34 years.22
Source: Authors.
One of the problems with the concept of a NAIRU is that it is very difficult estimating
the unemployment levels associated with it. Some economists believe that the idea is so
abstract (even incoherent) that it is of little practical significance. A further problem is
that the NAIRU hypothesis seems to fit the real-world experience less and less well as
time goes by. The notion of an equilibrium rate of unemployment which will result in a
stable inflation rate has been severely challenged by the experience of the 1980s in the
most advanced capitalist countries, when unemployment persisted despite a significant
reduction of inflation. Deviation from the NAIRU occurred almost as if the NAIRU
increased with unemployment itself. In any event, the serious rise in unemployment
globally as growth faltered after the 2008 economic crisis has refocused policy makers attention on boosting growth and employment rather than worrying about abstract
notions with little contemporary relevance such as the NAIRU. In India, for example,
employment rates were lower in 2011 than in 2007.23 This has serious implications for
inequality and growth.
labour market policy is more active, and welfare protection stronger in the co-ordinated
market economies of Germany and Scandinavia.24 The key idea is that firms in the more
co-ordinated market economies accept higher labour costs (and taxation)so long as
the system delivers the supply of skills, active industrial policies and long-term finance
they need to operate profitably.
This analysis implies that different paths to growth are possibleand that under
certain conditions, employment protection is consistent with rapid productivity and
income growth. A similar argument is made by those who distinguish between a low road
to growth in which firms have no incentive to get out of a low-wage, low-productivity
equilibrium, and a high road which uses employment protection (which raises the cost
of labour for business) to force firms to become more competitive. For example, with
regard to India, Sharma argues that: Only when the path to competition on the basis of
low wages and bad working conditions is barred by providing a floor of labour standards25
are firms likely to innovate.
DSouza argues that Indias approach to labour-market regulation rests on such theoretical foundations
Job security regulation in a state-led economic growth regime was treated more as a sunk
cost that was required to induce investment in skills and as a form of social insurance. It
was a means of spurring the acquisition of skills which makes labour more efficient in the
production process.26
In other words, what this argument suggests is that Indian policy makers assumed that
when faced with higher labour costs (caused by employment protection), firms would
improve productivity and innovate rather than simply move down the labour demand
curve and shed labour. We could represent this assumed positive effect of employment
protection (and other government policies which raise the cost of labour) in the ASAD
framework as a simultaneous outward shift in the nominal wage function and in the
labour demand curve. In the most optimistic version of this growth narrative, both the
AS and the AD curves shift outwards as income and productive capacity increases.
This high road to growth is possibleand indeed is a vision promoted by the International Labour Organisation (ILO). The ILO recommends that governments take active
steps to promote the growth of decent jobs (i.e., higher wage, higher productivity jobs)
24
Hall, P. and D. Soskice. 2001. An Introduction to Varieties of Capitalism. Varieties of Capitalism: The
Institutional Foundations of Comparative Advantage, pp. 168, eds. Hall, P. and D. Soskice. Oxford: Oxford
University Press.
Hall, P. and D. Gingerich. 2009. Varieties of Capitalism and Institutional Complementarities in the
Political Economy: An Empirical Analysis. British Journal of Political Science 39 (3): 44982.
25
Sharma, A. 2006. Flexibility, Employment and Labour Market Reforms in India. Economic and Political Weekly, May.
26
DSouza, The Employment Effects of Labour Legislation, p. 130. See also, DSouza, E. 2005. Are
Retrenchment Laws Inefficient? The Indian Journal of Labour Economics 48 (4).
165
166
Macroeconomics Simplified
and to accompany this with expansionary macroeconomic policies.27 However, increasing the cost of labour can undermine employment growth as some firms relocate all or
parts of their production processes to lower wage countries. And, in places where informal employment is tolerated, one might expect more firms to escape labour-market
regulation by moving into the informal sector. The challenge for government is to assist
in the creation of decent jobs whilst ensuring that its industrial policies provide support
to firms where necessaryand that the welfare system can provide for those unlucky
enough to remain unemployed. In this respect, Indias employment guarantee schemes
are of utmost importance in ensuring that growth is genuinely inclusive. Tolerating, even
supporting low-wage, labour-intensive sectors is also necessary where welfare budgets
are stretched and unemployment is high.
167
CHAPTER 9
So far, our theoretical discussion has been confined to a closed economy. This chapter
extends the Keynesian ISLM framework to include the international sector and concludes with a discussion of the role of prices.
14,825
23,946
9,121
2. Invisibles, net
5,362
3. Current account (1 + 2)
3,760
4. Capital account (A to F)
3,074
A. Foreign investment
2,418
120
421
4
582
465
685
685
Source: Reserve Bank of India: Key components of Indias balance of payments. http://dbie.rbi.
org.in/DBIE/dbie.rbi?site=statistics, downloaded 1 April 2013.
capital account reflects international financial transactions with the rest of the world,
and in effect shows how current account deficits are financed. The items in this account
include net foreign direct investment (i.e., investment by foreigners in undertakings in
India minus investments by Indians in undertakings abroad) and portfolio and other
investments.1 In the current account, invisibles include services (travel, transport, insurance, miscellaneous items like software, business, financial and communication services), transfers (official and private) and income (investment income and compensation of employees). The total current account is the sum of merchandise and services.
The capital account includes foreign investment (foreign direct investment and
portfolio investment in India and abroad), loans and external assistance to and from
India, rupee debt service and other capital. The overall balance is the sum of the current
account and the capital account. When this is negative (as was the case in 201112),
there will be a decline in reserves. Loans from the IMF can help to reduce the pressure
on reserves, but no such loans were obtained in 201112.
