A M A R K E T THEORY OF MONEY

This page intentionally left blank

A MARKET THEORY
OF MONEY

JOHN HICKS

C L A R E N D O N PRESS . O X F O R D
 This book has been printed digitally and produced in a standard specification
in order to ensure its continuing availabitity

OXFORD
U N I V E R S I T Y PRESS
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of Oxford.
It furthers the University's objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan South Korea Poland Portugal
Singapore Switzerland Thailand Turkey Ukraine Vietnam

Oxford is a registered trade mark of Oxford University Press

in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
©John Hicks 1989
The moral rights of the author have been asserted
Database right Oxford University Press (maker)
Reprinted 2007
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
And yon must impose this same condition on any acquirer

ISBN 978-0-19-828724-7
 CONTENTS

Introduction I
Part I
THE W O R K I N G OF MARKETS

1 Supply and Demand ? 7

Pre-Keynesian theories. Jevons and Walras, Edgeworfh
and Marshall, 'Corn* and "Fish',
2 The Function of Speculation 12
A seasonal market. Two senses of 'equilibrium'. Harvest
variations. Futures. Physical carry-over does the
smoothing.
3 The Pricing of Manufactures 19
Primary and secondary merchants. The secondary
market in Marshall. Imperfect competition in secondary
markets. The Introduction of new products. Advertise-
ment. 'Fixprice' {announced price) markets.
4 The Labour Market 27
What happens to the unemployed? Family labour and
the growth of a wage system. Country to town. Estab-
lished labour; potential competition. Collective bargain-
ing: the brokering function. Defensive character of trade
unions. Money wages and real wages. Influence on
Keynes of British experience. S-unemployment and F-
unemployment.

Part II
MONEY AND FINANCE

5 The Nature of Money 41

The representative transaction. Its three parts: contract,
two deliveries. Money appears in contract as standard of
VI CONTENTS

value, in delivery as means of payment. The former is

primary. Store of value no distinguishing characteristic,
International barter deals. Price-lists and the justum
pretium. Coinage.
6 The Market Makes its Money 47
Offsetting of debts, Bills of exchange. A market in bills.
Discounting of bills. Appearance of interest. Outside
borrowing and lending. Government borrowing,
7 Banks and Bank Money 55
Banks defined by activities: accepting deposits, discount-
ing bills, making advances. Why these come together.
Bank money as means of payment. Notes as transferable
receipts. Payment by cheque a safety device. The 'cre-
ation' of bank money, exposes bank to risk. Protections
against these risks: 'law of large numbers', liquid re-
serves, provision for borrowing from another bank.
Liquidity. As defined in Keynes's Treatise, a quality
attributed to an asset. Running assets and reserve
assets. From the Genera! Theory to monetarism.
8 Choice among Assets 64
General theory of liquidity. The spectrum of assets. A
six-way classification. Invisible assets (and liabilities).
Keynes's speculative motive. Solid and fluid. Trans-
action costs. Intermediation, Perfect fluidity approached
but not achieved.
9 Theories of Interest; Keynes versus Marshall 72
Keynes's classic was Marshall. Stocks and flows. A
simple model in which each finds a place. Which margin
is the more important is an empirical question, which
may be answered differently at different times, The situ-
ation of Marshall: the Goschen conversion. The situ-
ation of Keynes; an already achieved conversion. Our
situation.
10 Markets in Equities: Ownership and Control 80
Nucleus and fringe. The private company; limited lia-
bility without full transferability. The importance of
 CONTENTS vii

transferability. Shareholder and bondholder. Why buy

shares? Why issue shares? Why pay dividends? Take-
overs and mergers. Uses and misuses,

Part III
PROBLEMS AND POLICIES

11 The Old Trade Cycle 93

Nineteenth-century cycles not statistical cycles but a
succession of crises, readily understood as of monetary
origin. Mill. The Bank of England as Central Bank of a
sterling area, extending far outside UK. Gold Standard
provided a ceiling. The problem of floor: the Thornton
precept. Cycle began to look controllable. At end of cen-
tury, leadership began to pass to US, which had no Cen-
tral" Bank. The Ted' in 1920 and in 1930-1. Why the
Great Depression ended the old cycle.
12 The Credit Economy: Wicksell 102
Non-interest-bearing money ceasing to be a reserve
asset; no more than petty cash. Monetarism no short-
cut. An economy in which all money is credit is coming
into sight. Wicksell's model of such an economy. A
single bank with the rate of interest at which it lends
and borrows its sole means of control. Revisions of the
Wicksell model. In-rate and out-rate. An arm's length
system. Defects could be mended by closer co-operation.
13 Interest and Investment 112
Hawtrey and Keynes. What is left of the Hawtrey
channel. The Keynes channel. Building of dwelling-
houses. Liquidity of builder—and buyer. Industrial
investment, defensive or innovative. Defensive invest-
ment insensitive to interest. Types of innovative invest-
ment. Running-cost reducing investment most sensitive
to interest. The Ricardo machinery effect,
14 An International Economy 121
A theoretical model. The market finds its centre. Centre
and non-centre (Centralia and Penland), Remedies for
viii CONTENTS

weakness of a Penland currency generally dependent on

finding some underlying strength. Help from inter-
national agencies; uses and limitations. Either strength
or weakness of central currency may set problems for
others. The Gold Standard. Not inconsistent with some
national currency being central, as sterling was before
1914. Epilogue 1914-33. General relations between
strength and centrality. The current dilemma,
15 What is Bad about Inflation? 132
The traditional answer (given by Robertson) can be
stood on its head. The classes of markets (part I of this
book). When prices (or wages) have to be negotiated it is
a time-consuming process, so the effects of inflation are
subject to lags; they cannot move together. Almost
everyone at some stage in the process feels that he is get-
ting left behind. So inflation damages democratic
government.
Real causes of inflation. Removal of real causes neces-
sary for solution; the strong case of hyper-inflation, If
substantial external trade is to be maintained, external
stability comes first.
Appendix: Risk and Uncertainty 137
An exercise directed to clarifying the issue. The Ber-
noulli assumption. Cardinal probability and cardinal
utility. An MU curve with constant elasticity. More
interesting case sets u(0} = minus infinity: a finite
chance of such an outcome is always to be avoided. If
this disaster point is set at zero, distribution of portfolio
unaffected by capital value. If it is set at a higher level,
operator will take more risks when he is better able to
afford them, A four-way classification needed; measur-
able and unmeasurable uncertainties, risky and not.

This rather detailed table of contents has been devised to take the place of an
index, which for a book of this character proves to be inappropriate.
 Introduction

Once upon a time I was giving a lecture, based upon my Theory of

Economic History. When it was over, I was asked: are you a follower
of Marx, or of Adam Smith? I am glad that I replied, without hesi-
tation; both! Certainly I had learned something from each of them.
Here, if I were asked a similar question—are you, or are you not, a
Keynesian?—I should want to give a similar answer. For although,
as will be amply shown in the following pages, I owe a great deal to
Keynes, I also owe much to some of his predecessors, whom he
thought to have made back-numbers; and also to some later writers,
who were none of his 'school'. My own writings on money do indeed
go back to the days when his were innovations, I was one of those
who had to be converted. Perhaps, I have now come to think, I
allowed myself too much to be converted. I already had some of the
means to preserve a greater degree of independence, as I think will
here be shown.
What was the essence of the 'Keynesian revolution'? I would now
state it in the following way, It had been a common assumption of
his predecessors that the economy under study had a 'long-term
equilibrium" about which it would indeed fluctuate, but the fluctua-
tions would be limited and by wise policy their amplitude could be
damped, I think I can show that this was in their day a defensible
position; in the days of the old Gold Standard it made a good deal of
sense.1 By the time Keynes was writing his General Theory that stand-
ard was being abandoned; by his 'persuasions' he had contributed to
its abandonment, especially to the abandonment of its old authority;
he had no desire to go back to anything so rigid, so firm. Thus the
only equilibrium which survives in his theory is a short-term equilib-
rium, with no sheet-anchor to hold it.
It was not easy for those who were in my position, in 1936-7, to
accept this abandonment. So the version of Keynes which we
received into our own thinking was provided with another anchor, a
1
As i shall be explaining in Chapter 11. Thus I would not want to go so far in
attacking it as Kaldor did in his last book Economics without Equiltbrium, though there
is much that he said with which 1 am in sympathy. He sent me a copy of that book,
inscribed 'as a token of a life-long friendship'.
This page intentionally left blank
 INTRODUCTION 3

There is nevertheless another difference between Keynesian and

pre-Keynesian theories which is not attended to in these chapters. As
a consequence of abandoning their long-term equilibrium he con-
fined attention to the short-term consequences of changes, such as
policy changes. His predecessors had not been very good at that. Still
the Keynes theory suffers from its inability, or disinclination, to look
further ahead. A main way in which this shows up is in its concen-
tration of attention on the employment-creating power of investment
expenditure, while it is occurring. This will be much the same whether
the expenditure is productive or unproductive. His 'investment* is
just the use of current resources for purposes other than furthering
the present or near-future output of consumption goods (or services).
It will be just the same if it takes the form of building power-houses or
of building palaces for kings. Its effect on productivity in the longer
run is left aside. That does need to be remembered, even if we abstain
from thinking in terms of long-run equilibrium (I shall come again to
this matter in Chapter 1 3).
I began to write this book in 1985, soon after the death of my wife,
who had been my fellow-worker for fifty years. It has been written in
monastic seclusion, without the benefit of continual discussion. It is
true that I have sometimes been able to get to meetings, sometimes
far away, when I could get friends to help me. There are a few par-
ticular points which I owe to people I have met at those meetings;
they are acknowledged as they come up, There are just three general
acknowledgements I want to make. The first is to Axel Leijonhufvud,
who started me off on this undertaking, convincing me that what I
said on earlier occasions was not enough. The others are to Anthony
Courakis and to Stefano Zamagni. without whose support and
encouragement I could not have finished the book. None of them has
any responsibility for what has finally emerged.
I am well aware that there are many articles which have appeared
in journals, during the years while I have been writing, and earlier,
which are relevant to what I am saying: 1 regret that in my circum-
stances I have not: had much access to them, So I must leave it to
their authors to discuss the relations between us—the other way
round,
So for whom am I writing this book? Not for economic advisers,
and commentators on current, affairs, who have had opportunity to
learn from experience much of the substance of what I am saying.
4 INTRODUCTION

More for teachers of economics; I hope I shall help them to match

what they say on one day with what they say on another. And for
their criticially minded students, some of whom will be the teachers
of the next generation.
 PART I

THE W O R K I N G OF MARKETS
This page intentionally left blank
 1 Supply and Demand ?

The theories about price-formation in competitive markets, that were

available to economists at the time when Keynes was writing, had
been the work of the so-called 'neo-classics' between 1870 and
1900. There were several of these and they did not fit together very
well. All however accepted the distinction that had come down from
Adam Smith, between market value and 'natural' or normal value,
natural value depending on cost of production, market value on
supply and demand. Market value would 'tend' towards natural
value by adjustment of supply. It was accordingly held, for nearly a
century after Smith, that natural values were the only values that
required attention. The whole of Ricardo's system, to take the most
important example, runs in terms of natural values.1 The chief thing
which happened at the 'marginal revolution' of Jevons and his con-
temporaries was a shift of attention to market values. They were
determined, it was accepted, by supply and demand; but how? Just
how did the market work?
Though Jevons (1871) saw the problem, he failed to solve it. He
gave nothing better than a rather simple-minded mathematician's
answer: that if the article being sold was of uniform quality, there
could not be more than one price in the market, so the price must be
that at which the last (or marginal) unit would sell (his 'law of in-
difference'). As is obvious to the modern student, this implied that the
market was always in equilibrium. But how did it get into such an
equilibrium? Jevons gave no help.
Walras (18 74), writing without knowledge of Jevons, met the
same problem; he also gave a mathematician's answer, but his was
more subtle. The price must be established at the level where curves
showing demand and supply, as functions of price, intersect. But

' It may be objected that in his rent theory, Ricardo had the price of 'corn' deter-
mined at the margin, and so was dependent on demand. Nevertheless it is a normal
price which is taken to be so determined: otherwise how did he fail to mention the
weather? Ricardo is showing that in the case of an agricultural product, not market
value only but also normal value Is dependent on demand. The most Ricardian among
modern economists was his editor, Picro Srafla. It is significant that Sraffa's own
theory (Production of Commodities by means of Commodities 1960) runs entirely in terms
of normal values.
8 THE W O R K I N G OF MARKETS

how are these curves to be found before there is any trading? If the
equilibrium has not been found before there is any trading, much or
most of the trading must have been conducted at non-equilibrium
prices, so the average price over the day may be far from the equilib-
rium price; even the final price, at the end of the day, may be out of
equilibrium, Walras's answer to the puzzle was to suppose all parties
trading to disclose their propensities before trading started. The
'market organizer' (as I prefer to call him, for a reason which I give
below2), when he had this information reported to him. could calcu-
late the equilibrium price, and at that price actual trading could
proceed.
Though it must be accepted that it is possible for a market to be
organized in this manner (by some preliminary agreement between
the parties trading) this particular form of organization is not one
which at all commonly occurs. Neither Edgeworth (1882) nor
Marshall (.1890) were satisfied with Walras's answer, of which they
were aware. Each of them supposed that information was collected
in the course of trading.
What takes the place of Walras's organizer, in the market as ana-
lysed by Edgeworth, is ability to recontract. All contracts for sale are
provisional. That is enough (when there is no monopoly on either
side of the market) to show that a uniform price must be established;
for if there were no such uniformity, a buyer who had bought at a
high price could repudiate his contract, finding a seller who had sold
at a lower price, to whom he could offer more favourable terms.
Further, if at the uniform price thus established there were buyers
who had not exhausted their demands at that price, or at any price in
its neighbourhood, they could find sellers who had sold at the estab-
lished price but who could be tempted away by an offer a bit better.
Thus the market would come to an equilibrium with demand and
supply, at the end, equated.
This solution of Edgeworth's was a great step forward; but it was
unfortunate that the illustration he gave, with which to explain it,
was not well chosen. This ran in terms of a labour market; it must be
2
It should be noticed that the Walras market is not an auction market, with which
it has often been confounded. Most auction markets are for second-hand commodities,
such us pieces of old furniture or houses, of which each unit is to some extent unique,
so that Jevons's law of indifference docs not apply. The auctions which are nearest to
the Walras model are those sometimes used for new issues on the stock exchange. But
even in this case the auctioneer is an agent for the seller; he is not the independent
organizer postulated by Walras,
 SUPPLY AND DEMAND? 9
granted that In that application it does not make much sense, (I shall
be coming to the labour market later.) There are other markets,
which do exist, where it works much better. All that is needed to
make it realistic is to introduce intermediary traders, neither final
buyers nor sellers, who on occasion may either buy or sell. So they
are readily able to reverse their contracts. Arbitrage, which is pre-
cisely the kind of transaction on which Edgeworth relies to establish
his uniformity, is common enough in practice for a name to have
been given to it.
Edgeworth knew very well that the 'equilibrium' which is estab-
lished in his way at the "end" of the market need not be the same as
that which would have been established in Walras's manner. For
willingness to trade at the 'end* could well be affected by gains and
losses due to non-equilibrium trading on the way. This is the point
that was taken up by Marshall. In substance he accepted Edge-
worth's analysis;3 there was just one point, to which he seems to
have attached considerable importance, which he added to it. If it
could be shown that the gains and losses which will have attended
non-equilibrium trading "on the way' are unlikely to have much
effect on the willingness of buyers to purchase an additional unit,
will not the market finish up at an equilibrium price in the sense of
Walras, after all? If these gains and losses amount to no more than a
small part of the buyers' total expenditure, the 'income effect' (as we
should now say)4 of allowing for them should not be considerable. So
the market should finish up at something quite near to a Walras
equilibrium,
Marshall was very attached to this proposition of his; but it is not
as helpful as he supposed. When he tries to write it out in a realistic
manner, as was his custom, this shows up. He interprets it as the be-
haviour of a 'corn market in a country town'.5 Various realistic char-
acteristics of such a market are allowed to slip in; most do not matter,
but there is one that is of fundamental importance. He had admitted
among the characters of his story the farmers who bring the corn to
market and the consumers (or millers?) who take it off, also mer-
chants who act as intermediaries; all that, as we have seen, is as it
should be. But it also allows his merchants to carry over surplus
* Principles, mathematical appendix, note XII,
* In Marshall's terminology, the 'marginal utility of money' would be substantially
unaffected.
5
Part V, Chapter 2.
IO THE W O R K I N G OF MARKETS

stocks at the end of the day; and if at the end, why not at the begin-
ning? What then is the significance of the 'end of the day*? A market
in which carry-over is permitted is a continuing market: it does not
'finish up'.
When in the 1920s at Cambridge a new generation of teachers
were set to lecture on Marshall potting into their own words what
he had said, this was one of the things that troubled them. If
Marshall's proposition was to be used in the way he had used it,
there could be no carry-over; so it would be safer to make the article
traded perishable—"fish* not 'corn'.6 That is formally correct, but it
greatly reduces the scope of the theory. The Marshall market
becomes a very special type of market, a little better but not much
better than the artificial market of Walras.
What was then to be done? The right thing, surely, would have
been to go on to construct a formal theory of the market for a non-
perishable product: that indeed would have turned a corner. One
could still have followed Marshall and admitted intermediary traders;
and also have followed him in supposing that these were the people
who held the stocks. Since they would have been willing to come in
either as buyers or as sellers, there would be an 'inside' market which
would develop betweenthem.It would be on. this inside market that
a market price, variable not just from day to day but from moment to
moment, could make its appearance. It was surely the theory of such
a market which was the next thing which should have been formally
set out.
It did not happen, just like that, for two reasons, one general, one
more special. I take the special reason first.
It was bound to be noticed, as soon as the first step was taken
along the road to such a theory, that the market in question would
be a speculative market; and speculative markets, highly organized
speculative markets, for some particular commodities, did unques-
tionably exist. But they also seemed to be a very special kind of
market. Should they not also be regarded, like the 'fish' market, as a
peculiar case? That was indeed the way in which they came to be
regarded by Keynes himself. He had himself done some thinking
about the working of such markets, and in his Treatise on Money
(1930) he gives a good though incomplete account of them; 7 but
' I am sure it was Dennis Robertson who told me about the 'fish market',! think he
must have invented it himself,
7
Treatise, Chapter 29 (in Volume 2i.
 SUPPLY AND DEMAND.? 11

the thrust of his chapter is to explain why such markets are not
important -because it is only in exceptional cases that costs of stock-
holding are low enough for large stocks, of a particular commodity,
to be carried. So, in his General Theory (1936) he leaves them out, A
gap was thus left, between the 'fish' markets, where carrying costs
were prohibitive, and the regular speculative markets, where they
were so very reasonable; into that gap a great number of actual
markets must fall. And Keynes. in neither of his books, gave much
help in dealing with them.
Then there was also the more general obstacle. The theory that
was needed could not be developed without a considerable change in
point of view. The traditional view that market price is, at least in
some way, determined by an equation of demand and supply had
now to be given up. If demand and supply are interpreted, as had
formerly seemed to be sufficient, as flow demands and supplies
coming from outsiders, it is no longer true that there is any tendency,
over any particular period, for them to be equalized; a difference
between them, if it were not too large, could be matched by a change
in stocks. It is of course true that if no distinction is made between
demand from stockholders and demand from outside the market,
demand and supply in that inclusive sense must always be equal. But
that equation is vacuous. ft cannot be used to determine price, in
Walras's or Marshall's manner. For what matters to the stockholder
is the stock that he is holding; the increment in that stock, during a
period, is the difference between what is held at the end and what
was held at the beginning, and the beginning stock is carried over
from the past. So the demand-supply equation can only be used in a
recursive manner, to determine a sequence;* it cannot be used
directly to determine price, as Walras and Marshall had used it,

* It is a difference, or differential, equation.