Table 9.1 shows that the Indian economy in 201112 was running a trade deficit of
`9,121 billion. Capital inflows of `3,074 billion were insufficient to finance this deficit so
the overall balance was negative and Indian reserves fell by the difference (`685 billion).
The last time this state of affairs existed was in 200809 (Figure 9.1). Figure 9.1 shows
27,500
22,500
17,500
Imports
Exports
12,500
Invisibles
Current account
7,500
Capital account
2,500
Overal balance
Decrease in reserves
2,500
201112
201011
200910
200809
200708
200607
200506
200405
12,500
200304
Trade balance
7,500
169
170
Macroeconomics Simplified
that the main driver of the Indian BOP deficit from 200304 to 201112 was the steadily
widening current account deficit as import growth outstripped export growth.
SS1
`/$
SS2
e1
e2
DD1
a
all exchange rates (Figure 9.2). There will be an excess supply of foreign exchange equal
to ab, and foreign-exchange reserves will rise by that amount if the reserve bank supplies
the increased demand for rupees.
If there is no reserve bank intervention, the excess supply will result in a fall in the price
of foreign exchange from e1 to e2, that is, the rupee will appreciate. In the absence of compensating price changes, this will dampen export demand (and hence reduce the supply
of dollars to the foreign-exchange market) and boost import demand (thus increasing
the demand for dollars) until BOP equilibrium is restored.
This process of adjustment may also have implications for capital flows, and hence,
movements in the capital account will also affect the market for foreign exchange.
If, for example, interest rates decline along with aggregate demand, then capital will
probably leave the country, thus increasing the supply of rupees and further encouraging
depreciation.
According to economic theory, under a system of perfectly flexible exchange rates,
the nominal exchange rate adjusts instantaneously to equilibrate SS and DD, changes
in reserves are zero, and the BOP is therefore always in equilibrium. In practice, however, most central banks (that is, reserve banks) intervene occasionally in the foreignexchange market. Some attempt to contain the exchange rate within a specified target
range. Where a central bank stands ready to buy or sell foreign exchange at a fixed rate,
a fixed exchange rate regime is said to be in operation.
Between 1945 and 1973, most major economies operated fixed exchange rate regimes.
Prior to the Second World War, exchange rates were fixed by the gold standard. After the
war, the Bretton Woods agreement created the gold exchange standard whereby the dollar was tied to gold, and all other currencies were fixed in terms of dollar. The IMF was
established to provide support for countries experiencing temporary BOP difficulties.
Thus, if a country found its foreign-exchange reserves were running too low to allow the
central bank to continue intervening in support of the fixed exchange rate, then it could
go to the IMF for a loan. If economic conditions were serious enough to warrant a more
fundamental adjustment, deficit countries were expected to devalue their currencies.
The system worked well until the 1960s when the German and Japanese economies
started becoming very competitive. As the demand for American goods fell, so too did
the demand for dollars. At the same time, the need to finance the politically unpopular
Vietnam War resulted in a rapid expansion of the US money supply, high BOP deficits
and increased world inflation. Tying the dollar to gold was clearly no longer sustainable.
In 1971, US President Nixon broke away from the Bretton Woods agreement and by
1973 the dollar was floating. Other countries rapidly followed suit.
Since the collapse of the Bretton Woods system, most of the worlds currencies have
floated against one another, but not freely. Some countries tie their currencies to that of
their major trading partner. Argentina, for example, maintained a fixed exchange rate
with the US dollar between 1991 and 2002, and Senegal maintained a fixed exchange rate
171
172
Macroeconomics Simplified
with the French franc (and now the euro). Other countries tie their currencies loosely
to the dollar (or some other major currency), or attempt to keep their currencies in line
with some target range. Because depreciation has inflationary consequences and appreciation can cause harm to export industries, most countries adopt a system of managed
exchange rates in order to maintain some exchange rate stability.
During the 1990s, most European countries joined a common monetary system
in which the various central banks agreed to maintain their respective exchange rates
within a narrow band. This culminated in the European Monetary Union in 1999. Since
then, participating countries have fixed their currencies to the new European currency
(the euro), which is now controlled by the European Bank. This has made independent
monetary policy by member countries impossible.
Consider how a central bank may intervene in the foreign-exchange market in order
to maintain a fixed exchange rate. Assume the central bank wishes to maintain the
exchange rate at e1 in Figure 9.3 but is suddenly faced with an increased demand for foreign exchangeperhaps the result of lower interest rates that boost spending (including
on imports) and encourage capital flight. The demand for foreign currency shifts out
to DD2 as a consequence. There is now an excess demand for foreign exchange equal to
ab. In the absence of central bank intervention, the domestic currency would depreciate (rise to e2). If the central bank wished to maintain the current rate, it would have to
accommodate this excess demand by supplying the quantity of dollars ab to the market
out of its stock of foreign exchange reserves. In effect, it would be shifting the supply
curve to the right to intersect DD2 at point b.
`/$
SS1
e2
SS2
e1
DD2
DD1
a
Note that the central bank cannot sell foreign exchange reserves indefinitely because
sooner or later it will run out of reserves.
If, conversely, the supply of foreign currency exceeded the demand for it, then the
domestic currency would appreciate (e would fall). If the central bank wished to maintain the exchange rate, it would have to buy up all the excess foreign exchange in the
market. This excess (equal to the BOP surplus) would then be added to the central
banks stock of foreign exchange reserves.