 2 The Function of Speculation

An outline of the theory, which in the light of these considerations

should have been felt to be required, may be introduced by going
back to the corn market of Marshall. This is a closed system, so far as
the corn is concerned; there is no trade with other such markets; all
transactions are between the parties listed, producers, consumers,
and dealers. But then, as Marshall in his pursuit of realism should
surely have noticed, ought we not to think of new supplies coming
on to the market at harvest-time, while demand is relatively steady
over the year? Someone then must be holding stocks, high just after
harvest, gradually falling to a low point just before the next. It is
reasonable to suppose that they will be held by the dealers.1
We shall further find it convenient, for no deeper reason than con-
venience of statement, to suppose that the market is only open for
one "day* in each 'week'. Our market is accordingly endowed with
two significant periods, the week which elapses between successive
market-days, and the year which elapses between two successive
harvests. These are different, not just in duration, but also in their
relation to the working of the model,2
Any dealer, on any market-day between the harvests, has a choice
between selling (to consumers or to another dealer) and holding on.
If he is to do a bit of both, as some dealers must be doing if the market
is to continue, the advantage he expects (or plans) to get at the
margin from the two courses must be the same. Granted that there
are some extra costs involved from holding longer,1 it can only be
profitable to continue holding a part of the stock in his possession, if
the price he expects to get from selling later is higher than that which
rules at the moment. If this expectation is realized, and goes on being
realized, the market price must go on rising during the year—from a
low point just after the harvest to a high in the following summer,
after which it falls, as a result of the following harvest.

' For that is the arrangement which (usually) will minimize transport costs.
* They have very little to do with the 'short* and 'long* periods of Marshall, which
he introduced at a later point in his analysis to that we are here considering.
* Cost of sale to consumers may (realistically) be reckoned to be borne by the con-
sumers.
 THE F U N C T I O N OF S P E C U L A T I O N 13

It is already apparent; from consideration of what happens in this

very simple example, that there are two quite different senses of the
term 'equilibrium' which are coming up. One relates to the week, the
other to the year. It is perfectly proper to say that the market is in
equilibrium during a particular week if the price established at the
end of the week (at the next market-day) is the same as what was ex-
pected for it at the week's beginning. (This is the ex-ante/ex-post equi-
librium of the Swedish economists contemporary with Keynes,} It is
quite different from the iow equilibrium of Marshall and his pre-
decessors, which in our case can only refer to the year. As we shall
see, there is room for both.
For the market to be in 'Swedish' equilibrium during a particular
week, when that is taken by itself, is not a very stringent condition.
The foresight which is required is not considerable. It should further
be noticed that it need not be stated in so 'subjective' a manner. For it
only requires that the actions taken at the beginning of the week
(holding stock or disposing of it) should be such as would have been
taken if the end-price had been correctly foreseen. That is a matter
which in principle can be tested. It is true that to suppose it to hold
for all weeks within the year is a more serious matter. It does how-
ever lead, to testable results. In our case it leads to the course of prices
that was described—low just after the harvest, rising to a high point
just before the beginning of the next.
If flow equilibrium is interpreted, after the manner of Marshall and
his contemporaries, to mean a condition in which price remains un-
changed because flow demand and supply are in balance, it is im-
possible in our model for the market to be in equilibrium within the
year. It could however be in a flow equilibrium from one year to the
next, to mean that there was an annual cycle which repeated. The
conditions for this to happen are obviously very stringent. The stock
must be the same just before the first harvest and just before the
second; and also the same just after the first harvest and just after the
second; both of these conditions cannot be satisfied unless the two
harvests are equal. And also what is taken off by consumers in the
two years must be the same. If the flow demands of consumers have
any elasticity (their demands for corn are not absolutely inelastic) the
price over the year, its general level over the year, must be such as to
keep demand and supply in balance, over the year as a whole. But it
is only because of the (seasonal) fall in price when the new harvest
comes in that there will, in this equilibrium state of our model, be no
14 THE W O R K I N G OF M A R K E T S

carry-over from year to year. Only so can we be satisfied with what Is

shown in the conventional supply-demand curve diagram. In all
other cases there is more to be considered.
Before we can properly trace the course of the price from one year
to another, there is a little more to be said about the movement
within the year. If 'Swedish' equilibrium is maintained within the
year, the price must be rising just fast enough to cover the marginal
cost of holding stocks. This is in principle composed of two elements,
one physical, the other financial. The physical cost is the cost of
storage, the financial is the interest that is given up on the funds
locked up in the holding of the stock. Now it may well be maintained
that in the full equilibrium state of our model, marginal cost of
storage should be very low, for it is confined to the cost of measures
that are needed to prevent some physical deterioration. Storehouses
indeed will have to be provided, and they will have what may here be
regarded as a rental cost; but this will be nearly the same whether
the storehouses are full or nearly empty. So it may be that the main
cost of storage, during the week, is interest cost—interest which
could be earned, during the week, on the funds that are locked up
in the holding of the stock. If the rate of interest is low, this need
not be considerable; so the annual price-cycle should be nearly
flattened out. (This means, incidentally, that the fall in price, at the
time of the harvest—-during the period of the harvest-—need not be
great.)
But, now suppose that our model, having been for some time in this
'annual' equilibrium, is confronted, in some particular year, with a
harvest which is not so obligingly 'normal'. The cases of an excep-
tionally large and of an exceptionally small harvest are substantially
different, for though supplies can be carried forward in time, they
cannot be carried back.
Take first the case of a deficient harvest. If there had been no news
of the deficient harvest before the end of the annual cycle of the pre-
vious year, the difference which would be made by experience of the
new harvest being deficient would simply mean that the level of price,
over the annual cycle of the post-harvest year, would have to be
higher, There would be no more to be said than that. If however the
news came in earlier, before stocks left over from the previous har-
vest had been exhausted, there would be an incentive to hold them
back from the market; so the price would rise earlier than it would
otherwise have done. One can see that rise being blamed on 'specula-
 THE F U N C T I O N OF S P E C U L A T I O N 15

tion'. It would however do no more than smooth out a rise which

would have inevitably occurred.*
In the opposite case of a more than normal harvest, anticipation of
its appearance could not make much difference. For even if a normal
harvest had been expected, a fall in the market price at the time
when that harvest came in would have been expected, so the stocks
which were held near the end of the harvest year would have been
small; there would not be much that could be released in anticipation
of a further fall in price. The major difference from the equilibrium se-
quence would come after the harvest. The problem with which stock-
holders would then be confronted would be a matter of the future
disposal of their (now) exceptional stocks. How much should be dis-
posed of within the coming year? How much should be carried over
to a further future? That would have to be decided, directly or in-
directly, in the light of the way in which the appearance of the big
harvest was interpreted.
If the big harvest was taken to be exceptional, giving no indication
that future harvests would be anything but normal, it would be
profitable to carry over some part of the supply to future years. But if
it were read as meaning that future harvests also would be expanded
to the new level, or thereabouts, there would be nothing to be gained
from carry-over, so the price would have to fall sufficiently to
engender a low demand equal to the increased supply—just as in the
textbook version of the story where no attention is paid to holding of
stocks. (There would still be an annual cycle about this new level of
prices.) If some dealers read the situation one way and some another,
there would still be some carry-over to the further future; that would
moderate the fall in the level of price in the year after the big harvest
came in. The possibility of holding stocks over the future would be
acting as a buffer.
But then, as soon as we admit such differences of opinion, other
possibilities open. The optimists, as we may call them, who take the
low price to be temporary, will be more willing to hold stocks than
the pessimists, who take it to be permanent or more permanent. So
why should there not develop an Inside' market between the two
classes, pessimists selling to optimists, in which (by arbitrage) a regu-

* Historical cases of this phenomenon, complicated and of course alleviated by

some possibility of drawing supplies from outside sources, have been studied by A. K.
Sea in his Poverty and Famines (1981).
16 THEWORKINGOPMARKETS
lar market is established? That is indeed what happens on the specu-
lative markets, which (as we saw) Keynes put aside as a special case,
The reason why it is special is mainly because of a further compli-
cation. The optimist buyer may not have the facility to take physical
charge of what he (temporarily) acquires; he may have no store-
house at his disposal. What he requires is to have proprietorship in the
article traded, so that he can sell it when he chooses to do so; he does
not desire to have the custody. But how is proprietorship to be trans-
ferred without change of custody? Only, as commercial men have
long discovered, by expressing the proprietorship in a document,
which confers a right.
What is required here is that the right should be easily transfer-
able; so its terms must be precise, A right to take possession of a par-
ticular stock, the physical condition of which will have to be
ascertained at the time when the right is exercised, is not precise
enough. Nor is it enough to leave any obscurity about the date at
which the right is to be exercised. The practical solution is for the
right to be expressed in a "futures' contract, which takes the form of a
promise to deliver, at a specified date, a specified quantity of a stand-
ard grade of the article traded. Such a promise is admirably suited for
trading between dealers; anyone who acquires it can readily pass it
on, at any time before its expiry, at a market price. The holder of a
physical holding can then 'hedge' his position by issuing (or selling)
a corresponding 'future' at a current market price; he is then in a
similar position to that he would have been in if he had sold his
actual stock forward to the date of expiry of the 'future', with the
only uncertainty left consisting in the difference between what his
actual stock is worth at that date, and the value at that date of a
stock of standard quality. It is easy to see that he will usually be able
to protect himself if the 'futures' method is adopted more cheaply
than if he had proceeded by a simple forward sale, just because the
'future' is easily transferable. But for such re-arrangement to be prac-
ticable, there must be agreement among traders on the deinitions of
standard qualities; so the futures market must be an organized
market, in quite a different sense from the organized market of
Walras.
No more need be said in this place on the working of futures
markets. For futures are not themselves commodities; they are prom-
ises to deliver a certain quantity of a commodity (or rather the
current money value of that quantity, as it will be at the specified
 THE F U N C T I O N OF S P E C U L A T I O N 17

date). They are the same kind of thing as an indexed bond. It is in

that connection, when I come to financial markets, that I shall have
a little more to say about them (see Chapter 10). They come in here
simply as one device by which holding of stocks may be made easier,
so that the smoothing of price-movements which has been shown to
follow from carry-over is facilitated. It is however the physical carry-
over which does the smoothing; that can occur, to a signilcant
extent, even in the absence of futures markets.5
For this reason it must be insisted that whatever is the practical
importance of organized commodity markets, at one time (or place)
or another, they need to occupy a central place in a general theory of
markets, because they are the most sensitive markets we know. They
are the practical counterpart to the 'perfect competition' models of
the textbooks; but they do more for the economist than those models
can do, because they bring out so forcefully that most prices are
determined, not by mechanical matching of flow propensities, but by
the way they are interpreted, thus by the state of mind of those who
trade. There is not, at least there need not be, anything 'irrational'
about this. It is Just that knowledge of what may happen in the
future can never be complete. In my 'harvest' model I have simplified
this ignorance, since the date at which relevant information would
come in (the date of the next harvest) was supposed to be known. In
a more general case that also (or what corresponds to it) would be
uncertain. This makes the stabilization more difficult but does not I
think affect, the principle.
The chief things which our model has shown, in spite of this sim-
plification, are (1) that the stabilizing effect of stockholding is better
for dealing with unexpected surpluses than with unexpected short-
ages, and (2) that it is better for dealing with moderate surpluses
than with those that are large. For while the marginal costs of carry-
s
This is surely a point at which I should make an acknowledgement to Kaldor,
whose truly classic paper 'Speculation and income stability' (R. Econ. Studies October
1939, reprinted and revised in the second volume of his Essays 1980} has been of the
greatest help to me in this chapter. I look upon it as the culmination of the work that
was done in this field, not only by Keynes but also by Hawtrey, in the twenties and
thirties. It has not received the attention it deserves, largely I think, because he
plunged hfe readers, without much preparation, into the complexities of futures
markets, which taken like that are ferocious. I am trying here to be more gentle. I
should like to report that in the last letter I had from Kaldor, only a few weeks before
he died, he told me that he knew that Keynes did read that 'stabilization* paper of his
and planned to give it serious attention. But we know that from 1939 to his death in
1946 Keynes had other things to occupy his mind.
18 THE W O R K I N G OF M A R K E T S
ing normal surpluses, such as those of our seasonal cycle, may not be
very considerable, for dealing with an exceptional surplus new facili-
ties, such as storehouses, are likely to be necessary so that marginal
costs will mount up. It is this last which has led, in our day, to the
vogue of 'stabilization schemes' in which the cost of holding "a buffer
stock is met in some way out of public funds. The price is then set by
what amounts to a producers' co-operative. Since it is operated in the
interests of producers, there is always a temptation to set it too high,
so that stocks go on increasing and their costs mounting up. There is
then no way in which normal production can be resumed unless the
surplus is destroyed, or removed in some way or other.
 3 The Pricing of Manufactures

Three reasons why speculative markets in commodities are a special

kind of market have emerged from our discussion. They require, if
they are to flourish, (1) that the article traded should be fairly stand-
ardizable, so that supply from one 'outside' supplier should be a good
substitute for that from at least some others; (2) that dealings should
be on a sufficient scale for the costs of some organization to be easily
covered; and (3) that arbitrage should be possible, so that most of the
participants must be merchants who may either buy or sell. It is easy
to see that these conditions are most likely to be satisfied in the case
of raw materials; even the 'corn' of Marshall's market would be a
raw material for millers. Thus the market, so far considered, can be
no more than one link in a chain of transactions, extending from
primary producer (farmer or miner) to ultimate consumer. At each
link of the chain there is work for intermediaries. We may begin by
thinking of them as acting independently, though we shall find that
it is of the greatest importance that by Vertical Integration' they may
be brought together. It is accordingly suggested that in the simplest
model of a production system we should make places for two sorts of
intermediaries—one between primary producers and manufacturers
(this being the stage to which our previous discussion should now be
taken to refer), the other between manufacturers and consumers,
what are commonly called the distributive trades. Since it is the
resemblances and differences between these two sorts of intermedi-
aries to which I now desire to direct attention. I shall venture to call
these distributive trades secondary merchants, contrasting them with
the primary merchants, who deal in primary products.
Even among the secondaries there may well be disintegration,
wholesalers and retailers at least being distinguished. Wholesalers
sell to retailers and buy from manufacturers: retailers buy from
wholesalers and sell to consumers; that is the normal course of their
trade. Retailers rarely buy from other retailers; wholesalers however
may buy from other wholesalers, just like the merchants on a
primary market. Bach of them, unless he is dealing in a perishable
good, will need to hold stocks, A retailer who allowed his shop to be
emptied would soon be out of business.
2O THE W O R K I N G OF M A R K E T S

For the purpose of the retailer in holding a stock is different from

that which operated on the primary market. It has nothing to do with
speculation, with carrying forward, in the hope that at a later date
the price will be higher. It has nothing to do with movement of
prices. It would still exist in a world where prices never changed. By
setting up his shop, the retailer has given notice that he is ready to be
a seller of a particular class of goods. The penalty for being 'sold out'
is not mainly the loss of proit on the goods that might have been
sold; it is mainly a loss of reputation, or goodwill.
It is not only at the stage of retailing that this reputation motive
may make its appearance. In the primary commodity market it is
probably at its minimum; for the dealer on such a market, need not
run out of stock, since at a price he can always replace. It is by
making losses on unprofitable transactions that he may come to
grief. The same could be true of the secondary wholesaler stage, to
which I shall be returning shortly.
But what of the manufacturer himself? There are two kinds of
marketable stock that he may be holding, (Half-finished goods, goods
in process, will usually not be marketable.) These are stocks of mater-
ials and stocks of finished products.
If he has easy access to a primary market on which the goods he
uses as materials are traded, he will not need to hold much in the
way of stocks of materials. (The chief reason for holding them will be
to avoid the extra cost of frequent deliveries.) For what he buys from
the market he will pay the market price. The price that is formed, in
the manner we have examined, on that competitive market will be
just transmitted to him.
On the holding of stocks at the stage of finished product there is
more to be said. We can indeed conceive of an economy in which the
same would be true in the case of finished product as in the case of
materials. The product, as soon as it was completed, would be sold to
secondary merchants (wholesalers) at a price which was mainly
determined by trading among the wholesalers themselves. The
manufacturer would have little means of exercising a direct in-
fluence upon it. This would be the practically realizable counterpart
to the 'perfect competition' model of the textbooks.1 It is however not
* At least in the sense that the manufacturers would be 'price-takers', not making
prices by their own decisions. The textbook 'perfect competition' model is of course a
static or 'equilibrium" model: it is not concerned with adjustmeat to a changing en-
vironment, the problem with which we are here concerned.
 THE P E I C I N G OF M A N U F A C T U R E S 21
surprising, especially when, it is spelt out in this manner, that it
should appear to the modern student to be very strange,
I do not believe that it need always have appeared so strange,
(That is why it was able to get into the textbooks!) There was a time,
perhaps including a great part of the nineteenth century, when the
principal end-products of manufacturing industry were rather
simple: cotton and woollen textiles, sold by the yard, tools (knives
and forks and hammers), even some sorts of basic foodstuffs (flour
and sugar),2 It is true that these could be regarded, in a way that was
going to be important, as half-finished goods, left to be used or made
up in the home. As long as these were the chief sorts of goods in ques-
tion, it would be quite appropriate and convenient for stocks of them
to be held by the wholesalers.
I have elsewhere3 suggested that this was the system which may
well have been in Marshall's mind when he came to his short-period
theory of the Industry', a part of his work which had particular in-
fluence on Keynes. The short period, it may be remembered, is defined
as that which elapses before the fixed equipment of the manufacturer
has bad time to adjust. 'The producers have to adjust their supply to
the demand as best they can with the appliances already at their dis-
posal.'4 Such adjustments as can be made under this restriction are
taken to be rather rapid. Some time must nevertheless elapse
between the date when a decision to change the rate of output Is
taken and that when the actual change results. But it does not
greatly matter how the decision comes about. It could be that the de-
cision is made by the manufacturer directly on his own initiative,
reacting to a change in price on the wholesale market (and having to
take the risk that when the output comes to be ready the price on the
wholesale market may have changed); or it may be the wholesaler
who gives the order, at the price which is currently ruling on the
wholesale market, himself having to take the risk that by the time the
goods are ready the price may be changed. Not much foresight is

2
An octogenarian, like the present writer, can remember those days, I think of
going with my mother to do her shopping (about 1910). There were none of the pack-
aged goods which are the principal contents of the modem shop. There were bins and
jars from which the goods were taken out in ladies, Then they were weighed out, and
the quantity purchased was wrapped up In thick blue paper, I would like my reader to
imagine that; it is a condition which can exist, for it has existed.
* Capitol and Growth, p. 53.
* Marshall, Principles, p. 376.
22 THE W O R K I N G OF M A R K E T S

required to adjust to this trouble so long as the time taken by produc-

tion is short.
If conditions such as these are granted, there is nothing in
Marshall's short-period theory which seems to be open to serious criti-
cism. It needs not raise any puzzles about 'laws of returns'. There is
no need to assume that there is a single optimum output for which
the plant is designed; it is better, being more realistic, to think of it as
having a regular range of outputs (from x0 to x,) which it is reason-
ably well fitted to produce. It would then be a reasonable simplifica-
tion to suppose that over that range marginal cost is simply running
cost per unit of output, not including any contribution towards
covering the overhead cost of the plant itself, and this could be taken
as constant so long as the prices of primary factors are given. Call this
running cost c0. If the price that was offered for the product was less
than c0, to operate the plant at any intensity \vould be unprofitable,
(That is not to say that the plant must be idle; if the low price is ex-
pected to be no more than temporary, there may be a gain in some
sort of reputation or other to offset the loss.) If the price that is offered
is greater than c0, it will always pay to go to the top of the regular
range (to x,), for that is where profit is maximized, or loss (after
allowing for overhead) is minimized,* The overhead does not have to
be considered when the question is just one of the scale of production
from a given plant.
If the price that is offered is well above c0, it may pay to expand
output above Xj, for x, may be well below capacity output (x2). From
xl to Xj marginal cost is likely to be rising, so there should be a rising
supply curve of output from the individual plant6 But Marshall does
not need this refinement to get a rising supply curve for his 'indus-
try'. It would be enough that there should be different plants with dif-
ferent levels of basic running costs (c0), more of them coming into
5
The borderline case when price = c0 does not need attention.
* There are several reasons for this, which should be distinguished. In the first
place, it is useful to have some spare capacity in case of accidents (this is analogous to
the need for reserves, on which much will be said when we come to the financial
sector). Secondly, even if no question of replacement of plant is at issue, it will require
maintenance to keep it in order. The cost of that maintenance could be reckoned into
running cost; it is easy to admit that this would quite generally rise when maximum
regular output was exceeded. SA distinction between maintenance expenditure saved
by temporary closure and by permanent would have to be noticed.) Thirdly (the most
difficult, in practice as well as theory), there is the effect of over-usage, in excess of
'regular' output, on expected life of equipment, the'user cost'of Keynes, For the value
to be set upon this depends on expectations, of a future that may be far away.
 THE P R I C I N G OF M A N U F A C T U R E S 2J

production when the price that is offered rises. (Just as happens in

Ricardo's model of agricultural production, which Marshall is imitat-
ing.)
I have thought it right to give this amount of attention to the
short-period theory of Marshall, not just out of respect to it as a his-
torical monument, but because of the considerable impact that it had
had on later work, not least on a puzzling chapter of the General
Theory itself.7 Surely however by the time of Keynes the structure of
production and marketing which it had assumed had become quite
out of date. The typical end-products of manufacturing industry no
longer consisted of objectively standardizable goods, which could be
traded on competitive wholesaler markets: they had become much
more various, new products and new varieties being continually
devised.
There were indeed a number of economists who were attempting
to construct theories to deal with this diversity; some of them were
working in Cambridge, close to Keynes himself, Joan Robinson, in
particular, was a leading member of Keynes's own circle. Keynes
however made no use of her Economics of Imperfect Competition
(1933). I believe that for his own work he was quite right to pass it
by. For her theory, like most of the others than becoming available,
was a static theory. It was confined to a comparison of states, in each
of which there was a diversification already established. It did not
show how an imperfectly competitive system would work; but that
was what Keynes required.
There was indeed one economist who had attached himself to this
group, who had seen the problem and made an attempt to face it.
This was Roy Harrod. In his article 'Doctrines of imperfect competi-
tion' {Quarterly Journal of Economics, 1934) he tried to deal with it on
Marshallian lines, distinguishing positions of short-period and of
long-period equilibrium; no doubt it appeared too late for Keynes to
be able to take advantage of it. Harrod himself was later to become
dissatisfied with it; it was nevertheless the best thing in the field
which was available to those who took part in early discussions of
the General Theory. But the direction in which it led did not prove in
the end to be fruitful.
One can indeed now see that the stumbling block in the Harrod
theory was already present in the Marshall theory on which it was
7
Chapter 21.
24 THE W O R K I N G OF M A R K E T S

based. How did Marshall himself suppose that his industry was to get
into his long-period equilibrium? It would have to be supposed that
his firms, or those controlling them, were endowed with remarkable
foresight. They would have to see the equilibrium, coming, and adjust
to it in advance. For if they got it wrong, they would have the wrong
equipment and would have to start all over again. That is one reason
why the long-period equilibrium of an industry is a less useful con-
cept than many neo-classics (and Harrod) imagined.8 It is better to
go back to the start and enquire how it could have been that the
diversification came about.
There must have been a sequence of occasions on which decisions
to introduce new products had been made. The maker of such a de-
cision would have been an entrepreneur or innovator, a character
who has not yet appeared in our story. For the manufacturer who
simply responds to a signal given to him by the market, doing so
almost automatically, is not called on to innovate. Our entrepreneur
has to devise a new product, make arrangements for manufacturing
it, and also make arrangements to get it sold.
For since the product is specialized, no other manufacturer produ-
cing anything exactly like it, any merchant to whom he it dir-
ectly must be dependent on him for supply. The merchant must thus
be acting, in this part of his business, as the manufacturer's agent. So
we have here an important example of the vertical integration pre-
viously noticed; manufacturing and selling come in substance under
the same control.
There were two functions which we were attributing to our
secondary merchants and their market: stockholding and price-
formation. As we saw, they are nearly allied; so it is here. The selling
department is able to set a selling price and make it effective by hold-
ing stocks. That is to say, it can do its own buffering; and can do it
relatively easily, since producing and stockholding have been
brought so close together. So the price that is set can be chosen, as a
matter of policy,
It is of the greatest importance that while the Marshallian manu-
facturer was selling in the first place to professionals, who would be
able to assess just what it was they were buying, it is now the pro-

" It should perhaps be underlined that this is not only a problem of manufacturing
industry. It is a problem of any form of production which uses fixed capital oo a con-
siderable scale.
 THE P R I C I N G OF M A N U F A C T U R E S 2§

ducer himself who has to take responsibility for the quality, and use-
fulness, of what he is selling; for he is selling, at least at the end of the
chain, to a consumer who is not an expert. That is why at this point
there is a function for advertisement, which is basically a promise
about the character of the thing being sold. It is a promise like that
which is given by the retailer, when he opens his shop. In each case it
is gi¥en by a professional to a non-expert, so it quite ordinarily needs
to do more than just give information. The attention of the customer
has to be attracted, by a smart shop-front in the one case, by pretty
pictures and suchlike in the other. But he has then to be persuaded to
buy on the strength of the information given to him, including a
promise, explicit or implicit, that the information is correct.
The price is one aspect of the offer that is made; there are some
characteristics of other aspects which are shared by it. The chief is
that it must not be changed arbitrarily, at a moment's notice.
Arbitrary changes 'unsettle* the consumer. He may be taking time to
decide to buy; so if, when he finally decides, he finds the price has
risen against him, his confidence is lost, and the seller's reputation is
damaged. And it can happen that there is a similar obstacle to price-
reductions; they cast suspicion on the quality of the product, they
suggest that something is wrong. Thus the diversified market had a
tendency to be what I have called* a flxprice market, meaning not
that prices do not change, but that there is a force which makes for
stabilization, operated not by independent speculators, but by the
producer himself,
It is important (as Okun10 has emphasized) that the stabilizing is
more effective against price-reductions than against rises; the latter
can be put through without loss of reputation, if an objective reason
can be given for them. The most obvious is a rise in costs, which has
affected not only this particular producer, but his (imperfect) com-
petitors also. What he must not do (as he so often seems to do in the
textbooks) is to admit that he is putting up his price because demand
has increased: 'I am charging you more, because I can get more out
of you.' The other side to this is the lack of necessary response to a fall
in costs. It is tempting, then, to take a monopoly proit just by taking
no action. The only safeguard against that which is offered by a
diversified market is the appearance of new varieties which, If costs

* Crisis in Keynesian Economics (1974), pp. 22-40.

10
A, Okun. Prices and Quantities {1981).
26 THE W O R K I N G OF M A R K E T S

in general have fallen, can be offered at an appealing price. I suspect

that this is the main way in which {in normal conditions") prices
can come down in such a market."
I
* Exceptions are {1} when there has just been a very sharp rise in prices, which has
not had time to get established, aad (2) when there has been a great fall in the prices of
raw materials, on 'lexprice' markets. In the slump of 1921 both of the* conditions
were present, in that of 1930 only the second, and the fall in consumer good prices
was much less.
II
This should be reckoned by the economist as a fall in prices; but his statistician
partner, who makes his price-index numbers on the basis of a physically unchanged
bundle of commodities, makes it hard for him to do so. There can be little doubt that
real incomes, over the present century, have almost everywhere risen much more
than appears from the statistics.
 4 The Labour Market

There remains one extremely important non-financial market which

has so far escaped our attention—-the market for labour itself. How
does that fit in? How does, or rather ho\v can, a labour market work?
It might have been expected that the author of the General Theory
of Employment would have given some help towards an answer, but
he gives us very little. Nearly all he says is negative, Just that on the
labour market there is no equilibrium of demand and supply. But
demands and supplies of labour are flows, work to be done over a
period, and we have been seeing that there is no inevitability, in
other markets, that flow demands and supplies should always be in
balance. A difference could be made up by variation of stocks. Labour
however is not a stock that can be carried forward. As a Victorian
economist once said,1 it is 'more perishable than cut flowers', more
perishable, we might say, than 'fish'. So if there is to be an equilib-
rium, a continuing equilibrium, with unemployment, something
must be implied on what is happening to the unemployed labour.
At the time when Keynes was writing, provision for 'unemploy-
ment benefit' was being extended in many countries; so it was
natural for his early readers, and many later readers also, to take it
for granted that the unemployed were being supported by some kind
of public assistance. So long as that continues, the unemployed
would not need to find a way of supporting themselves, so they need
not compete with the labour which remained employed. In the
model, accordingly, the level of wages could be taken arbitrarily, as
Keynesians often appear to take it. This would correspond, in, prac-
tice, to the wage being fixed by the power and policy of trade unions.
On a wider historical, or geographical, perspective that was surely
a special case. Joan Robinson, a leading Keynesian, found that it was
when she started to think about India. Her 'disguised unemploy-
ment' was a fitting of Indian experience into a conventional Key-
nesian mould,
I have found it more instructive to begin the other way round,2 For

1
W, T. Thornton, On Labour (} 868).
2
At greatest length to my Theory of Economic History, pp. 1 34-40.
28 THE W O E K I N G OF M A E K E T S

surely there were wages before there were trade unions. Consider for
instance the labour market in Britain in the days of Adam Smith,
Even then wages were beginning to coagulate into some sort of a
pattern, We shall understand the wage-system better if we begin by
considering how it could have started in those days, and then go on
to see how, in what circumstances, and to what extent a trade union
system could have grown out of it, (That is similiar to the procedure
we found it useful to adopt in the case of the market for consumer
goods; I shall follow that procedure here.)
A standard model of this early stage of deYelopment would show
no more than a part of the whole labour force, or potential labour
force, being paid wages; the rest would have been supported other-
wise. They could be regarded in Marx's manner as a 'reserve army',
but they need not be idle. They might be working on family farms
(which could have been paying rent to a landlord, but would not be
paid wages by the landlord) or they might be doing domestic work in
a family home. In either case it is by family connections that they are
being supported.
One can see that a considerable movement, from family work to
wage-labour, would frequently be matched by a movement from
country to town. It is in that context that I find it convenient to begin
to consider it.
It is a matter of major importance that there are two ways in
which the movement could occur—according as the initiative is
taken by a potential employer, or by the immigrant himself. These
are fairly distinguishable in practice, since if the initiative is taken by
the immigrant, he must himself bear the costs of movement, so he
must almost inevitably come from fairly near at hand; while if the
initiative is taken by the employer, labour can be brought from afar.
(There are exceptions to this rule when the movement is subsidized.)
It will however not pay for an employer to bring labour from a dis-
tance, or expensively, unless It is expected that the new arrivals will
go on working, for the employer who has paid for bringing them, for
some considerable time. So their employment must be, at least to
some extent, lasting or (as I shall call it3) established employment.
As for the people who bring themselves, they will have no such
assurance. Some of them, perhaps most of them, will just make a pre-
carious living by picking up odd jobs. Though they are paid for doing
1
All possible alternatives have associations which I do not, for the present, require.
 THE LABOUE M A R K E T 29

those jobs, there is no market on which a regular rate of pay can be

formed. For there are no Intermediaries who can 'make' a market;
neither 'buyer* nor 'seller' is professional. If we stretch the term
'market' to include it, this is the most disorganised market that can
be conceived.
Nevertheless, even after beginning like this, our immigrant can
sometimes make progress. If he has brought with him some particu-
lar skill, or can persuade people that he has some particular skill, he
can let it be known that he claims to be competent to do that sort of
work. He can then open what amounts to being a shop for his ser-
vices, and much of what I have been saying about retail trade will
apply,* The quality (or standard) of the service provided cannot
easily be stipulated, so it becomes a matter of importance to the seller
of such services to establish his reputation; the price-policy which he
adopts is in part a means to that end. To quote a low price is a means
of entering the market, but it is the low' end of the market which is
most easily reached in that way. To work 'up' the market is a matter
of gaining reputation,
Accordingly even here there is a wide range of possible outcomes,
between failure and success. Those who fail are on the edge of starva-
tion;5 those who succeed may make fortunes. It is indeed from those
who succeed that the entrepreneurs, who become the employers of
established labour, may well be recruited.
So let us look again at the established sector, where there is a rela-
tion between employer and employed which promises some con-
tinuance. The employer expects the employee to stay with htm, at
least long enough to make a wage-bargain on that assumption, and
the employee the same. It is here that there can most obviously be a
market on which a wage is competitively determined.
If the market is to be a competitive market, each must be free to
change his partner. The employee must be able to go away if he can
find what he thinks to be a better offer, and the employer to dismiss
him if he desires to do so. But by making their contract, the two have
agreed to work together; if it is denounced by either party, each has
suffered a defeat, The loss of the worker who is obliged to look for