An alternative or complementary way to maintain a fixed rate is for the government
to implement economic policies that shift the demand or supply curves for foreign
exchange. For example, when faced with a BOP deficit, the government could embark
on expenditure switching and expenditure reducing policies as part of a programme
of structural adjustment. Expenditure switching policies (such as export subsidies and
import quotas) discourage imports and encourage exports, and expenditure reducing policies (e.g., restrictive fiscal policy) reduce the level of expenditure, and thereby
dampen import demand. Both improve the current account of the BOP.
The exchange rate regime has important implications for money supplyand for
the conduct of monetary policy. Under a fixed exchange rate regime, foreign exchange
reserves will rise when there is a BOP surplus and fall when there is a deficit. When
reserves rise (as the central bank provides domestic currency to meet demand), so does
the money supply and vice versa. Under flexible exchange rates, by contrast, there are no
overall deficits or surpluses. Changes in reserves will always be zero, and there will be
no impact on the money supply from the BOP. These dynamics are modelled explicitly
in the ISLMBP framework developed later on in this chapter.
173
174
Macroeconomics Simplified
(Box Continued)
the exchange rate, the objective being to minimise volatility.2 A dual-exchange
rate regime was initially introduced (which required that 40 per cent of receipts
be converted at the official rate and deposited with the Reserve Bank) but this
was soon replaced by a unified exchange rate system.
Source: Authors.
(9.1)
X, Z
Z = m + uy
} deficit
surplus {
y1
depreciation (which it is by definition in this Keynesian model where prices are assumed
constant). Furthermore, the current account will only improve following a depreciation if exports increase enough, and imports decrease enough, to compensate for the
increased price of imports.
Consider the simple one-sector Keynesian open-economy model in Figure 9.5. The
economy is initially at rest at y1. The bottom section of the diagram indicates that y1 is
associated with a current account balance equal to zero. Now assume that the government pursues an expansionary fiscal policy. This shifts aggregate expenditure from E1 to
E2, and output rises from y1 to y2. As shown in the bottom diagram, imports rise and the
current account moves into deficit. Under a flexible exchange rate regime, this will result
in depreciation (as the demand for foreign currency from importers exceeds the supply from exporters). This will boost exports (from X1 to X2) and reduce imports (as the
marginal propensity to import (u) falls) from Z1 to Z2. This adjustment continues until
equilibrium is established. Under a more managed exchange rate system, the monetary
authorities would have to facilitate a depreciation by increasing the supply of domestic
currency onto the foreign-exchange market.
Note that the BOP will improve only if export and import demands respond sufficiently to the change in exchange rate. This requirement is formalised in the Marshall
Lerner condition3 which states that the current account will improve in response to
depreciation if the sum of elasticity of demand for exports and elasticity of demand for
imports exceeds unity. In the following discussion, we shall assume that the Marshall
Lerner condition holds (i.e., we adopt an elasticities approach to the BOP).4
See Marshall, A. 1923. Money, Credit and Commerce. London: Macmillan, and Lerner, A. 1944. Economics of Control. New York: Macmillan.
4
An alternative approach is the absorption approachsee Alexander, S. 1952. Effects of Devaluation on
Trade Balance, IMF Staff Papers, April.
3
175
176
Macroeconomics Simplified
E2
E1
y
X, Z
Z1
A depreciation
reduces the
propensity to
import (u) and
hence reduces the
slope of the Z
function
Z2
X2
X1
y1
y2
Like consumption, government spending and investment, exports represent an injection of demand into the economy. Imports, however, are a withdrawal. In deriving the
IS curve in an open economy context, the equilibrium condition will thus change from
I + G = S + T to
I + G + X = S + T + Z. (9.2)
Substituting Equations 6.18 and 9.1 into the right-hand side of Equation 9.2:
S + T + Z = a + (1 b + bt)y + m + uy. (9.3)
The S + T + Z function thus has a steeper slope than the S + T function (because
imports rise with income) and the intercept rises by the amount m (i.e., autonomous
imports). This is shown in Figure 9.6.
The I + G function now becomes the I + G + X function. As exports and government spending are not a function of the interest rate, the I + G + X function obtains its
slope from the investment function. Government spending and exports merely affect
the position of the curve. The equilibrium condition becomes
j di + G + X = a + (1 b + bt)y + m + uy.
Open economy IS is y = (j di + G + X + a m)/(1 b + bt + u).(9.4)
The multiplier is 1/(1 b + bt + u).(9.5)
Note that the open economy multiplier is smaller than the closed economy multiplier because some of the increased demand is lost to domestic producers as a result of
propensity to import (u). In other words, lower the multiplier effect of an expansionary
S+T+Z= m a+(1 b+bt+u)y
S+T= a+(1 b+bt)y
ma
a
177
178
Macroeconomics Simplified
S+T+Z
S+T+Z
S+T+Z=I+G+X
S+T+Z=ma+(1b+bt+u)y
45
I+G+X
ma
i
IS2
IS1
I+G + X
Figure 9.7 The Open Economy IS Curve Before and After Depreciation
Source: Authors.
fiscal policy (or any increase in demand), the greater the propensity of the economy
to import.
Figure 9.7 derives the open economy IS curve by means of a diagram. It also shows
how depreciation affects the IS curve (assuming the MarshallLerner condition holds).
Depreciation will reduce the marginal propensity to import and hence flatten the slope
of the S + T + Z function, and boost exports, thus shifting out the I + G + X function.
Taken together, these shifts push the IS curve out to IS2. Note that the new IS curve is
more elastic than the pre-depreciation curve because of the fall in propensity to import.
Remember that this is a Keynesian model and hence prices are assumed constant and
the economy can come to rest at a less than full-employment equilibrium.