* The shops of merchants and the workshop of artisans may be found side by side,
as I have seen them myself, in Isfahan, in the days of the Shah.
* I think of the London poor of the nineteenth century, such as figure in the novels
of Dickens. Why were things so different, for most of that time, in the United States? I
shall cotne to that question later (p. 34),
3O THE W O R K I N G OF M A R K E T S

another job is no doubt much more severe than the loss to the
employer, if the latter is only obliged to seek for a replacement; but
when the dismissal is incidental to a decision by the employer to
reduce the scale of his operations, that is bad for him too. Thus in all
cases of premature ending there is at least some loss for each, a loss
which is better avoided. So it pays to take some trouble, and even to
incur some expense, in order to avoid it. There is not much that the
employee can do by himself to protect himself except to "give of his
best', and that can be no more than a partial protection. The
employer, on the other hand, can see that he pays a wage which is at
least as good as what is being paid by his competitors, so that an
employee who has become established is unlikely to be tempted
away. This is surely the principal way in which competition works,
in the established sector—not by actual change of partners, but by
potential change.
I think one can show that this is a matter of major importance,
that it is indeed the essential way in which the labour market, when
it is an established labour market, differs from the markets in goods,
which we have been considering hitherto. Competition on markets
for goods works for the most part, as we have seen, in terms of actual
transactions; this is particularly so when there are intermediaries,
whose actual dealings 'make' the market. So much is sold, and such
a price is given for it. But potential competition does not work
through actual transactions; it works through the influence of ideas
about transactions which might be made, but are not. On these vari-
ous parties may have different ideas. It is here, I believe, that the
single employer, confronted by an unorganized labour force, has his
chief 'bargaining advantage'. It is simply that he can afford to be
better informed, better informed about the alternatives which for this
sort of labour are open. He is, in terms of our previous discussion,
more of a professional than his employee. But Ms bargaining advant-
age is diminished if the employee also can find means of getting pro-
fessional advice.
So this is the first way the trade union comes in. At this stage the
function of the trade union official (and, still more obviously, of the
shop steward) is similar to that which used to be attributed to the
broker on the London Stock Exchange. There is little place for jobbers
on a labour market, any more than on a fish market; but the broker-
ing function, the provision of professional advice to the non-
professional party, is needed to make the competitive market work.
 THE L A B O U R M A R K E T 31

But a manual tvorker, by himself, can rarely afford to pay for profes-
sional services; thus the obvious solution, in the market for estab-
lished labour, is for a number of workers to get together, jointly
employing an agent—collective bargaining. It could be that this just
made the competitive market work more smoothly.
But like other economic activities this function is subject, up to a
point, to scale economies, so the trade union is made more effective
by increasing its membership; that leads on to a second stage. For
here as elsewhere increase in size affords opportunities for mono-
polistic behaviour; by using the strike weapon, or threatening to use
it, a union may be able to extract gains from employers and through
them from their customers. But to analyse their actions in these
terms, though it is tempting for an economist to do so, since he has
his monopoly theory at his disposal (an essentially static theory, it
should be noticed) does not bring out aspects of the problem which
experience has shown to be of importance. Trade union members
cannot easily be mobilized to take action, which is costly to them-
selves when they are on strike, just to get a relatively small gain in
the ensuing period. So they are characteristically better at defence
than at attack.6 This has consequences that can be traced.
First, it is easy to resist a formal reduction in (money) wages; that
is the most obvious. So it is that in a well-unionixed market, a
straightforward reduction in wage-rates hardly ever occurs, except
on a few extreme occasions, mostly when the reduction is under-
stood to be temporary, and employment could hardly continue at all
without it.' Other methods of reducing labour costs will normally be
preferred.
Secondly, there are what are nowadays called relativities. It could
be that in a perfectly working competitive labour market, when the
wage of one sort of labour rose, the wages of similar sorts would be
drawn up with it. But that implies that there is a fairly easy move-
* This was not enough allowed for, though it got some attention, in the thumbnail
sketch of trade union history in Britain, up to the date of writing, which I gave in
Chapter VIII of my Theory of Wages (1932), That is not bad as far as it goes, for it is
based on empirical work I had been doing in the years preceding (since 1925), But the
accounts I have given in later work, such as 'Wages and inflation' (1955, reprinted in
the second volume of my Collected Essays), are more mature,
' The classical example of this, with an arrangement of this sort being formalized, is
the selling-price sliding scales which were used for the regulation of wages in British
coal-mining between 1870 and 1900, This was a very fluctuating industry with
labour costs peculiarly hea¥y; it was for a while accepted that labour must take its
share in adjusting to the fluctuations.
32 THE WORKING OF MARKETS

ment, or transformation, of the one kind of employment into the

other. If the movement is difficult, it will not happen without union
pressure. But it is understandable that people should think it ought
to happen: if it does not, that is "unfair*. This gets extended to cases
when mo¥ement, within the period for which the wages are to run,
is quite impossible; so it turns into a pressure for many wages to rise
when some wages are rising.
These two taken together could be sufficient to account for the
wage-inflation, or threatened wage-inflation, which has been with
us, in so many countries, since 1945. Even if it is not resisted, it need
not (so we are told) lead to price-inflation, if there is a sufficient rise in
productivity. But it is quite independent of whether or not there is a
rise in productivity, so it would not be surprising to find that it gener-
ally soaked through to price-inflation. But for that there has in fact
been another source, the desire to defend not just a money wage, but
a real wage, the money wage deflated by a price-index.
It is often said by economists that it is real wages in which trade
unions should be interested, not money wages. But the fact is that
until the beginning of the present century, there were no consumer
price-index numbers, so there could be no question of an appeal to
them: but quite strong trade unions go further back. All that could
then be noticed by trade union members and their representatives
were sometimes rises in particular prices, due to recognizable causes.
It was not easy to base upon these claims for rises in wages in quite
different industries.
The beginning of a change was in World War I. The general rise in
prices, due in the first place to war-time shortages, was reflected in
the (still very primitive) index numbers that were beginning to be
available; it was clearly an element that had to have attention in
wage-bargaining, sometimes going so far as to attach the money
wage to a cost-of-living index, But that was taken at first to be just a
matter of war-time disturbance. It must nevertheless have facilitated
the great fall in money wages which occurred in Britain in 1921-2,
when the wage-index (with 1914 as 100) fell from 280 in 1920 to
194 in 1923. Though that was matched with a corresponding fall in
consumer prices, it was surely enough to disgust the trade unions
with 'cost of living'. Little was heard of that in the later twenties and
the thirties, when the price-index was steady or falling. So it was that
when Keynes was writing, it was the maintenance of money wages
for which unions were pressing, since that suited them better. It
 THE LABOUR M A R K E T 33

turned out to be most unfortunate that Keynes and his followers

should have allowed this transitory state of affairs to be embedded so
deeply into the structure of his theory,
During the Second World War, the experience of the First was re-
peated, but the sequel did not repeat, since (no doubt because of
better management) there was no post-war slump. So there was no
disenchantment with 'cost of living'. War changed to peace much
more gradually, so war-time habits of thought did not so sharply dis-
appear. It is conceivable that they could slowly have disappeared, so
that the wage-system, like other parts of the economic system, could
have been slowly steered into something more stable. In the British
case, to which I cannot help referring (since it is that with which I
am most familiar) it nearly was done. When I wrote about the matter
in 1955, a good moment to be writing about it, I allowed myself to
hope that this was going to happen, and for the ifteen years that fol-
lowed it largely did. I quote what I said on that occasion:
The continual rise in money wages since 194 5 . . . is sufficiently explicable to
terms of factors that are peculiar to the time through which we have been
passing. Especially during the last four years, the main factors pushing up
wages have been (1) the dismantling of the controls, with its somewhat
'phoney' effect on the cost of living, and {2} the difficulty which has been ex-
perienced in the establishment of a new system of relative rates after the
war-time disturbance. But these are difficulties which, in the absence of ex-
ternal shocks, we can expect to overcome ... But it can hardly be doubted
that any serious disturbance of our rate of progress would itself push the
level of money wages in an upward direction.8
This is surely what actually happened in the British case. The "good*
years of the late fifties and the sixties were brought to an end by an
external shock—a rise in the prices of several important imports,
relatively to those of exports—which implied a reduction in the real
wage that British labour was able to earn.* It was this which set off
the following troubles. How much can be done in such an emer-
gency, by monetary or semi-monetary measures, will be examined in
later chapters,
But before concluding this, a little more should be said about the
distinction, fundamental to my argument, between established and
* 'Economic foundations of wage policy' (Economic journal, 1955), reprinted as 'In-
flation and the wage structure' in my Collected Essays, Volume II (1983).
* It is very striking that in those good years the British terms of trade were nearly
constant, while from 1970 to 1973 there was an adverse swing of about 20 per cent.
34 THE W O R K I N G OF M A R K E T S

non-established, or less established, labour. ! have tried to explain

why trade unionism, and the 'social' influences on wages that go
with it (for employers as well as employees are affected), are natur-
ally characteristics of an established sector. The proportion of the
labour force that is established can vary, from one time to another,
and from one country to another. This is not just a matter of the
degree of industrialization. It is by no means necessary that the
whole of an industrialized labour force should be thoroughly estab-
lished. It is possible to organize industry with no more than a nucleus
of established labour, the rest of those who are employed in produc-
tion being more loosely attached. Semi-skilled labour, which can be
quickly trained to do its work, need not be tightly attached. Modern
technology may well tend to make relatively increasing use of less
attached labour. But I do not have the information to make more
than a guess at that,
It is however worth noticing that if Britain (and possibly other
European countries also) can be reckoned to have a long tradition of
established labour, while in the United States the tradition goes the
other way, much would fit into place. One can see, irst of all, how it
could have been that such a difference arose. In Europe, in the form-
ative period, labour was being drawn out of agriculture into indus-
try; and when people had moved, they could not easily go back. They
had made a gain by moving, but a gain which they soon felt that
they needed to defend, fa the United States, on the other hand, the
'frontier' as it was called long remained open; there were also
expanding opportunities in farming and in occupations ancillary to
farming, so there was a two-way movement between the sectors. So
the industrial worker had less need to defend his position; for if he
were not paid what he thought to be a competitive wage, he would
just go away, 'Exit'! Of course there were and are trade unions in
America; but they have never sought to play such a part in the
American economy as British unions have for so long in the British.
That has repercussions which go very deep.10
10
There are examples from other countries which seeni to support this thesis.
apparent exceptions which prove the rule, I think of Australia and Argentina, 'new'
countries with very strong trade unions. Each has had a large, and for a while a grow-
ing, agricultural sector, but in each case the farming was highly capital intensive,
which gave little opportunity for 'exit* from industry. The Australians have sought a
release from their predicament by high protection; the Argentinians have failed to find
even that. But I only mention these examples to show that they are not inconsistent
with the general line of what! am saying.
 THE L A B O U R M A R K E T 35

It has for instance had its echo in the world of economists, making
it hard for American economists and British {for example) to under-
stand one another, not only on the particular matter of labour rela-
tions but more widely also. 'Search' theories of employment, which
have had quite a success in America, do not in, other countries have
much appeal. 'Full employment*, that sheet-anchor of the Keynesian
system, looks quite a bit different according as one is thinking of the
one sort of labour market or the other,
Keynes himself was surely thinking of employed labour as estab-
lished labour; his unemployed were people who expected to be estab-
lished but for the present were not. If their unemployment is only
temporary, they will still have an eye on their established places; an
increase in effective demand should bring them back, where they
were. They would not have lost the capacity to fit into those places;
they would have not Just the skill, but also the use of the experience
they had possessed. So if the Keynesian prescription was just directed
towards helping a recovery from Depression (which is how many of
its first readers including myself were inclined to take it) it is beyond
question correct." But it went on to claim that the same would hold
if the Depression were long lasting; that, in the light of later experi-
ence, is less convincing. Perhaps it Is more convincing In the case of
the American-style economy, where most of those who are employed
are less established; they will then have lost less in employ ability by
being unemployed; if they are 'out', it is not hard for them to get 'in'
again. The same may not be true of an economy with a large, very
fully established sector, the sort of economy which we thought that
Keynes had in mind.
I am getting short of words with which to explain myself, so had
better have resort to adjectives which go better into symbols. Let us
give the established employment the name solid employment or
S-employment. and the labour which expects to get S-emptoyment
the name S-labour. Similarly, for the less established employment let
us use the term fluid employment or F-empIoyment, and for the
labour which expects to get F-employment the term F-labour."
In our initial state of Depression, there is S-unemployment and
F-unemployment. When effective demand is increased, according to
Keynes, both S-unemployment and F-unemployment should be
11 The only problem is financial—how to finance the expansion.
" We shal! find it useful to use 'solid* and 'fluid' in ways that correspond, when we
come to the financial sector.
36 THE W O R K I N G OF M A R K E T S

absorbed. But suppose that for some reason (just a particular kind of
technical change, or it may be that because of labour troubles
employers would like to get out of employing S-labour) it is F-labour
that is more readily adsorbed, 'Keynesian full employment* is theo
more quickly reached in the F-sector than in the other. Wages, as
Keynes would expect, then start rising in the F-sector; and the rise
carries O¥er to the S-seetor, on trade union principles, in spite of
the unemployment which there persists. Further, since labour for
F-employment does not need much training, it is drawn in from what
was formerly outside the labour force (female labour and immigrant
labour are obvious examples). So the labour force, as statisticians
measure it, goes on expanding, and unemployment, as statisticians
measure it, S-labour going on being substituted by P-labour, gets
worse. The obstacle, it will be noted, is the blockage in movement of
labour from S to P."
Whether this is what has been happening, in Britain and perhaps
in some other European countries, during the last tee years or so, is
not a matter on which it is appropriate for me to pronounce. The
business of theorizing, such as I am engaged in in the present work,
is to ask questions and to formulate questions, not to answer them;
still less to make recommendations on what should be done to meet
the challenges which appear to have been raised.14

** A similar problem may indeed arise when there is increased demand for S-
labour, but the demand is for S-labour to work in a different part of the country from
that where S-labour is unemployed. Transfer of labour then implies transfer of resi-
dence, always expensive to the worker—made more expens«¥e in current British con-
ditions, by the subsidized or rent-restricted housing which is a hang-over from the
welfare state of the years before 1980, A subsidy on stagnation!
14
The Keynes theory on wages and employment, on which in parts of this chapter I
have been commenting, has been taken in a form which is not exactly that in which it
appears in his famous book. There he makes concessions to critics which, as sub-
sequently appeared, he need not have made. This is most evident in his curious
Chapter II. now known to have been written after the rest, in reply to criticisms which
had been made on the other chapters by his Cambridge colleague Pigou. Pigou was
arguing from a fully Marshallian position, on the formation of prices of manufactures,
that in the "short period' an increase in demand must raise their prices. So if money
wages are given, an increase in 'effective demand* must lower real wages. Pigou main-
tained that it was this reduction in real wages which raised employment. Keynes,
accepting that this would happen, claimed that Pigou had got the chain of cause and
elect the wrong way round. All this would have been quite unnecessary if it had been
accepted thai (as I have tried to show in the preceding chapter) there is likely to be an
important phase in recovery from depression, when firms who have been holding their
selling prices to what they think to be a normal level, have no incentive to raise them
when demand returns to normal. Over a range, that is, they will operate in a "flxprice"
 THE L A B O U R M A R K E T 37

manner. This was brought home to economists, soon after the GT was published, by
the work of J.T. Dunlop {'The movement of real and money wage rates'. Economic
loumat, 1938) and M. Kalecki (Essays in the Theory of Economic Fluctuations, 1939), I
do not claim that when I published tny version of Keynes In 1937 (what has since
become known as the ISLM diagram) I had myself got clear ort the matter. It was just
that I saw that the best way of simplifying Keynes was to take money wages (pro-
visionally) to be constant. I am grateful for the help I have had from Professor Tom
Wilson on this matter.
This page intentionally left blank
 PART II

MONEY AND F I N A N C E
This page intentionally left blank
 5 The Nature of Money

It will no doubt have been taken for granted that in the markets we
have been discussing, the typical transaction was an exchange of
some article (good or service) for something that was recognized as
being money; and it may also have been taken that the money was
simply handed over, as one does when one buys a newspaper in a
shop, A useful way of introducing the monetary theory, which will
be the subject of the chapters which follow, is to begin by calling into
question these two assumptions, asking how far they are justified,
Is is convenient to start with the second. It is clear from the most
common personal experience that spot payment—payment 'on the
nail' or 'on the spot'—is by no means the only, or perhaps even the
most important, way of doing business, I may pay spot for a news-
paper as I walk along the street, but I may also give an order to a
newsagent to deliver a copy to my house each morning. I should not
then pay for each issue as I received it: I should wait until the en4 of
the month when he sent in his bill. At that time I should have been in
debt to him for the papers I had got from him; the payment 1 made to
him would have been in settlement of a debt. Surely it is the latter
which should be taken as the general form of a transaction (of sale or
purchase); it covers the case of the spot transaction, when the debt is
settled immediately; but there are many more complicated trans-
actions it also covers.
It is probably true that it is only for small transactions—small, that
is, from the point of view of one or other of the parties concerned—
that the spot method of payment is ordinarily preferred. People are
not, and never have been, in the habit of carrying about them a suffi-
cient quantity of coin or notes to pay for a house or to pay for
furnishing it. Even if the notes are of large denomination it is unsafe
to carry them about without precaution. Further, when the article is
of considerable value, the right of ownership in it has to be trans-
ferred; arrangements, usually legal arrangements, have to be made
to get the transfer recognized. Each of these considerations tells
against the use of the spot method. In transactions between firms
which, as we have seen, are likely to be a considerable proportion of
all transactions, each applies.
4-2 MONEY AND F I N A N C E

I shall therefore insist on regarding the representative transaction,

of sale or purchase, as in principle divisible into three parts, The first
is the contract between, the parties, consisting of a promise to deliver
and a promise to pay (both are needed to make even a constituent
part of a transaction); the second and third consist of actual delivery,
one way and the other. In the case of the spot transaction, all are
simultaneous; but. they do not need to he simultaneous. If there is
any difference in timing, promises precede deliveries; that is the only
rule which applies throughout. Delivery of the article may come
before it is paid for, as in the case of 'consumer credit'; or it may come
after, or partly after, as when the buyer "puts down a deposit1. All of
these are conveniently covered,
What remains, in general, immediately after the making of the
contract, are on the one hand a debt 'in real terms* from the seller
and on the other a debt in money terms from the buyer. Money is
paid for discharge of a debt, when that debt has been expressed in
terms of money. Thus money comes into the transaction in two
ways, first in the part it plays in formation of the contract, then in the
part it plays in paying. Do not these correspond to the classical func-
tions of money, as laid down in textbooks, to be (1) a standard of
value and (2) a means of payment? By taking the representative
transactions in the form proposed, we have put them into their
places in relation to one another.
What however of the third 'function', usually taken to go with
them, of being a store of value? That money, on occasion, can be a
store of value—that, as one used to say, it can be hoarded—is of
course not to be denied. But this is no distinguishing property of
money as such. Any durable and resellable good can be a store of
value. A picture by an Old Master can be a store of value, but no one
would want to say that it was money. Nor can money be dis-
tinguished, along this line of thought, by saying, as Keynes did, that
money is the perfect store of value, that it is the only asset which pos-
sesses perfect liquidity, so that it does not have to bear interest in
order that it should be held. For liquidity in turn cannot be defined,
as we shall have much reason to see, except in terms of exchange-
ability for money. So to delne money as an asset with perfect liquid-
ity is to argue in a circle. It is the other functions of money which are
intrinsic; the liquidity property follows from them.
A fourth 'function', to be a 'standard for deferred payments', has
by the arrangement here adopted already been covered. It is included
 THE N A T U R E OF M O N E Y 43

in 'standard of value' if, as I am supposing, payment is usually more

or less deferred, 'Unit, of account', which has often been, taken to be a
synonym for 'standard', accordingly says much less than what is
needed.
We seem thus to be left with two distinguishing functions of money:
standard of value and medium of payment. Are they independent, or
does one imply the other? It is not easy to see that there can be pay-
ment, of a debt expressed in money, unless money as standard has
already been implied in the debt that is to be paid. So money as means
of payment implies money as standard. But could a debt expressed in
money be discharged otherwise than in money? Surely it could.
It could for instance be set off against another debt, the debt from
A to B being cancelled against a debt from B to A. If the two debts
have arisen from similar transactions, the net result is a barter trans-
action, an exchange of goods with no money changing hands. That
can happen, even if the debts are expressed in money terms; it is
what has in effect happened, in international trade, on many occa-
sions in the present century, particularly in Eastern Europe. There
have been two countries which have run out of supplies of an inter-
nationally acceptable money, but have kept trade going between
them by a more or less successful offsetting of debts. The debts are
expressed in a money which is recognized by each of them but maybe
not by others. Though this is called a barter deal, it is different from
the small-scale swaps that figure in economic textbooks (such as the
'nuts for apples' in Marshall's barter appendix); for these make no
use of money, even for acounting purposes. In the international
barter deals, money remains as a standard, at least as a unit of
account, It is money as a means of payment that is missing.
I will match that fairly recent example with another, also I think
illuminating, from much further away. There have been societies, so
anthropologsts in particular tell us, in which cattle have been used as
money. What is the evidence for this? It is not like the evidence for
coins, where actual coins have come down to us; it is not derived
from bones of cattle that have been dug up. It is derived from what
are in essence legal prescriptions, expressed either in written docu-
ments or in oral tradition, which set out the fines or compensations
which are to be paid on particular occasions, as for offences of vari-
ous types. If these are expressed in terms of cattle, it need not be sup-
posed that they had always to be paid in cattle. The prescriptions are
price-lists; they depend upon a notion of what things are worth. The
44 MONEY AND FINANCE

things which were delivered in payment had to have recognized, or

at least acceptable, values. (And values, it should be noticed, which
were fairly unchanging over time,1)
Indeed, as both of these examples show, the function of money as a
standard, if it is no more than a standard, is to make it possible to
form a price-list, in which the values of a number .of commodities are
reduced to a common measure. Without its help, there would be a
distinct price-ratio between each pair of commodities, and these
would not need to be consistent with one another. And that is a need
which (as the cattle example shows) can arise without arising from
trading; and (as the other example shows) can also arise from
trading between two parties, none others being, even indirectly, con-
cerned. It does indeed seem proper to lay down that barter, in the
narrow or even in the wider sense, is appropriate only for bilateral
trading,
Bilateral trading, as every economist knows, is an inefficient way
of trading; it is at the gateway to multilateral trading that we come to
money as means of payment. One might indeed conceivably con-
struct a model in which the effect of multilateral trading was
achieved through a sequence, or circle, of bilateral barter trans-
actions. But it would be a ¥ery artificial model and we may be sure
that if anything like it was ever achieved in practice, it would soon
break up. For, as we have seen in Chapter 2, a competitive multilateral
system depends on the activity of intermediaries, or merchants who
are ready either to buy or to sell. Here, as in other activities, there are
gains to be got from specialization; so we may think of the individual
merchant coming to trade to a particular line of goods. These are the
things he buys and sells; but for what is he to sell them? From whom
and to whom is he to buy and then sell? It cannot always be mer-
chants who have exactly the same specialism as he has, for that
would get him nowhere. It must in the first place be merchants who
have other specialisms, though the whole body of merchants will
have dealings 'outside'. At least tor dealings between merchants, a
medium of payment is needed which is not a speciality, something
which is acceptable by a merchant just because he is a merchant, so
1
So the practice seems to depend upon some notion of a justum prttium, a proper or
normal price. This is a notion which indeed is comfortable, still remains comfortable,
In a legal environment. It is congenial to a lawyer to take prices from precedent, since
that is what he does with other things. It is hard for him to accept the fluid prices
which are formed on markets.
 THE N A T U R E OF M O N E Y 45

it must be something which can be readily passed on to a trader of

any sort. That the precious metals, gold and silver, should have been
found to be the most suitable commodities for this purpose depends
on physical characteristics which are set out in all the old gold stand-
ard textbooks; they need not be repeated here. What is important is
that they were surely able to establish themselves through 'market
forces'; no one had to order that they should be used in that way.
There was nevertheless a most important step, on the way to the
establishment of a metallic medium of payment, which had still to be
taken: the invention of coinage, which appears to be traceable to
lands of Greek culture, about 650 BC. A coin is a piece of metal that
has been stamped by the issuer; by the stamp it is guaranteed. The
guarantee was in the first place one of weight and fineness, of quant-
ity and of quality. In its absence the metal would have to be tested, in
ways that were bound to be expensive, almost every time that it was
used as money. That would have greatly impeded the use that could
be made of it. (We do however have evidence that, at least some-
times, transactions were conducted in this way; for hoards of pieces
of uncoined silver have been discovered—archaeologists call them
'dumps'.} Coinage was a great step forward from that.
The stamp, in practice, has nearly always taken the form of an
image, or emblem, of some ruler; the guarantee that is given is a state
guarantee. How did that come about? Did it have to be a state guar-
antee? It had tobe given bysomeone, andtherewould seem to have
been only three alternatives: it might be given by one of the mer-
chants, it might be given by some sort of association set up by mer-
chants, or it might be given by the government in whose territory the
merchants were working. One can see that the second of these, if it
were available, would be better than the first, since the circle of
people who might be expected to have faith in the guarantee would
be wider; and the third, again if it were available, should for the same
reason be better than the second. So it is not surprising to find that it
was the third which won out.
But the fact that a guarantee was given did not mean that it was
always to be relied on; one does not get the impression that the kings
of olden times were a reliable lot. So it was that in practice metallic
money had many adventures. They make quite a story; it is eco-
nomically interesting, but I shall not pursue it here/ It is sufficient to
2
I hai'e said most of what I have to say about it in my Theory of Economic History
(1969), pp. 64-8. The chief thing which emerged from that discussion is that there
46 MONEY AND F I N A N C E

emphasize that metallic money, if it was to be usable, depended on a

guarantee. In that respect it does not differ so much from paper
money as is often supposed,
I shall instead follow up another development, already implied in
the foregoing, from which, as we shall see, modern moneys were
Indeed to be evolved.

can be no assurance that a guarantee will be kept if the guarantor has a monopoly
position; the effort, often made by rulers, to prevent the export of the precious tnetab
was an effort to protect their monopoly. So it was that governments tended to be better
behaved in this monetary matter when external relations were of major importance to
them, or to the peoples over whom they ruled. This holds for the mercantile republics
of Venice and Holland, and came to hold for England also. What appears to be a strik-
ing exception to this rule, the centuries-long stability of the gold coinage of the Byzan-
tine Empire, may be less of an CKception than it looks; for it would be explained by the
dependence of those emperors on mercenary soldiers, coming from abroad and return-
ing, (ft appears that many of the Saxon army, defeated by William at Hastings, took
service at Constantinople.)
 6 The Market Makes its Money

We have seen that one way in which a debt can be discharged is to

set it off against another debt. Debt is then 'paid' with debt. If there is
a perfect match—the two debts, expressed in a common standard,
being exactly equal—the net effect, as we saw, is a 'barter deal'. But
there could fail to be a perfect match, yet payment by exchange of
debts could still be feasible, if another debt could be brought in.
This would have to be a debt from some third party (C) other than
the A and the B initially concerned. A is then asked to accept part-
payment in the form of a debt from C to B, which is to offset the
balance of debt between A and B, a balance we take to be in favour of
A. But A can hardly be expected to consent to such an arrangement
unless he considers that C is to be trusted. So there is a question of
trust, or confidence, as soon as a third party is brought in.
But may not such a question arise even in bilateral trading? A is
selling to B; each has promised to deliver; a time comes when A has
delivered but B has not yet paid. It is understood that B has some
time allowed him before he is obliged by the contract to pay; but it
may happen that this time has elapsed and still he has not done so.
How is he to be made to pay? The legal answer is that A then has the
right to take back what he has delivered. But that (though, as we
shall see, it has a part to play in the story) can easily fail to be an
effective sanction; B may have hidden the stuff away, or may have
consumed it. Nevertheless, if the transaction is not an isolated trans-
action, but is part of a continuing business, there is another and
often a better remedy: if one party to the trading defaults, the trade is
unlikely to continue. That may be enough if the trading is bilateral;
but if a third party, not concerned in that trade, is introduced, it
cannot work. So the issue of conidence is chiefly one of multilateral
trading.
In the standard economist's model of multilateral trading (the
n-good m-person market of Walras) it is avoided; for there all
transactions are spot transactions, taking place-—somehow!—
simultaneously. But if payments are made by offsetting of debts, and
the debts are owing from different people, it cannot be taken for
granted that all will be paid, or will be paid exactly when promised;
48 MONEY AND F I N A N C E

so the debts may well be of different quality. That need not prevent
the establishment of a market in debts, a debt of low quality becom-
ing exchangeable for one of higher quality at a discount. It follows
that a trader, whose promises are judged by the market to be of poor
quality, cannot get as much for his promises as he could if his prom-
ises were better regarded. So he has an incentive to improve the
quality of his promises.
The quality of a debt from a particular trader depends on his
reputation; it will regularly be assessed more highly by those who are
in the habit of dealing with him, and know that he is accustomed to
keep his promises, than by those who do not have the advantage of
this information. Thus we may think of each trader as having a circle
of traders around him, who have a high degree of confidence in him,
so that they are ready to accept his promises at fuli face value or near
it; there is no obstacle to offsetting of debts within that circle from
lack of confidence in promises being performed. If he wants to make
purchases outside Ms circle he will not be so well placed. Circles how-
ever may overlap; though C is outside A's circle, he may be within
the circle of D, who himself is inside the circle of A. Then though A
would not accept a debt from C if offered directly, he may be brought
to accept it if it is guaranteed by D, whom he knows. D is then per-
forming a service to A, for which, he may be expected to charge. A
would have to pay more for a guarantee from a trader who is
"further* from him; but he should often be able to get it at a reason-
able price from some who are 'near'.1
We can recognize the market on which such prices are established
as a market for acceptances of falls of exchange, I am taking that as the
fkst of financial markets to be considered, not only for the historical
reason that it is the fkst which we know to have flourished, but also
because one can explain why it had to come first. Unlike the more
familiar financial markets which will be shown to follow after it, it
needs no specialists in financial dealings (bankers or even brokers)
for it to work. It can come into existence through dealings between
merchants (who may indeed be specialized in dealings on a particu-
lar line of goods, but are not specialized financiers); it can come
about, without any particular attention being paid to it, in the ordin-
ary course of trade,
Let us accordingly take that as the beginning and see what follows
1
The mathematical reader, if there are such, may enjoy the parallel with Ms con-
cept of analytical continuation in his theory of functions of a complex variable!
 THE M A R K E T M A K E S ITS M O N E Y 49

from it. We should be thinking of a fully monetized economy, which

includes a sector of merchants, who use bills as media of payment
between each other, while the rest of the economy uses cash, pre-
sumably coins. Then let the mercantile sector get large enough to de-
velop opportunities for division of labour, on the famous principle of
Adam Smith, There are at least two sorts of financial operators who
should then begin to appear. One works within the mercantile sector,
the other on the frontier between it and the rest.
The first are just intermediaries, with the regular function of inter-
mediaries, in the market for bills. We should think of the mercantile
sector as being made up of many, only partially overlapping circles,
so that, in order to get the best value for a particular bill, a fairly
roundabout route has often to be found. It is the business of this first
kind of intermediary to find that route, getting a sequence of guaran-
tees, as cheaply as possible.
The other kind of intermediation, which has more of a future
before it, is the discounting of bills for cash. Any bill has a date of
maturity, so it can (if it is honoured) be turned into cash simply by
waiting. But the dates at which a trader finds himself in need of cash,
to make purchases outside the mercantile sector, are unlikely to have
a perfect match with the bills he happens to hold. So there is a need
for intermediaries, between the bill market and the rest of the
economy. They can only operate if they hold stocks, both of bills and
of cash. Some at least must be doing so, so that when any one of
them runs out of cash, he can replenish his stock of cash by selling to
others.
For this to be easily and quickly possible, the quality of the bills he
holds must be high; there must be no question of lack of confidence
in them, no fear of default. So the effect of this second kind of inter-
mediation is for the bill market to develop a 'core*, consisting of
'prime' bills, as it should be appropriate to call them—bills which are
such that there is no question of lack of reliability. That is a point at
which most important things happen.
Until that point, the principal reason why the market value of one
bill should differ from another is difference in reliability; but bills,
between which no difference in reliability is perceived, may still differ
in maturity. A trader who is in need of cash needs it now, not (say)
six months hence. So there is a discount on a prime bill which is a
pure matter of time-preference—a pure rate of interest
I have chosen what may be thought to be this unnecessarily com-
5O M O N E Y AND F I N A N C E

plicated way of introducing interest, because there is a reason why a

simpler approach would not do. It is tempting to say that financial
transactions are always, in some or other, loans; so the sim-
plest form of loan contract—money being paid over now, in return
for a promise of repayment, with interest, at some future date or
dates—is the element from which we should start. One could start
that way, and go on to admit that the amounts, and dates, of repay-
ment may be not fixed but conditional on things that may happen in
the future; so proceeding to insurance contracts, subscription to
equities, and so on. Much of the matter we shall be proceeding to dis-
cuss could be reached if one started that way.
The trouble is that the establishment of a competitive market for
simple lending is not at all a simple matter. The lender is paying spot,
for a promise the execution of which is, by definition, in the future.
Some degree of confidence in the borrower's creditworthiness—not
just his intention to pay, but his ability to pay, as it will be in the
future—is thus essential to it. There cannot be a competitive market
for loans without sonic of this assurance.2
We may suppose, in accordance with, what was said at the begin-
ning of this chapter, that any particular potential lender will have a
circle of potential borrowers around him, whom he knows, and feels
that he can trust. We can conceive that there will be competition
between these borrowers for loans from him. And we can imagine
that a particular borrower might be so fortunate as to belong to the
circles of several lenders, so that he can choose between them. But
for both of these conditions to be satisied, without some further com-
plication of the story, looks most unlikely. I accordingly maintain
that a necessary, or nearly necessary, condition for the existence of a
competitive market for loans is that there should be intermediaries,
such as, in our discussion of competitive markets for commodities,
we found ourselves obliged to introduce.3

2
The rural money-lenders, who so obviously do not have confidence in the credit-
worthiness of those to whom they lend, who therefore charge usurious rates of inter-
est, in order to have a prospect of profit to spite of their expectation that many of those
to whom they lend will default, are of course a well-known phenomenon. But they do
not form a competitive market. Their clients accept their terms because they have no
choice,
* It may be that some of my readers, having personal experience of the way in
which, at the present day, a bank will offer loans, to such people as students, with
hardly any security, will doubt whether my emphasis on trust is not overdone. Why
does the bank not charge such people a much higher rate than that at which it usually
 THE M A R K E T M A K E S ITS M O N E Y 5!