Under a flexible exchange rate regime, the nominal domestic money supply remains
unaffected by developments in the foreign-exchange market. The supply and demand for
foreign exchange is brought into line by changes in the exchange rate, and hence, there
will be no net impact on the nominal money supply. A flexible exchange rate regime
LM3
LM1
LM1
LM2
y
A. The LM Curve under a
Fixed Exchange Rate Regime
y
B. The LM Curve under a
Flexible Exchange Rate Regime
ensures that the amount of domestic currency people want to trade for foreign currency
is exactly matched by the amount of foreign currency demanded.
In a fixed exchange rate world, however, the central bank is obliged to supply domestic and foreign currency on demand at a fixed rate. If more people offer foreign currency
in exchange for domestic currency than vice versa, then foreign exchange reserves will
riseas will the nominal money supply. Assuming no change in prices (which is the
case in a Keynesian model), the LM curve will thus shift out from LM1 to LM2 in panel
A of Figure 9.8. If the demand for foreign currency exceeds the available supply, then
foreign reserves will decline (as will the nominal money supply) and the LM curve will
shift backwards from LM1 to LM3.
The BP line
As noted earlier, the open-economy ISLMBP model includes a third function, the
BP line, which consists of combinations of y and i for which the BOP is in equilibrium.
Given our assumption of perfect capital mobility, there is effectively one (internationally
integrated) bond market, and one ruling world yield or interest rate (i *) in the model.
Perfect capital mobility implies that residents of one country can easily purchase bonds
in another country and that foreign and domestic bonds are perfect substitutes. Thus, if
yields are higher in country A than in country B, people will sell bonds in country B and
buy bonds in country A until yields are the same in both countries. Country A will thus
experience capital inflows, and country B will experience capital outflows, until yields
are equalised.
Under such conditions, any deviation of domestic yield from the world yield will
result in BOP problems. Hence, it follows that capital account equilibrium is only possible at the world interest rate (i *) and that movements in the capital account completely
dominate the current account position. The BP line is thus perfectly elastic at i *.
179
180
Macroeconomics Simplified
X,Z
Z
X
current
account
surplus
current
account
deficit
y
S2
i
S1
i1
D1 D
2
D3
BP
capital account deficit
i2
For example, if the domestic interest rate marginally exceeded i * (e.g., i1 in Figure
9.9), then capital from abroad would flow instantly into the country (thus boosting the
capital account into surplus) and into the bond market. This will push out the demand
for bonds from D1 to D2. The overall BOP would thus be in surplus at any yield above i *
irrespective of whether the current account was in deficit or not.
If the domestic interest rate was marginally below i *, that is, at i2 in Figure 9.9, then
there would be instant capital outflow. People would sell bonds (thus pushing out the
supply of bonds to S2 and driving up yields). The impact of a yield below i * on the capital
account would be so severe and instantaneous as to ensure that the overall BOP would
be in deficit irrespective of whether the current account was in surplus (as would be the
case at any level of income below y1) or not. Overall BOP equilibrium is thus dictated by
the capital account in this model.
Capital is of course not perfectly mobile between countries nor are foreign and
domestic bonds perfect substitutes. However, as globalisation gathered pace, most
countries reduced their capital controls and capital markets became increasingly
internationally integrated. Capital account transactions dwarf current account transactions: total foreign exchange transactions in a single day exceed the annual GDP of
many industrialised countries and amount to an equivalent of over one-quarter of the
annual value of world trade. Assuming perfect capital mobility (and hence that the capital account dominates the current account when it comes to BOP equilibrium) is thus
not as unrealistic as it may at first appear.
Note that the ISLMBP framework can incorporate a change in the risk profile of
the economy. Thus, if people decided that holding domestic bonds was risky compared
to holding bonds abroad (perhaps because of fears that domestic bond-holders are going
to be taxed), then there would be a risk premium on domestic bond yields. The BP curve
would thus shift upwards by an amount of that risk premium. In other words, i * (which
is the nominal domestic interest rate) becomes the ruling world bond yield plus the
risk premium necessary to make bond-buyers indifferent between holding domestic or
foreign bonds.
So, for example, if the domestic interest rate was equal to i *1 in Figure 9.10, then
there would be no tendency for capital to flow in or out of the country. However, if bond
buyers subsequently became nervous about the domestic bond market, then the supply
of bonds will increase (as people try and reduce their domestic bond holdings) from S1
to S2. The demand for bonds may also decline exogenously (from D1 to D2). Domestic
yields would thus rise until bond-holders are compensated for the increased risk and
hence are once again indifferent as to whether they hold domestic or foreign bonds. The
BP line would thus shift up to i *2.
The full ISLMBP model is shown in Figure 9.11. Equilibrium in the goods and
money markets occurs at the intersection of IS and LM. As long as the equilibrium interest rate is i *, the BOP will also be in equilibrium. If IS and LM intersect above i *, then the
i
i2
BP2
i1
BP1
D2
D1
S2
y
The BP Line
S1
The Supply and Demand for
Domestic Bonds
181
182
Macroeconomics Simplified
i
LM
BP
i*
IS
BOP will be in surplus (as capital flows into the domestic bond market from abroad).
If they intersect below i *, the BOP will be in deficit as capital flows out of the domestic bond market and into foreign markets. This will spark off further adjustmentthe
nature of which varies depending on whether a fixed or flexible exchange rate system is
in operation.
i
LM1
LM1
LM2
i1
i*
LM2
BP
IS2
IS1
y1
y2 y3
i1
IS
BP
i*
y1
y2
are higher, i.e., at i *). The BOP thus moves into deficit. The money supply will decrease
(as people exchange domestic currency for foreign currency, thus eroding the foreign
exchange reserves of the central bank). The LM curve thus shifts backwards and hence
erodes the intended expansion.