Let us accordingly go back to our bills. The simplest model, on that

approach, is the model we were on the point of constructing—an
economy consisting of (1) a mercantile or commercial sector, which
uses bills as means of payment among its members, and (2) an out-
side sector, which uses cash. Let us further, to sharpen the issue,
admit that the bill-using sector has a complete system of guarantee-
ing bills, along the lines described, so that all the bills it uses are fully
reliable. There will still, as we saw, be a need for a special class of
dealer who will discount bills for cash. But has not the model then
settled into a familiar form, these dealers being similar to dealers in
foreign exchange? 'Inside* and 'outside' are like two countries, each
having its own money. The determination of the rate of interest, or
discount, on the bills is equivalent to the determination of a rate of
exchange.
We have learned from experience, though it has not been easy to
learn it, that the rate of exchange between two currencies, though it
is affected by the current balance of payments between the countries
which use them, is also affected by speculative 'capital movements',
which are sensitive to expectations of the future course of the
exchange rate. So it should be here. Consider the position of the
exchange dealers, on the boundary between the sectors, who make it
their business to trade bills for cash. Changes in their holdings of bills
(taking the whole subsector of the exchange dealers together) come
about in two ways; first on the initiative of traders who are not
exchange dealers, whose net demand for cash will rise or fall accord-
ing to the balance of their trade with the other sector; and secondly
on the initiative of exchange dealers themselves, because of changes
in their relative willingness to hold bills or cash. We shall find that
this distinction runs right through the theory of interest. 'Classical'
theorists looked only at the one, Keynesians only at the other. For a
proper theory of interest, neither should be forgotten. ! shall have
much more to say about this in Chapter 9.
It may however already at this stage be objected: is there not a fun-
damental difference between the market for foreign exchange and
our market for bills? The former, if it is a freely competitive market,
may surely establish the rate of exchange at any level, high or low;
but if our bill market is to be used as an approach to the study of
lends, on the usurer's principle? Surely because it hopes to persuade the young person
to become a regular cSient The transaction is steiiar to the ofa" of a free sarapte by a
manufacturer. The price of arc at which that is offered is not in itself a market price.
52 M O N E ¥ AND F I N A N C E

actual bill markets, or 'money markets', it needs to incorporate a

reason why bills, in practice, nearly always stand at a discount in
terms of cash, the rate of interest on them being positive. A sufficient
reason, within our model, might perhaps be found in the considera-
tion that bills are only acceptable within the mercantile sector, while
cash is acceptable within that sector and also outside. So whether the
mercantile sector is large or small, cash must always have a wider
acceptability.* But it is probably more fundamental that cash is a
standard of value as well as a means of payment, so it is fully money;
it is the standard in terms of which contracts are expressed and
enforced at law; bills, being only a means of payment, are no more
than quasi-money. The discount is the expression, by the market, of
this inferiority.
The purchaser of a bill is, in effect, making a loan to the issuer; he
is willing to lend, in this form, because he is assured, and those who
have guaranteed the bill are also assured, that the loan, when the
time comes, will be repaid. Bills have usually run for quite short
periods, at the most for a year or so; it is easy to see that such a
method of inance is peculiarly suitable for commercial enterprises,
the capital employed in which is turned over quite fast. The lender
has just to wait until the 'ship comes home*.5 Even before that
happens, the bill is represented by the cargo, or some part of it, so
that the lender can think of himself as entitled to something more
than a promise; indeed, as we saw, he has something against which
he can exercise a legal claim. But it is surely the fact that the bill is
guaranteed by people who are known to the lender, people who are
within his 'circle*, that gives him better security,
One can see that there would be people, not within that mercantile
sector, who would want to borrow (and possibly, though perhaps
not so obviously) to lend. Some of those wanting to borrow would be
private people, often no doubt quite wasteful borrowers, just wanting
to 'anticipate' an expected inheritance;* more importantly there
* I think this is not upset by the point, which is often noted by historians, that it
may be safer to hold bills, in transit from buyer to seller, since cash is more easily
stolen. (A thief, or highwayman, will not find it easy to cash a Mil that has come into
his possession.) That is indeed a consideration which must have facilitated the growth
of a bill market; but one must conclude from what happened that it did not outweigh
the others.
* Problems of Insurance, as the history also shows, are almost from the first in-
volved,
*'The long-expected death of some old lady . , , Who has kept us youth waiting too
long already' (Byron).
 THE M A R K E T M A K E S ITS M O N E Y 53

would be a demand for loans by the government itself. This would be

typically a demand to meet emergencies (wars and other disturb-
ances), expenditure on which it is nearly always hoped at the start
will not be long continued; so the ruler needs funds to tide him over,
in just the same way as a merchant needs them. So he would like to
turn to the bill, or 'money' market. Kings, as a matter of history,
have often attempted to get loans on these terms; but there have
been two obstacles in their way, each arising from difficulty in pro-
viding a credible assurance of repayment
One was that while the creditor of a private debt could take legal
action to recover from a defaulter (and this, though as we haveseen
it was not very effective, was nevertheless some protection), it was
harder to use the king's courts of law to recover from the king him-
self. This, though intended to be a protection to the king, actually
made it harder for him to borrow. The other was that to cover ex-
penditure by raising a loan, to be paid back later, was bound to set
the prospective lender to worry: if he cannot get the money now,
otherwise than by borrowing, why should he be able to get it when
the time comes to repay? ft was by finding ways round these
obstacles that obligations of the state became 'gilt-edged*.
It so happened, in English history, that ways around them were
discovered, more or less simultaneously. The Bank of England and
National Debt were founded, together, in 1693-4. A National Bank
was the answer to one of the difficulties: to borrow long was the
answer to the other. But as the experience of other countries shows,
the two do not need to go together,7 Each requires particular con-
ditions for it to work, but the requirements are different,
A National Bank, which need not be a Central Bank (the Bank of
England can hardly be reckoned to have been a Central Bank for the
first century of its existence) is an intermediary between the govern-
ment and potential lenders, themselves most conveniently being the
rest of a banking system. Since it is legally separate from the govern-
ment (though it may be owned by the government) its debts are com-
mercial debts, which in principle are subject to legal action. The
government however in a sense stands behind them; so what this in
effect amounts to is a way by which the legal privilege of the govern-

' The story in the United States (essentially no doubt because it was the States,
rather than the Federal Government, who at first were the needy borrowers) has been
notoriously very different.
54 MONEY AND F I N A N C E

ment as a debtor is indirectly waived. There is much that follows

from that which I shaE be discussing in Chapter 9.
In the absence of such an arrangement, short-term borrowing by
government must be difficult, for the other reason; it will be taken for
granted by a lender that when the time for repayment comes, the
government will have no way out but to reborrow, so the trouble will
start all over again. That can be avoided if it is faced from the start, if
the lender engages himself to relend, that is to say, if he agrees to
lend long. The promise is then more credible, since it should be easier
for the borrower to repay in the form of a moderate, though continu-
ing, interest payment, than to repay the capital sum all at once. That
there have long been people who are willing to lend on those terms
seems to be shown by experience; but it does not seem to be inevit-
able that there should be plenty of them. That can hardly be taken
for granted.
How far it is the banking system which has come to the rescue is
one of the things which will be considered in the following chapter.
 7 Banks and Bank Money

What Is a bank? This is a question which has lately become quite

topical; is one sort of business or another to have the right to call
itself a bank? But this is because of the rights and duties which have
been conferred on banks by legislation; for the purpose of an enquiry
such as the present, these may at first be disregarded. For surely
banks existed before there were any such regulations.
So we must define a bank as a firm which does banking business.
But what is that? There is one kind of near-banking business with
which we are here already familiar—that of the exchange dealers on
the edge of the bill market who discount bills for cash. As we saw,
this amounts to making loans to the issuers of those bills. We have
been thinking of the promise expressed in the bill being credible,
mainly because it has been guaranteed by a number of merchants,
but also because it has arisen out of a sale of goods, which in prin-
ciple can be reclaimed if the buyer does not pay. We have seen that
borrowing would be more difficult if the borrower could not give
something of that assurance.
What then is to happen if trade expands, so that more bills are
drawn, and more come in to be discounted? Where is the extra cash
that is needed to coone from? Any one of the dealers could get more
cash by getting other dealers to discount bills that he holds. But the
whole body of dealers could not get more in that way. They must get
cash from outside the market; they must themselves become
borrowers. But what is the assurance which they can give, if they
confine themselves to the business so far described, to the outsiders
who are to lend to them?
The solution was to combine this business with another sort of
business, which in the days of metallic money we know to have
already made its appearance.
In excluding 'store of value' from definition of money, I did not of
course mean to deny that money, any sort of money, could be
hoarded. It would be quite rational to hoard it as a reserve against
emergencies—the 'precautionary motive' of Keynes. But hoarding of
gold or silver would not have been a simple matter. There would
always have been a problem of keeping it safe from theft or pillage.
S& MONEY AND F I N A N C E

and yet accessible to the owner, so that he could lay hand on it

whenever he wanted it. The obvious solution was to entrust it to a
custodian, who could make expert arrangements for looking after it
He would be involved in expenses, of strong-rooms and guards to
watch over them; so if the deposit was a commercial transaction, he
would have to be paid for his services. If the deposit is looked at as a
loan (and it is very like a loan) it carries negative interest. But that is
not the way in which at first it is likely to be looked at.1 It will not be
looked at like that until custody has become a regular business.z
Then, once that happens, there will be a clear incentive to bring
together the two activities—lending to the market, and 'borrowing'
as custodian from the general public—for the second provides the
funds which in the first are needed. At that point the combined con-
cern will indeed have been becoming a bank.5
But it will not have quite got there even yet. For there is a further
step, what looks like being a risky step, which it is almost bound to be
tempted to take. The funds which had become available to it could be
more, even much more, than it could use for its business on the bill
market; why not look for other borrowers? Borrowers outside the bill
market could not give that market's kinds of assurance; but surely
there would be some who look like being reliable. We certainly find
that the earliest banks, which merit that description, were doing at
least some outside lending.*
So I shall. I hope acceptably, reckon a firm to be a bank, a fully
formed bank, when it is doing all these things: (1) accepting deposits,
(2) discounting bills, and (3) making advances to customers. I have
tried to show how these could have come together. But what of the
fourth function, commonly attributed to a bank, that of providing a
medium of payment; how does that fit in? Let us see.
If a bank, as so far described, is to extend its business, it must in-
crease its lending, in the one form or the other; and when it has

1
The leading custodians, in ancient times, would probably have been temples, or
other religious foundations. To put your treasure in the care of a god would have been
a prime way of keeping it safe. But this would not be thought of as a commercial trans-
action. It would have been mixed up with outright gifts.
2
It survives as such to the present day, as when a bank makes a charge for keeping
a small account.
* Custody is sometimes described as 'cloak-room banking". But it surely makes for
clarity to regard it as no more than a step on the way to banking.
4
I am thinking of early banks in Renaissance Italy.
 BANKS AND BANK MONEY 57

exhausted the funds which have been entrusted to it for safe keeping
(and any perhaps which are in its own possession) it cannot go
further without increasing its deposits. Thus it has an incentive to
encourage deposits. There are two main ways in which this can be
done.
One is to offer a (positive) rate of interest on deposits. The interest it
pays must be less than what it earns on advances, or it could not
make a profit. Here the bank is acting as an intermediary on the loan
market, between those who lend to it and those who borrow from it.
This is expensive to the bank, but competition will often ensure that
it has to be done.
The other is to make it easier for depositors to make use of the
funds which they have deposited. They have been thinking that their
deposits were available to be called upon when needed, character-
istically to pay a debt. If this meant that cash (gold or silver) had to be
taken out of the bank, and then posted to the creditor, the safe keep-
ing (which was the purpose of the exercise) would be most im-
perfectly achieved, since the package could get lost or stolen on the
way. It would however always have happened that when cash was
deposited in the bank, some form of receipt would be given by the
bank. If the receipt were made transferable, it could itself be used in
payment of the debt, and that should be safer. But for this to become
a general practice, the bank must co-operate. It must issue receipts
in standard amounts (bank notes). It would indeed be necessary
that the creditor should have confidence in the bank, so that he
accepts the bank's promise to pay as being as good as money. There
might at first be sufficient confidence for this only within a narrow
circle.
Nevertheless, as time went on, the circle could widen. The bank
notes could become a quasi-money, in rather general use. (Historic-
ally, when that point was reached, the government could begin to be
interested, and could put restrictions on bank-note circulation.) Even
apart from that, the more widely acceptable the bank notes are, the
more tempting it is to steal them. So the bank-note device, intended
as a protection, would defeat itself. A further protection was there-
fore required.
This was found in getting the bank itself to make the transfer—a
device which in the end became payment by cheque. It would at first
be necessary for the payer to give an order to his bank, then to notify
the payee that he had done so, then for the payee to collect from the
58 MONEY AND F I N A N C E

bank. Later it was discovered that so much correspondence was not

needed, A single document, sent by debtor to creditor, instructing the
creditor to collect from the bank, would suffice. It would be the
bank's business to inform the creditor whether or not the instruction
was accepted, whether (that is) the debtor had enough in his account
in the bank to be able to pay. In most business dealings the debtor
would have looked after that before drawing his cheque. But if he
had overdrawn, the bank would inform both parties that the cheque
was ineffective, so no payment had been made.
It is easy to see why this has become so common a way of making
payments, at least in an economy where most people have bank
accounts, for it is a superior way of minimizing transaction costs. But
the consequences of its general adoption are notable. For it means
that the whole of the bank deposits which are withdrawable at sight
become usable as money. They are usable as such by the depositors
in the bank, and—what is even more remarkable—they are usable
as money by the bank itself. It is true that they are not a store of
value for the bank, since they figure on the liabilities side of its
balance-sheet, not on the assets side. But they can be used by the
bank itself as a medium of payment.
When the bank makes a loan it hands over money, getting a state-
ment of debt (bill, bond or other security) in return. The money
might be taken from cash which the bank had been holding, and in
the early days of banking that may often have happened. But it could
be all the same to the borrower if what he received was a withdraw-
able deposit in the bank itself. The bank deposit is money from his
point of view, so from his point of view there is nothing special about
the transaction. But from the point of view of the bank, it has
acquired the security, without giving up any cash; the counterpart,
in its balance-sheet, is an increase in its liabilities. There is expan-
sion, from its point of view, on each side of its balance-sheet. But
from the point of view of the rest of the economy, the bank has
'created' money. This is not to be denied.
But before concluding at once, as many do, that this increase in the
'quantity of money' is inherently inflationary, or 'dis-deflationary',
we should further examine the effect on the bank itself.
We have seen that the bank can be regarded as an intermediary,
between those (depositors) who lend to it and those who borrow
from it. The lenders are to be attracted by facility of withdrawal; but
what corresponds to that on the borrowing side? An outside
 B A N K S AND BANK MONEY 59

borrower, who wants money now, will usually* want it in order to

spend it; but however profitably it is that he spends it, he cannot
expect to be able to repay until some time has elapsed. Thus the bulk
of the advances that are made by the bank will have to be for appre-
ciable periods. The banker cannot expect to be able to recall his ad-
vances just when he wishes to do so. He may try to arrange them so
that repayments are coming in fairly steadily; but that does not
change the essential point that the money he has advanced will not
come back until some date in the future, which he has accepted in
the past but cannot now be changed.
If deposits are withdrawable on demand, or at short notice, while
advances are relatively immovable, the position of the bank is inher-
ently risky. It must always be exposed to some danger of a 'run'—
many withdrawals coming together.
There seem to be three main ways in which a bank can protect
itself against these risks, risks which are inherent in the kind of busi-
ness it is doing.
One is to take advantage of the 'law of large numbers'. There must
probably be something of this if banking, as a continuing business, is
to function at all; but the protection which it offers does not by itself
extend very far. For all that is said by this statistical principle, applied
to banking, Is that when a large number of similar transactions are
being undertaken, in each of which there is a chance of some kind of
failure, but the risk in one is independent of that in another, the loss
that needs to be allowed for over the whole, when that is taken to-
gether, should be fairly predictable. This applies on both sides of the
business of the bank. In the case of advances, failure consists in the
borrower failing to repay at the appointed date; that can be looked
after, on the statistical principle, by making a provision for bad debts.
In the case of deposits, the risk that is undertaken by the banker is
uncertainty of date of withdrawal; that also can be spread, if there
are many depositors, and what makes one withdraw does not affect
the behaviour of others. There have nevertheless been important
cases when independence, on one side or the other, has been counted
on but has failed. If the customers, who receive the advances, are
most of them doing the same sort of business, when one is in trouble
5
Usually, because in practice there is an exception.. He may borrow, although he
does not plan to spend until a iater date, if he thinks he can get the money more easily,
or on more favourable terms, than he would be abte to get it iater. 1 shall leave this
aside for the present
60 MONEY AND F I N A N C E

many others may be also; thus a bank which is specialized on lending

to farmers, itself gets into trouble in an agricultural depression. With-
drawals by depositors, who have begun to suspect that the bank may
not be able to pay, are very likely to be imitated by others. So on each,
side there are possible conditions when the statistical protection does
not work,
A second way in which banks have commonly protected them-
selves is to avoid allowing too much of the funds entrusted to them to
be tied up in advances. Some may be held in the form of cash; but
even if no interest is being paid on deposits, to hold a money, which
bears no interest, as corresponding asset is clearly unprofitable. Bills
are obviously a better alternative; and something of the same
advantage can be got on suitable occasions, from longer-dated secur-
ities also. They can be expected to be sellable in an emergency,
though the price at which they can then be sold is uncertain.
A third recourse, which to modern times has become of major im-
portance, is to borrow from another bank. If there exists a group of
banks, which are prepared when called on to lend to one another,
the group is stronger than any of its constituents would be by itself.
Strength is needed; so a certain amount of association of this sort has
an economic function. By those who stand for competition 'though
the heavens fall' it is under suspicion; but the virtues of competition,
in cases where failure has wide repercussions, are open to qualifica-
tion. How far this dilemma is resolved by the creation of a Central
Bank—in fact, even if not in name, a Government Bank—in which
the monopoly element is concentrated, I shall be considering in
Chapter 11.
It will be observed that of these three protections, it is only the
second which has the quality that the extent to which it is used can
be continuously under the control of the bank itself. To get a loan
from another bank requires the consent of that other bank; to vary
the independencies, or interdependencies, between the risks involved
in its advances, or in its deposits, can only be matter of long-term
policy. But it is open to the bank at any moment to vary the size of its
cash holdings, by buying or selling securities. It is therefore inevit-
able that operation upon this margin should be central to the man-
agement of the bank.
It is true that its advances will be 'rolling over'; some, at any time,
will be being repaid, and (normally) being replaced by new advances,
or continuations. Thus one way in which a bank may replenish its
 BANKS AND B A N K MONEY 6l

cash holdings is by cutting down on replacements. But the most that

can be got from this source, at all quickly, will be limited. Advances
are not so liquid as investments (in securities) are.
So this is where we come to the concept of liquidity, and here we
have it in the banking context where it first appeared in the work of
Keynes,* Bankers, he told us in his Treatise, have
three categories [of assets] to choose from: (1) bills of exchange and call loans
to the money market, (2) investments, and (3) advances to customers. As a
rule, advances to customers are more profitable than investments, and
investments are more profitable than bills and call loans; but that order is
not invariable. On the other hand, bills and call loans are more liquid than
investments, i.e. more certainly realizable at short notice without loss, and
investments are more liquid than advances.
I regard this passage as extremely important, not merely because it is
the first place where Keynes spoke of liquidity (and it may also be the
first place where any economic or financial writer spoke of liquidity7)
but also because it is better than the simplified version of which
Keynes himself made use in his later work. But it has not had the
impact it deserved, for it needs some explanation and working-out.
First, explanations. 'Realizable' means convertible Into cash, or
money; but why money? Because it is in money that its liabilities—
particularly its sight liabilities, deposits withdrawable on demand—
are expressed.8 Thus cash, that is held among its assets, would seem
to be treatable as perfectly liquid. Advances we have been reckoning
not to be cashable at all "at short notice'. Thus until the time for
repayment comes, they are completely illiquid; their liquidity is zero.
Degrees of liquidity can only pertain, in the case of the bank, to the
securities segment of its assets (bills, bonds or maybe equities). It is
only these which can be more or less liquid.
When liquidity is defined in this manner, it becomes clear that it is

* Treatise on Money, Volume 2, p. 67.

' I maintained this priority, in a paper entitled Liquidity, published in the Economic
journal (1962) and reprinted in my Collected Essays, Volume II (1983), pp. 238-47. It
is difficult to prove it.I can only claim that I looked for it in a number of writings of the
twenties, where it would have been appropriate for the writer to have used It, and did
not find it. No one has told me I was wrong.
8
if there is more than one sort of money to which its liabilities are expressed, such
as its 'own' money and a foreign money, and there is no fixed rate of exchange
between them, things become more complicated. I shall have a little to say about that
very contemporary problem later {see Chapter 14).
§2 MONEY AND F I N A N C E

a quality which is attributed to an asset, according to a judgement

that is made by its proprietor, or by some other interested party, at a
particular moment. The loss', of which the definition speaks, must
be the difference between the current market price of the asset and
what it might fetch if it were to be disposed of at an unfavourable
moment. (The need for such disposal might arise on the liabilities
side, such as on withdrawal of deposits, or on the asset side, for
example a new opportunity for profitable lending in the form of ad-
vances.) just when it would be desired to make such a disposal is
unknown.
Let us now look again, in the light of these considerations, at the
balance-sheet of the bank. Its 'investments' we are now to reckon as
"more or less liquid'; its advances as nearly illiquid (at least so far as
the near future is concerned); its holding of cash, on the principle just
explained, as perfectly liquid; but is that right? Most, and sometimes
even all, of its cash is normally employed on its regular business,
covering gaps between deposits and withdrawals; these go on all the
time, without creating any "emergency". It needs to have a money
holding for this purpose, but this is not a liquid asset, from the bank's
point of view. When this is allowed for, we ought to say that liquidity
is a characteristic of an asset that is held as a reserve. The money that
is held for current transactions is not a reserve asset; it is what cor-
responds to the working capital of a manufacturing business. I find it
convenient to call this a running asset, (Advances also, when it is ex-
pected that they will go on being replaced, or renewed, are in this
sense a running asset.)
In these terms, it can readily be seen what we should mean by the
liquidity of the balance-sheet as a whole. It must be a matter of the
quantity, and quality, of the reserves. This must be measured against
possible calls on those reserves, which are essentially a matter of
withdrawal of deposits. So it is tempting to say that the liquidity of
the bank should be measured by the ratio of its reserves to its
deposits. If there are advances, so that not all deposits are covered by
reserves, then on this measure the bank is always imperfectly liquid.
But liquidity is a matter of quality as well as quantity. Among the
reserves there will be some which have high liquidity, some (per-
haps) very much less. They shade into one another. So though it is
true that banking liquidity is a matter of comparison between re-
serves and deposits, it is not a comparison that can readily be
reduced to an arithmetical ratio. For the liquidity index which is to be
 BANKS AND BANK MONEY 63

attached to a particular security will vary over time and over state of
mind—even the state of mind of the market as a whole,
It is nevertheless understandable that people who make decisions
about liquidity (and people, such as economists, who think about
people making those decisions) should want to work in such terms as
can be put into an arithmetical form. If the 'more or less' liquid assets
could be shepherded into two classes, on the one hand those that are
very liquid, and on the other those that are decidedly illiquid, an
arithmetical comparison between the very liquid and the deposits
would serve as a good proxy for liquidity in general. But such a
separation may not be easy to make, or to maintain. The best place
for making it may shift from one time to another.
This has a bearing on what has happened to one part of the eco-
nomics of Keynes, I have been greatly helped in this chapter by what
he said on liquidity in his Treatise; but in his later and more famous
book he seems to have fallen into the trap Just described. And how
many of his monetarist followers—in this respect they were his
followers!—he led into it. One can see how it happened. During the
years 1932-8 Oust when Keynes was writing his General Theory and
defending it against its first critics) the market rate of discount on
bills, in London, was hardly more than one-half per cent. So bills
were standing at a discount which was practically negligible; to treat
them as being money, as Keynes implicitly did, was very natural. If
bills were money, there was just one margin to be considered, among
the reserves of a bank (or other financier): that between money so
extended and other investments (bonds). So he could show his long-
term rate of interest being determined at that margin. But this was a
state of affairs which did not persist; from the perspective of fifty years
later it appears an aberration. As things have been since the 19 50s,
not only in Britain but in other countries, short rates have been
much higher, and there have been numerous issues with medium
maturities, by governments and others, all the way between the bills
of Keynes's time and his long-term bonds, or the nearest to the latter
which still exist. Where, in this continuum, do we draw a line? it is
no wonder that there has been such a fuss about the sorts of claims
that are to be reckoned as money, Mt and M,, and so on! In what has
become the modern world, there can be no answer to that question,
We have to go back to the qualitative concept of the Treatise.
And it is not only for theory of banking that we need it, I shall be
looking at it in a wider way in the chapter which follows.
 8 Choice among Assets

It was rather obvious that the liquidity concept, which in Keynes's

Treatise (1930)—in the passage quoted in the previous chapter—had
been confined in application to the behaviour of banks, could be rein-
terpreted so as to give it a wider reference. Any sort of financial firm
would be confronted with similar alternatives, and so could have
assets of various degrees of liquidity. One could think of it being faced
with a 'spectrum* of such assets from which to choose, a spectrum
which could be wider or narrower than that on which a bank would
usually work. And having gone so far, why not go further? Why not
look for liquidity elements in the decisions of all sorts of firms, and
even in the management of his property by the private capitalist?
So there were two directions in which the concept might be
generalized—to the decisions of more sorts of choosers and between
more sorts of assets.
This generalization was not made in Keynes's General Theory
(1936); the generality from which that took its title was in a different
direction. The Treatise had been a theory of the influence of money on
prices; its analysis was extended, or transformed, in the later work, to
the influence of money on activity of production also. Nothing like
that happened on the side of liquidity. His new Liquidity Preference
theory was another special case, where choice is restricted to two
assets, money and securities; the latter being interpreted sometimes
as long-terra bonds, sometimes as something which is to stand for
securities in general. So Liquidity Preference became a theory of the
rate of interest. Differences in liquidity between different securities,
on which attention was concentrated in the Treatise definition, have
disappeared from sight. (I have explained the historical circum-
stances which could have led Keynes to suppose that this procedure
was sufficient.)
Nearly a year before the appearance of the Genera/ Theory I had
myself published (in Econotnica, February 1935) my 'Suggestion for
simplifying the theory of money'. This, though it does not use the
word liquidity, was a step in the direction of what was required. I
think it has become quite widely accepted that this is so. It was stated
that the theory should apply to any decision-maker; the spectrum on
 C H O I C E A M O N G ASSETS &5

which he operates is drawn very wide, extending from money at one

end, through more or less liquid securities, to real capital goods at the
other, I am still quite happy with that way of formulating the prob-
lem. There does not seem to be much thai is wrong with what I said
about it,'
Nevertheless there is something which ought to have been there
but is missing. It is something which is in Keynes (of 1936); that of
course at the time I was writing was not at my disposal, It is his
famous classification of motives for holding money—his trans-
actions, precautionary and speculative motives—which should have
found a place. It could have been generalized in just the way I was
generalizing; but it was not until many years later2 that I saw how
that was to be done. It can now very usefully be made to fit.
Keynes's motives were purposes for which money may be held. The
step which had to be taken, corresponding to that taken in my
'Suggestion', was to notice that money is only one of the assets that
may be held for corresponding purposes. We have already seen, in
the case of the bank, that its cash may be a running asset (corres-
ponding to Keynes's transaction motive) or it may be a reserve asset
(corresponding to his precautionary); also that, in the case of the
bank, there are other assets which may be held for just these pur-
poses. It is possible, in the case of the bank, that these two may
exhaust all the purposes for which assets are held: for its advances, if
they are turning over regularly, would count as running assets. For a
bank to be holding cash for a speculative motive must, as we shall
see, be uncommon. But there clearly is another purpose, not so far
classified, for which assets may be held—just to get an income from
them—as a private capitalist, to take the most obvious example, may
hold a portfolio of securities. It has not been easy to find a good name
for this third category of assets. The private capitalist calls them his
'investments': so it probably gives the right impression to call them
Investment assets'. That has been my usual practice, which I shall
here maintain, I am however by no means entirely satisfied with it.1
1
As the spectrum was written out in my "Suggestion", consumption goods appear
at the more liquid end! This was clearly a mistake.
1
First in the third of Two triads' (Critical Essays, 1967), A later version is 'Solidity,
fluidity and liquidity', a new paper put into the second volume of my Collected Essays
11983). I am still quite satisfied with nearly the whole of the latter, so I shall not
abstain front quoting quite a lot of it verbatim.
* It does not fit the usage of the Treatise, where the bank is reckoning what clearly
are reserve assets as 'investments'.
66 MONEY AND F I N A N C E

Some kinds of assets are financial claims; others are property

rights in real goods. This is a different classification from that we
have been making, but both are needed. So what results is a six-way
table:
Running Reserve Investment
Real A B C
Financial D E F
What kinds of asset should we put into each of these boxes? Let us
look for some examples,
We should not expect that in every balance-sheet there would be
instances of all six. A manufacturing firm will have real running
assets (A) in the form, of goods in process, and of plant and machinery
in use. They are running assets because they are in active use, in the
work on wic tirm is engaged. It will probably have some real
reserve (B) in the form of reserve stocks of materials, and per-
haps some equipment, machines and suchlike, which will only come
into use in the event of a need for repairs, or some similar emergency.
It is not easy to see that there is a need for such a §rm to have real
investment assets (C). But there are other entities which clearly have
them. An obvious case, to the present writer, is the land which is
held by some Oxford and Cambridge colleges as an investment. Land-
holding is not the regular business of the college; the land is just held
to get an income from it.
Real assets (B) that are held as reserves are already embodied in
specific forms, so they are only usable as reserves in suitable emer-
gencies. But the kinds of emergency that may arise can never be
exactly foreseen; so a form of reserve that is not so tied down must in
general be required. This is the function of the inancial reserve assets
(E) which we should thus expect to be very widely held. One is indeed
tempted to say that any business, which is not to be in danger of col-
lapse, must have some financial reserves. But this does not. necessar-
ily mean that it must have these reserves as assets, for there is a
possible alternative. As we saw in the case of the banks themselves,
assured borrowing power, probably from a bank, will do as a substi-
tute, This may take the form of an agreed overdraft, but it need not
be so formal. We may perhaps reckon it as an 'invisible asset'. It does
not appear on a company balance-sheet, in the form in which that is
published; it is one of the reasons why the balance-sheet needs to be
'annotated' if it is to be properly informative. The corresponding item
in the balance-sheet of the bank is an Invisible liability'.
 C H O I C E A M O N G ASSETS &7

Financial investment assets (F) have already been mentioned, as a

principal asset of the private capitalist:—or of the pension fund or
charitable trust. It is an essential characteristic of the assets in this
compartment that they are held for their yield, for the interest or divi-
dends that are to be derived from them,
There remain the inancial running assets (D) which are clearly of
major importance for monetary theory. For money, in some sense or
other, in the till, in the pocket or in the bank account, is without
doubt a leading financial running asset, for almost any entity. Indeed
there must be many for which it is the only item under D, If a firm
has non-money items under D we should probably reckon it to be to
some degree a dealer in money—a inancial firm.
Nothing has been said directly so far about the speculative motive of
Keynes. If (as Keynes often did) we regard it as an essential charac-
teristic of money that it does not bear interest, it would seem at first
sight that no money should be held under F. For these assets are held
for their yield,-for the income that is expected to be derived from
them, Keynes nevertheless maintained that money could be held
under F for his speculative motive. This was quite a shock to some of
his first readers, who had been thinking that an investor, who
bought a security to order to derive an income from it, would be
intending to hold it indefinitely. If he had that intention, there would
be no point in holding 'barren money'. Keynes however was not
thinking of that sort of investor, but of one who plans to manage his
portfolio, selling and buying again as opportunity offers. There can
surely be no question, when one faces the issue, that both types are
types which occur. We shall need to have names for them; suitable
names have already come to hand in the context of the labour
market in Chapter 4. So I shall say that the one type invests solid, the
other fluid.
There can be no doubt that there will be occasions when a luid in-
vestor will act as a bear speculator. If he expects a fall in the prices of
securities (a rise in rates of interest) he will sell and then, if his ex-
pectation is correct, buy back at a lower price. During the interval he
will have held his funds idle, for a speculative motive. Even if he gets
no interest on the funds held idle during the interval, he still makes a
gain, over the transaction as a whole. Nor is it true, as some of
Keynes's critics objected, that the gain is a capital gain while the loss
is an income loss; and that they are incommensurable. It is only by
tax legislation that they have been made incommensurable; other-
wise they are commensurable.
68 MONEY AND F I N ANCE

Suppose it is expected that the rate of interest on a (long-term)

bond will rise, from 4 to 5 per cent, within a year. The sum of f 100
invested now will yield £4 per annum; if invested later, it looks like
yielding £5 per annum; thus by delay there is a gain of £1 per
annum, in perpetuity, against a loss of £4, at the most, during the
period of delay. Considered purely in income terms this is a very
profitable investment. The barren money, properly accounted for, is
not without yield. So one must accept that there is this case when
money may be held as an investment asset. Nevertheless, though so
much attention has been paid to it, this is not the chief way in which
the solidity-fluidity distinction is important.
The general liquidity theory, of which a first sketch was given in
my 'Suggestion', gains greatly in force when attention is paid to it.
A solid investor, though he may choose his investment with care
at the moment when he makes it, can only do so in the light of what
he knows at that moment. He denies himself the opportunity of
changing his decision, in the light of further information available
later. The fluid investor is not denied that opportunity; this to him is
an advantage. Why should the former deny himself that advantage?
The obvious answer is transaction costs, Every act of investment
(and disinvestment) in securities ordinarily involves some costs—
which may be subjective, the time and trouble involved, but may
also be quite objective, brokerage charges in particular. If a portfolio
is managed in a very fluid manner, these transaction costs must pile
up. They have to be deducted from the gains from fluidity if there is to
be a net advantage of changing over to a fluid policy. It is therefore of
the first importance that there are likely to be great differences in the
transaction costs that are faced by investors of different sorts.
One can assert with some confidence that transaction costs, of
most kinds, increase less than proportionally to the volume of funds
to be invested, at least for large differences in volume. This is shown,
if it needs showing, by the fact that when the portfolio is large, it pays
to have specialized departments, or equivalent organizations, to do
the managing; it does not pay when the portfolio is smaller. Thus it is
essentially the large investor who can profitably pursue a fluid
policy; it does not pay the small investor to do so. It is better for him
to invest at least fairly solid.
This indeed implies that he is investing for income, that what he is
buying is to be an investment asset under F. For his reserve assets he
is more in the position of a bank—in which, in practice, most of his
 CHOICE A M O N G ASSETS 69