In short, when exchange rates are fixed, monetary policy is ineffective. The central
bank could try and counteract the contraction of the money supply (resulting from the
decline in reserves) by buying bonds from the public. The central banks holdings of
government bonds and foreign exchange will fall by equal amountsthus leaving the
money supply unaltered. Such sterilisation of the BOP deficit is, however, not sustainable and can last only as long as foreign exchange reserves are positive.
183
184
Macroeconomics Simplified
LM1
i
LM
LM2
BP
i*
i*
BP
IS2
IS2
IS1
y1 y2
A. The Effect of Expansionary
Fiscal Policy
IS1
y
y1
y2
Given that prices are assumed constant, this nominal appreciation is a real appreciation.
It thus results in an increased demand for imports (and a drop in exports). The IS curve
will thus shift backwards. This process will continue until the interest rate is brought
back into line with the world interest rate i * (i.e., at the intersection of LM1 and IS1).
Now consider the case of an expansionary monetary policy that shifts the LM curve
to LM2 in Figure 9.13(B). This lowers the interest rate and hence encourages capital
flight out of the domestic bond market. As the supply of domestic currency increases on
the foreign-exchange market (as people demand more foreign exchange in order to buy
bonds in other countries where the yield is higher), the exchange rate will depreciate.
This will dampen the demand for imports and increase the competitiveness of exports.
The IS curve thus shifts outwards to IS2. Thus, both the IS and LM curves respond to a
monetary stimulus and output rises to y2.
This brings us to the concept of real exchange rate. The real exchange rate (REX) is
equal to the average nominal exchange rate between a country and its trading partners,
with an adjustment for difference in inflation rates between that country and its trading
partners.
REX = e . P*/P,(9.6)
or (approximately)
REX = e + P* P,(9.7)
where P* = world price index (i.e., index of prices of the countrys trading partners);
P=domestic price index; = change in.
Thus, if world inflation was 5 per cent and domestic inflation was 10 per cent, then a
20 per cent nominal depreciation would approximately result in a 15 per cent real depreciation (20% + 5% 10%).
The more competitive the foreign trade sector, the closer the relationship is likely to
be between nominal exchange rate movements and changes in domestic prices (relative
to world prices). In a perfectly competitive world market where consumers have easy
access to products produced in other countries, the prices of traded goods should be the
same everywhere (after adjustment for customs duties and transport costs). If so, then
purchasing power parity (PPP) would hold, and nominal exchange rates would adjust
quickly in response to changes in relative prices. Under such conditions, real exchange
rate changes are impossible.
185
186
Macroeconomics Simplified
This implies that all goods will cost the same when expressed in a common currency
and that the nominal exchange rate will be equal to the ratio of domestic to international
prices.
This theory of absolute PPP underpins The Economists tongue-in-cheek Hamburger
Standard evaluation of exchange rates. Using the McDonalds Big Mac as a proxy for an
international good sold everywhere, The Economist evaluates the dollar exchange rates
of various countries to see whether they do indeed equal P/P* (see box below).
(9.10)
In other words, relative PPP holds when percentage change in the exchange rate is
equal to inflation differential between the domestic and foreign countries. The implication is that countries with relatively high inflation rates will have depreciating exchange
rates (and vice versa). Because we are dealing with percentage changes rather than levels,
this is not as strong an assumption as the absolute version of PPP theory. Note that if
absolute PPP holds, then relative PPP will also hold (and relative PPP may hold even
when absolute PPP does not). This is because the level of e may not equal P/P*, but the
change in e could still equal the inflation differential. If exchange rates are fixed, then
price levels will remain equalised (absolute version) or inflation rates will be equalised
(relative version) throughout the world.
A key implication of PPP theory is that real exchange rate movements are impossible
or at least cannot be sustained. Any nominal depreciation will result in increased demand
for domestic goods, and hence domestic prices will rise until the previous equilibrium real
exchange rate is restored.
There are various reasons why PPP theory does not hold very well. Price adjustment (as Keynesians would argue) may not be as quick or as efficient as assumed by the
theory. Furthermore, measures of inflation tend to include a wide range of goodsnot
all of which are traded internationally. This is true, for example, with regard to building construction and retail services. Thus, in the case of the Big Mac, the meat, bread
and potatoes may be traded goods (and hence subject to international arbitrage), but
the retail costs of selling the hamburgers (e.g., rent and labour costs) are likely to differ
between countries.
There are, nevertheless, several insights to be drawn from PPP theory. In particular,
it predicts that a country that allows its domestic inflation rate to exceed the world rate
(i.e., experiences an appreciating real exchange rate) is likely to suffer from a nominal
exchange rate depreciation (as the demand for its currency falls). This is not a bad rule
of thumb to start with. But bear in mind that it will not always result in a correct prediction because the demand for, and supply of, foreign exchange depends on more factors
than merely the ratio of domestic and foreign price indices. For example, the US dollar
appreciated in the 1990s because of strong capital inflows (attracted in part by a strong
economy and by the invention of new products such as internet software and projects)
and not because of changing relative inflation rates.
Policy Implications
In practice, most economies operate somewhere between a fixed and a flexible system,
and most experience price adjustments (but not to the extremes suggested by PPP
theory). Thus, one is most likely to be using a hybrid ISLMBP framework when
approaching a real-world case.
For example, consider the following possible real-world scenario. An economy is at
less than full employment and the monetary authorities pursue an expansionary monetary policy. This leads to a drop in interest rates, and hence some capital leaves the
country. The central bank intervenes to cushion the impact on exchange rate, but allows
the nominal exchange rate to depreciate to some extent. Assume also that prices rise, but
not enough to prevent a real exchange rate depreciation.
This story could be incorporated into an adapted ISLMBP framework (Figure 9.14)
as follows: The expansionary monetary policy pushes LM1 to LM2 as the nominal money
supply rises from M1 to M2. The drop in interest rates sparks some capital flightand
hence boosts demand for foreign currency. The central bank supplies some, but not all,
of this increased demand out of its reserves, thus contracting the nominal money supply
(to M3).
Given that the central bank does not provide enough reserves to cover the increased
demand, the LM curve does not contract all the way back to LM1 but instead comes to
187
188
Macroeconomics Simplified
y1
y2
rest at LM3. Instead, the exchange rate depreciates (from e1 to e2). Prices rise (from P1 to
P2), but not enough to prevent the real exchange rate from depreciating (i.e., e > P).
Exports rise and imports fall, thus resulting in an outward shift in IS to IS2. The increase
in price erodes the real money supply, thus pushing the LM curve to LM4. Notice that
output rises from y1 to y2 which is more than would be predicted using an extreme neoclassical model (assuming full employment and PPP), but less than would be predicted
using an extreme Keynesian model (assuming constant prices).
convert their yield payments at a different exchange rate, then domestic interest rates
have to reflect that expectation.
Consider a situation where e = 1 and interest rates in country A and country B are
both 10 per cent. If the exchange rate is expected to remain the same, people will be
indifferent as to where they buy bonds. But what happens if country As currency is
expected to depreciate by 5 per cent over the period between buying bonds and receiving the yield payments? The yield payments would thus be worth 5 per cent less in terms
of foreign currency than they would be if the bond had been purchased in country B
where the exchange rate was expected to remain constant. Under such conditions, bonds
would only be purchased in country A if the yield at the time of purchase was sufficiently
high to compensate for the expected exchange rate depreciation.
The implication is that nominal domestic interest rates (adjusted if necessary for
country-specific risk) should equal (approximately) the sum of the nominal interest rate
abroad, i *, plus the anticipated depreciation of the domestic currency (x).
i = i * + x. (9.11)
Equation 9.11 is known as the uncovered interest parity condition. It can be transformed into
i i * = x. (9.12)
Equation 9.12 suggests that the interest rate differential among comparable assets
denominated in different currencies can be interpreted as representing anticipated
exchange rate movements. One would thus expect countries with depreciating currencies (and high inflation rates) to have comparatively high nominal interest rates (as in
Latin America), and vice versa.
189
190
Macroeconomics Simplified
unanticipated, which explains why it is so difficult to predict exchange rate movements accurately.
For an excellent discussion of the causes of the 2008 financial crisis and its implications for India, see
Reserve Bank of India. 2010. Report on Currency and Finance 200809: Global Financial Crisis and The
Indian Economy.
5
15%
Real money supply
(M1)
10%
Real GDP
5%
Inflation (GDP
deflator)
0%
Nominal interest
rate
201112
201011
200910
200809
200708
200607
200506
200405
200304
200203
10%
200102
5%
Real effective
exchange rate
Real interest rate
15%
191
192
Macroeconomics Simplified
(Box Continued)
the notion that trade liberalisation often led to increased growth. Research had
shown that share of the least developing countries in world trade had actually
been declining, he said.
Source: Authors.
We could think about the situation facing India in ISLMBP terms as follows: the
sudden drop in confidence and flight to safety by bondholders (a process exacerbated
by the global liquidity crisis that also sparked mass sell-offs of financial assets) resulted
in the risk premium for Indian bonds rising (the BP curve shifting upwards). The resulting depreciation did not boost exports (though it probably prevented a greater fall).
Indian fiscal policy injected demand into the economy as deficit rose. We can represent this in Figure 9.16 as an outward shift in the IS curve. The growth in the money
supply shown in Figure 9.15 would also have injected demand into the economy (i.e., the
LM curve shifted out). However, after 2010, tighter monetary policy resulted in slowing
inflation, rising real interest rates and a fall in the real money supply.
LM1
i
LM2
BP2
BP1
IS2
IS1
y1
y2
CHAPTER 10
The 2008 global economic crisis highlighted a key feature of capitalism which mainstream economic theory pays limited attention to: its highly unstable character and proclivity for financial crises. Indeed, the neoclassical approach, which holds that financial
markets are efficient (prices on traded assets are assumed to reflect all known information) and that rational expectations prevail (outcomes do not differ significantly from
what they are expected to be), is totally ill-equipped to deal with financial crises. This is
a serious problem because speculative bubbles and busts are common under capitalism
and their effects are harmful. As Keynes warned us long ago:
Speculators may do no harm as bubbles on a steady stream of enterprise. But the position
is serious when enterprise becomes the bubble on a whirlpool of speculation. When the
capital development of a country becomes the by-product of the activities of a casino, the
job is likely to be ill-done.1
In their thorough analysis of financial crises over the past eight centuries, Carmen
Reinhart and Kenneth Rogoff show that financial crises are a regular occurrence caused
by overconfidence, excessive debt accumulation and a lack of appreciation of rising systemic risk.2 They argue that policy makers and investors ignore the lessons of history,
believing erroneously that this time is different. This illusion is costly as financial crises
reverberate through the real economy, causing extended recessions. Their central theme
is that:
Excessive debt accumulation, whether it be by the government, banks, corporations, or
consumers, often poses greater systemic risks than it seems during a boom. Infusions of
cash can make a government look like it is providing greater growth to its economy than
it really is. Private sector borrowing binges can inflate housing and stock prices far beyond
their long-run sustainable levels and make banks seem more stable and profitable than
they really are. Such large-scale debt build-ups pose risks because they make an economy
vulnerable to crises of confidence, particularly when debt is short term and needs to be
1
Keynes, J. M. 2006 [1936]. The General Theory of Employment, Interest and Money. p. 142. New Delhi:
Atlantic Publishers.
2
Reinhart, C. and K. Rogoff. 2009. This Time is Different: Eight Centuries of Financial Folly. Princeton
and Oxford: Princeton University Press.
194
Macroeconomics Simplified
constantly refinanced. Debt-fuelled booms all too often provide false affirmation of a governments polies, a financial institutions ability to make outsized profits, or a countrys
standard of living. Most of these booms end badly.3
Reinhart and Rogoff highlighting the role of credit as a source of great instability.
This aspect of monetary economy is not accommodated within the standard ISLMBP
and ASAD models (which effectively boil the financial sector down to a bond market).
To explore the role of credit and debt as a source of stability internal to capitalism itself,
we need to turn to Joseph Schumpeter, who built it into his theory of economic cycles,
and his student Hyman Minsky who developed the ideas further through his concept of
financial fragility.
Joseph Schumpeter
Writing in the first half of the 20th century, Schumpeter was critical of the economic
mainstream for its lack of attention to profitability and the forces behind the continued
Ibid., p. xxv.
Wallerstein, I. 2008. The Depression: A Long-Term View. Commentary no. 243, 15 October 2008.
http://www.sendika.org/yazi.php?yazi_no=20228
3
4
Hyman Minsky
Minsky believed that as long as an economy is capitalist, it will be financially unstable,8
hence his financial instability hypothesis:
In contrast to the orthodox Quantity Theory of Money, the financial instability hypothesis
takes banking seriously as a profit-seeking activity. Banks seek profits by financing activity
5
Schumpeter, J. 1961 [1911]. The Theory of Economic Development. pp. 3032. Oxford: Oxford University Press.
6
Schumpeter, J. 1994 [1954]. A History of Economic Analysis. London: Routledge.
7
Ingham, G. 2008. Capitalism. p. 39. Malden: Polity Press.
8
Minsky, H. 1982. The Financial Instability Hypothesis.. Financial Crises, eds. C. P. Kindleberger and
J.P. Lafarge. p. 36. Cambridge: Cambridge University Press.
195
196
Macroeconomics Simplified
and bankers. Like all entrepreneurs in a capitalist economy, bankers are aware that innovation assures profits. Thus bankers (using the term generically for all intermediaries in
finance), whether they be brokers or dealers, are merchants of debt who strive to innovate
in the assets they acquire and the liabilities they market. This innovative characteristic of
banking and finance invalidates the fundamental presupposition of the orthodox Quantity
Theory of Money to the effect that there is an unchanging money item whose velocity of
circulation is sufficiently close to being constant....9
the economy transits from financial relations that make for a stable system to financial
relations that make for an unstable system.12
Schumpeter and Minsky emphasised that because capitalist enterprise is debtfinanced, crises are normal functioning events. The typical cycle begins with the expansion of production and revenues of the prudent hedge finance enterprises. The growing
cash balances in their bank accounts enable banks to lend money to enterprises which
fall into the speculative and Ponzi categories. As the level of unserviceable debt rises,
the probability of a chain reaction of defaults increases (or the monetary authorities
might try to curb lending by raising interest rateswhich could also trigger defaults).
So, assets are sold to realise cash, which sparks a collapse in asset prices, also making a
financial crisis likely. This translates directly into the real economy as firms try to cut
losses by reducing output and employment, thus collapsing demand.
Minsky saw this as a problem, but not a terminal one in that governments and central
banks could (at least hypothetically) become more effective at monitoring and so prevent financial fragility from translating into crisis. Reinhart and Rogoff, echo this view,
arguing also for better and more transparent monitoring of the level of national government debt.13 Central banks can prevent a total crisis by advancing loans to the banking
system (to prevent the unravelling of the credit network and payments system). But the
obvious danger here is moral hazard as banks feel insured because they are seen as too
big to fail. Under such conditions, money managers have every incentive to continue
with risky financial innovations.
197
198
Macroeconomics Simplified
argued that the rise of fascism in Europe was a response to insufficiently regulated and
crisis-ridden economies).
Polanyis analysis certainly resonates today as citizens around the world put pressure
on governments to restrain the activities of money managers (the banksters) and to
increase taxes on the rich. There is also growing interest in the ways in which governments can intervene more aggressively in economies, both with regard to investment
and welfare spending.
Capitalism has come under a great deal of criticism since the 2008 crisis, but it is not
about to disappear anytime soon. Rather the challenge facing policy makers is to manage the capitalist economy so that it delivers more benefits, and fewer costs, for ordinary
people. The old tension between neoclassical and Keynesian ways of thinking will
continuebut this is a creative tension and can help us as we think through what policies
make sense in particular contexts.
INDEX
Consumption, 7
In India, 810
Keynesian consumption function, 5051
Other theories of consumption, 5152, 54
Real balance effect on consumption, 146
Coupon rate, See bonds
Credit default swaps, 64
Distribution and demand, 13335
Easterlin, Richard, 16
Easterlin paradox, 1718
Ex ante and ex post, 12, 49, 82, 116
Economic growth, 15566
Demand- and supply-side policies, 155
Endogenous growth theory, 155
In India, 9, 13, 13738, 165, 17374, 19092
Low and high roads to growth, 16466
Sustainable growth, 15556
Effective demand, 4546, 60, 143
Efficiency wages, 59
Endogenous and exogenous, 5051
Equations (reduced form and structural), 52, 53
European monetary union, 172
Exchange rate, 17074
Economic policy under fixed exchange rates,
18283
Economic policy under flexible exchange rates,
18384
Real exchange rate, 18485
Expectations
Adaptive expectations, 16162
Expectations and exchange rates, 189
Naive expectations, 15860
Rational expectations, 16263, 193
200
Macroeconomics Simplified
Expenditure on the GDP, 67
Expenditure reducing and expenditure
switching, 173
In India, 910, 13738
See also effective demand, aggregate demand
Index
Open-economy ISLMBP, 17692
Possible dynamic adjustment paths, 12225
Pragmatic approach towards, 13133
Japan, 61, 100, 131, 171
Jorgenson, Dale, 67
Keynesian economics, 12, 12, 19, 31, 33, 38,
4061, See also Keynes, ISLM model, monetary
policy, fiscal policy, neoclassical synthesis, New
Keynesian macroeconomics
Accidental Keynesianism in India, 138
Closed economy one sector Keynesian Cross
model, 4958, 11213, 15354
Consumption function, 5052, 5455, 84, 134
Dynamic Keynesian models, 15863
Equilibrium and adjustment in the money
market, 101, 10306, 11011, 14850
Equilibrium conditions, 12, 49, 52, 5556,
8183, 9091, 148, 177
Hydraulic and fundamentalist Keynesians, 58
Inflation, 143, 187, See also Phillips curve,
dynamic Keynesian models
ISLM Keynesianism or bastard Keynesianism,
7980
Keynes effect (effect of prices on output), 145
Keynesian aggregate supply curve with wages
rigid downwards (Keynesian economics as a
special case), 140, 14243
Keynesian demand management, 57, 61, 12021,
12425, 136, See also fiscal policy, monetary
policy
Liquidity preference and the demand for money,
97101, 10708
Liquidity trap, 100
Multiplier, See multiplier
New Keynesian macroeconomics, 2, 41, 43
Open economy one sector Keynesian Cross;
model, 17476
Planned and unplanned investment, 49, See also
ex ante and ex post
Post-Keynesians, 58, 60
Quantity (or output) adjustment in the
goods market, 4849, 83, 9597, 11011,
14851
Savings function, 8485, 91
Transmission mechanism (of money supply on
output), 122
Versus the classics, 43 (footnote 7), 4445
201
202
Macroeconomics Simplified
Monetary policy, 36, 10607, 113, 118
In a closed economy Keynesian ISLM model,
12122, 158
In a closed-economy Neoclassical ISLM model,
12931, 158
In an open economy model under fixed
exchange rates, 18283
In an open economy model under flexible
exchange rates, 18384
Keynesian transmission mechanism, 122
Open market operations, 10607, 121
Printing money to finance government
expenditure (monetising the deficit or
quantitative easing), 106, 113, 118
Moral hazard, 197
Multiplier, 53, 54, 55
And government economic policy, 57, 9293, 121
Balanced budget multiplier, 9293, 119
In Indias Mahatma Gandhi National Rural
Employment Guarantee Act, 5455
In an open-economy context, 183
In one-sector, two-sector and three-sector
models, 113, 14950
In underdeveloped economies, 55
Larger for bond-financed than tax financed
government expenditure, 119
NAIRU, 16364
In India, 16364
National accounts, 518
and welfare, 1516
green accounting, 15
Identities, 1112
Not to be confused with equilibrium conditions,
33, 82
Real and nominal values, 1213.
Neoclassical economics, 1, 2, 3, 1938, 52,
See also ISLM model, fiscal policy, monetary
policy, neoclassical synthesis, new classical
economics
Aggregate supply curve, 2631, 141
As an impossible Utopia, 197
Full-employment outcome, 2634, 3637
Keynes critique of, 4245
Labour market policy, 152
Microfoundations, 1920, 3637, 41
Neutrality of money, 21
Supply and demand for loanable funds, 3134,
82, 85
Index
Ttonnement, 20
Taylor, John (and the Taylor rule), 124
Thatcher, Margaret, 152
Tobin, James, 16, 60, 80, 133
Tobins q, 67
Uncovered interest parity condition, 189
Unemployment, 2526, See also labour market,
Phillips curve, neoclassical synthesis
Involuntary unemployment, 2526, 46, 55
Ura, Karma, 17
Value-added, 5
Varieties of Capitalism, 16465
Walras, Lon, 20, 40
Walrasian processes, 36, 40
Wallerstein, Immanuel, 194
Yields, See bonds
Yield curve, 7276
Expectations theory of, 7275
In India, 7576
203
Nicoli Nattrass is Professor of Economics and Director of the AIDS and Society
Research Unit at the University of Cape Town, and a regular visiting professor at Yale
where she teaches courses on development and the AIDS epidemic. She has published
widely on macroeconomic policy, inequality, AIDS, and the struggle for antiretroviral
treatment. Her work on the economics of preventing the transmission of HIV from
mother to child was used in a successful court case launched by South Africas Treatment Action Campaign against the then government of Thabo Mbeki. In her subsequent research on the role of ideas in shaping South Africas response to AIDS, she
focused on AIDS denialist and AIDS conspiracy beliefs.
G. Visakh Varma is an Associate Professor, Department of Economics, KKTM Government College, University of Calicut, Kerala. He teaches macroeconomics and his
research interests are in the field of labour economics and macroeconomics. He is a
member of the governing body of the Centre for Socio-economic and Environmental
Studies, Kochi.
Professor Varma has published articles in Economic and Political Weekly (September
2008 and January 2012) on Nurksian contributions and macroeconomics curriculum.
He co-authored the book, An Economic Approach to Social Interactions (2007, Ernakulam)
with Professor Martin Patrick, the preface of the book being contributed by 1992 Nobel
Prize winner in Economics Professor Gary Becker.