reserve is likely to be deposited. Reserve funds are held against emer-

gencies, whose extent and time o( occurrence are not exactly fore-
seen. If the date of requirement is completely uncertain—-it may just
as well be in the near future as further ahead—it will be essential
that the funds should be held in a form which makes them realizable
easily, at any moment. And this does not only mean that they must
be held in securities that are readily marketable; it also means that
the value of the portfolio held for this purpose is not likely to vary too
much over time. So they must, in the sense of the Treatise definition,
be at least fairly liquid assets.
Banking itself, from this point of view, can be regarded as a device
by which outsiders (outside the financial sector) can obtain the
liquidity they require by using the bank as an intermediary. Both by
operating on a larger scale and by being more professional, banks
reduce transaction costs—and so can, at least in principle, operate
more efficiently. From the point of view of the depositor, there is a
substitute for the transaction cost, which he would otherwise have to
bear himself, in a price for its service which he pays to the bank.
So it can be also with respect to investment assets (F). Here also
there is an opportunity for an intermediary.
It is true that the manager of a pure investment fund does not have
to attend to the value of his portfolio, as it may be in the event of
some external emergency, arising at an uncertain date, as the
manager of a reserve fund has to do. But he also needs to be in a posi-
tion to be able to take advantage of opportunities for profitable
investment, which may arise in the future but cannot now be fore-
seen. If he is illiquid, being tied up in securities, the value of which at
an uncertain future date is uncertain, he may find that an opportun-
ity arises, just when the value of his portfolio is abnormally low. He
can then expect to make a (capital) gain just from staying where he
is; that which is promised by the new opportunity must be greater
than this if it is to be acceptable. So he will often have to let it pass.
For this reason, and for the more obvious reason—especially im-
portant in the case of investment in equities, and in foreign securities
affected by rates of exchange—that new information about the pro-
spect of particular securities is continually forthcoming, also by the
manager of an investment fund (F) there is something to be gained
from operating in a fluid manner. Even the small saver, who is saving
to buy himself an income for his retirement, or to leave to his depend-
ants, would gain from fluidity, if he could afford it. If he has no choice
7O M O N E Y AND F I N A N C E

but to invest in securities directly, not being able to afford fluidity, it

is better for him to make his choice and stick to it, thus investing
solid. He may nevertheless be able to gain some of the advantages of
fluidity, by using an intermediary—by depositing or buying shares In
an intermediary, such as an investment trust. Modern financial
systems are rich in such devices: so it may seem at first sight as if
solid investment, except by institutions that are legally bound to it,
must be on the way to disappearing. It is for instance well known
that the proportion of the value of all securities, traded on the
London Stock Exchange, that is held directly by private people has for
long been diminishing. But does this mean that the economist can
take it for granted that an assumption of perfect fluidity is good
enough for his purposes? I think one can show that this is not the
case.
I have insisted (following the Treatise) that liquidity, applied to the
securities sector of a balance-sheet, is a relative concept; money
indeed may be taken to be perfectly liquid (because it is the standard
in which debts are reckoned), other securities however only more or
less. Should not something similar be true of our solidity and fluidity?
Perfect solidity raises no problems; perfect fluidity is what we appear
to have just encountered; but have we any right to it? As we shall
find in the next chapter, it is an instructive hypothesis for some
purely theoretical models; but is it more than that? I do not think it is.
The formal reason is that time is continuous. However short-
sighted speculators may be, however negligible their transaction
costs, they are always speculating with an eye to the future, a future
which is surely always some finite distance away. It is not good
enough to say that tomorrow is future, while today is present; for
that leaves room for an active speculation on differences between
morning and afternoon. However short we make the 'horizon', room
can always be found for something shorter. It should further be
remembered that our transaction costs are partly subjective; even if
objective costs are negligible, there must surely be some of those
operating on the market who would prefer not to have to give their
decision-making sleepless attention,
! accept that modern 'financial innovation' has given stock
markets great fluidity. But if that fluidity was perfect—certainly if it
was always perfect—they would not work in the way we find them
doing, A bull market might go on creeping upwards, with fluid in-
vestors continually making profits, until it had reached a level which
 CHOICE A M O N G ASSETS Jl

anyone with a longer horizon could see straight off was just silly. But
as long as there was a balance of optimist buyers over pessimist
sellers it could go on rising", it would not turn until the balance
turned the other way, AH that is consistent with perfect fluidity, It is
also consistent with perfect fluidity that the fall, once it came, should
be sharp. For everyone would be trying to get out at once. As long as
such fluidity remained dominant, the fall would continue. It must
continue, under perfect iuidity, until all the securities traded, had
become waste paper. But even in the most free and most sophistic-
ated market, that does not happen.
What stops it, without any 'intervention', is that some of the
dealers (granted that they are professional dealers) start to look
further forward, and discover that they can make a very reliable
profit just by holding on. They introduce, in doing so, a little solidity.
It is not enough to prevent the market being very unstable, but it is
enough to prevent complete collapse. It is against this background
that the monetary policy, which I shall be considering in later
chapters, should be thought of as in principle operating.
 9 Theories of Interest:
Keynes versus Marshall

To choose Marshall as leading opponent of Keynes on interest may

seem peculiar, for Marshall had died a whole decade before Liquidity
Preference came into the world. I have nevertheless come to feel sure*
that when Keynes spoke of 'classical' theory It was Marshall's he had
in mind. He had learned and then taught in the school of economics
that Marshall had founded at Cambridge; it was the doctrine that he
himself had been teaching that he was now deliberately casting off,
This was not easily understood by those who did not have that
background. They supposed a 'classic' to mean any earlier influential
economist who was a stranger to Keynes's innovations. This indeed
is the interpretation which has largely survived.1 But by writing
down the part of Marshall in the story something, as we shall see,
has been lost.
It can be recovered once we recognize that there was an aspect of
the 'revolution' which was internal to Cambridge. It led to a split in
the Cambridge school. There were'some of the Cambridge economists
who followed Keynes; but there were others, of whom Robertson was
the most persistent, who remained faithful to Marshall.* They did

1
A footnote is the proper place for my personal part in the story. I have to admit
that I am myself responsible for some of the misinterpretation. My 'Mr Keynes and the
classics' was published in Econometrics la 1937; it was written for econometrists who
were no MarshalHans. So it was that when Keynes saw it, though he found my version
of his own theory fairly acceptable (it is what has later become known as the ISLM
diagram), he insisted that I had got the 'classics' all wrong, My 'classical' was much
more primitive than his 'classical' theory. 1 now regard this as evidence that his
'classic' was Marshall. Though when I wrote that piece I had been teaching for some
months in Cambridge, I was not acclimatized to Cambridge; my background was still
what I had learned in my years at LSE. As it turned out, there were many economists,
not only econometrists, who were in much the same position as I was, so what I said
went down welt with them. It has gone on going well.
' \ was a close friend of Robertson's, so we had many discussions together; but
never in his lifetime (he died in 1963) did we get the matter straight. The first step
forward, in tny own thinking, came in the writing of the chapters on stocks and flows
in my Capital and Growth 11965). But it was only as a result of discussions with that
fervent Robertsonian, Professor S. C. Tsiang, that I was impelled to make the effort to
go over the old ground again. A first sketch of what I shall be saying here was pre-
 THEORIES OF INTEREST: K E Y N E S VERSUS M A R S H A L L 73

have something to say; but they could have put it more convincingly
If they had put it in the form of saying that there was something sig-
nificant in Marshall that had not survived in Keynes. (They could
then have admitted that the same might be true the other way.) That
is what in this chapter I shall try to show,
The formal difference between Keynes's theory and that of the
others is that Keynes looked at stocks—what is needed, at a particu-
lar moment, for a given stock of bonds to be held—while the others
looked at flows, of borrowing and lending, over a period. It will be
useful to begin with the simplest possible model in which the issue
comes up,
This is a model of a closed economy, in which there are Just two
things which are exchangeable for money—one being goods (includ-
ing services), the other bonds, which are (reliable) promises to pay a
regular income, fixed in money terms, to whoever is the holder. The
bonds are homogeneous, and we can make the goods homogeneous,
by supposing that their relative prices are fixed exogenously. So there
are just two prices to be determined, the price of goods and the price
of bonds, the latter being (arithmetically) expressible as a rate of
interest,
So as to give the flow aspect due attention, let us consider the
working of the model during a period. A period must have a start and
it must have a finish, If we consider the start by itself, we must look at
stocks. There will be in this initial position, a certain stock of bonds
and a certain stock of money. The price of the bond must be such
that there are holders who taken together are willing to hold that
particular stock of bonds, neither more nor less, A change in their
holdings of bonds must, in the first place, imply a purchase or sale of
bonds for money. So if we just look at that by itself, it is a change
which must be effected by people who have a choice between holding
money or holding bonds. The clearest case in which there is such a
choice is when the holding of money and the holding of bonds are
directed to much the same purpose; and it is hard to see that this can
be anything else but that which figures in Keynes's theory, a require-
ment for liquidity, for a reserve against emergencies.
It thus appears that if we just look at the start of the period, what

sented to a conterence. organized by him, at Taipei, Taiwan, at the end of 1985,

Through collaboration with him I think I have found a piank, by which the stream
can at last be crossed.
74 MONEY AND F I N A N C E

happens then must be determined by stock relations, which lead to

Liquidity Preference in Keynes's manner, But the flow party will
rightly say that we must not stop at that point; we must go on to see
what happens during the period,
The period indeed has a finish as well as a start; and there seems to
be no reason why the finish, if taken by itself, should not be treated in
the same way as appeared to be appropriate for the start. But the
difference between the position at the end, from what it was at the be-
ginning, must be a matter of" what has happened during the period. If
new bonds have been issued during the period, the stock of bonds at
the end must be different from what it was at the beginning. The
change in the rate of interest during the period must then be a matter
of what is needed for this additional stock to be held.
It would have to be said by a thoroughgoing Keynesian theorist
that these extra bonds would have to be taken up by people who
were willing to add them to their reserves, holding them as a substi-
tute for money reserves which they would otherwise have been hold-
ing. Thus, unless there were new supplies of money coming in, from
outside, during the period, the change in the rate of interest would be
a matter of the elasticity of a Liquidity Preference curve.
A quite consistent stock-flow theory could be constructed on those
lines; but I think it is clear that a good Marshallian, such as Robert-
son, would have refused to accept it. For he would have refused to
accept that the holding of bonds as a reserve asset is the only reason
for holding bonds.
I have myself insisted (in the previous chapter) that there is
another reason for holding them; they may be held as investment
assets, just to get a regular income from them. Surely it is this which
the bond, which here we are postulating, most obviously promises.
If one is just looking at the (initial) stock equilibrium, the introduc-
tion of this 'investment* motive is a little awkward; so one sees how it
was possible for Keynesians to leave it out. But when one passes to
the question of the rate of interest which can be maintained over a
period, it must come into place. For it cannot then be denied that
extra bonds might be held, not as reserve assets, but as sources of in-
come, accumulated out of savings. If the new bonds, issued during
the period, were matched by new savings, made during the period,
there need be no additional bonds to be held by those who were hold-
ing them as a reserve. The level of the rate of interest could then
remain unchanged over the period. Equality between 'saving' and
 THEORIES OF INTEREST: KEYNES VERSOS M A R S H A L L 75

Investment' would allow "it to remain unchanged, and would allow

the supply and demand for 'goods' to remain in balance,
A model in which all these things were taken into account could
not ! think have failed of acceptance, at least in principle, both by
Robertson and by Keynes, But, having got so far in agreement, they
would still have had their differences. These would not have been
theoretical differences, in a narrow sense; they would have been dif-
ferences of judgement on which of the choice margins that had been
identified were of practical importance. This, being an empirical
question, could have been answered, quite properly, in different ways
at different times. It could be that it would have been proper to
answer it in one way in the 1880s (Marshall), in another in the
1920s and 1930s (Keynes and Robertson), to another in the 1980s.
to which we have come. And maybe to answer it in different ways in
different countries.
To deal at all adequately with the historical questions which thus
come into sight would require research, for which I do not have
afacility. There are nevertheless a few which are available to me,
from which it might begin.
Any British economist, at the time of Marshall, would have
thought that he knew what the pure long-term rate of interest was; it
was the yield on consols. Consols (market jargon for consolidated
stock) already had a history. It had been as long ago as 1752 that the
National Debt of the British Government, which in the first half-
century of its existence had been largely owed to a few big corpora-
tions (Bank of England, East India Company, and so on) had been
converted into this single transferable stock which became known as
consols. Information about the price of consols (and hence about the
long-term interest rate) from that date to his could have been avail-
able to Marshall, One may surely suppose that he would have had at
least some idea about it.
The nominal rate of interest on consols, for nearly all that time,
had been 3 per cent. The holder could not demand repayment,
though he could sell his stock on the market; but the Government
had the right to repay at par, if it was in a position to do so. That
came to mean, in practice, if it could reborrow at less than 3 per cent.
That never happened until the time of Marshall, when in 1888 there
was a conversion, the Goschen conversion, to 2| per cent.
We happen to know what Marshall thought about that
conversion—or rather about the fall in interest rates which preceded
76 M O N E Y AND F I N A N C E

it, for the conversion was announced in the budget speech in March,
and Marshall's statement was in the preceding December,3 It will be
useful to quote it, though it is more flamboyant than is usual with
Marshall, since it shows the context in which he looked at his theory:
It seems to me that the great economic feature of this age, more important
than every other fact put together, is that the amount of capital is increasing
many times as fast as that of population. It is increasing faster than ever in
England and, what is much more important, there is a very rapid increase in
America where everyone almost is saving. The 'extravagant' American is
saving more than any other person. In spite of the inventions which are con-
tinually making new uses for capital in the form of machinery and in other
ways, this vast increase forces down the interest that can be got in business.
The rate of discount, in my opinion, is merely the ripple of a wave on the sur-
face. The average level is the rate of interest which can be got from the
investment of capital, and this is being lowered by the rapid and steady
growth of capital—I do not mean the growth of credit, I mean the growth of
things, the actual excess of production over consumption. I do not see any
necessity at all why interest should be more than 2 per cent a century hence.
I should not be surprised at all if a railway company could borrow on deben-
tures at 2, or even less than 2 per cent, in the next century.
So that was where the 'euthanasia of the rentier' came from, Keynes
got it from Marshall!
Marshall, when he made that statement, knew that interest rates
had been falling, and that for some time the 'three per cents' had
been standing above par. But what could he have known about the
flow of savings? The National Income statisticians, who were to give
a later generation some information about that, had hardly yet been
born. But he was not just guessing; he was deducing his 'fact' from
his theory. Interest had been falling; so saving must have risen.
But how could he have explained, on his theory, what had
happened further back? The financial history of the first quarter of
his century would not have troubled him. Heavy borrowing during
the Napoleonic War, with high rates of interest, accompanying it; a
surviving problem of debt, after the war, so alarming that Ricardo
considered dealing with it by a capital levy; its successful absorption
in the twenty years after Waterloo; all of that he could understand,
But what of experience between the thirties and seventies?
' It was given as evidence to a public enquiry into monetary policy (Gold and Silver
Commission). It is reprinted in the collection of Marshall's Official Papers (1926), p. 46.
I owe the reference, of course, to Dennis Robertson.
 T H E O R I E S OF I N T E R E S T : K E Y N E S V E R S U S M A R S H A L L JJ

During all that time there was great stability in the rate of interest
(or yield on consols); it hardly moved outside the range of 3 to 3| per
cent. There was very little government borrowing: but there was
massive borrowing by railway companies, some of whose obliga-
tions, as Marshall recognized, were taken to be 'gilt-edged'. It would
be to these that the 'flow of savings' would be largely directed. But
why should low demands and supplies of savings, over all these
securities, have been so nearly in balance?
It is hard to find an explanation entirely on Marshall lines,
Keynes's "speculative motive* gives more help. It could be that
already at this date an appreciable part of the stock of consols would
be in the hands of professionals, who would be holding it fluid,* ready
to buy or sell in pursuit of short-term gains. Consider the position of
such a dealer, at some date in the middle of the century, consols
standing at (say) 95. The maximum capital gain which he could
expect from holding on would be relatively small; for if the price went
above 100 there would be a threat of conversion. The maximum
loss, on the other hand, might appear to be quite large. If possible
gain and possible loss were to balance, the chance of a large loss
must have been thought to be quite small. Why should that have
been so?
It is not the case that during those years the British financial
system, as a whole, as so stable. There were crises, of marked cyclical
character, and-of considerable severity (see Chapter I I ) , In those
years there would be large movements of short-term rates on the
money market (going up to 7 per cent in 1867). But their effect on
the price of consols was remarkably small. Why should that have
been so?
It would be enough if no one expected the crisis short rates to last
long, if they were just taken to be 'ripples on the surface' as Marshall
said, lithe price of consols had been standing at 95, and was confid-
ently expected to come back to that level in a few months, a fall of a
point or two in the current price would offer a capital gain from hold-

* I have a bit of evidence for that, derived from Goschen himself. In the budget
debate in the House of Commons, which followed upon the conversion, he was being
attacked by an. ignoramus on the Opposition side, who said the whole thing was what
would later have been called a "bankers' ramp'. Widows and orphans were being
exploited for the benefit of bankers, Goschen replied thut the bankers also were going
to be hit, for a lot of the debt was already in their hands (A. D. Elliot, Life of Goschen,
1911. Volume 2, p. 1 <S 3 Ut should be noted that Goschen had been a 'city man' before
he went into politics.
78 MOiNEY AND F I N A N C E

ing which would balance a quite high short rate for that time. This
would make little impression on the course, over years, of the long
rate of interest.
Nevertheless, when we come to the fall in the long rate in the
seventies and eighties, there is more to be said for Marshall. These
were years of retarded growth (in Britain); the railway boom was
ending; so a fall in the demand for savings (much more plausibly
than an increase in supply) could well have been responsible for a fall
in interest. It is true that the fall can, at least at first sight, be
explained the other way. Over the twenty years before the Goschen
conversion there had been no important commercial crises, so there
was less danger of high short rates affecting long rates, even tempor-
arily. (This also would have been a consequence of the retardation.)
In a former discussion of the matter I gave much weight to this
alternative.5 but it is my present opinion that it does not get the time-
sequence right. For surely the fall in Interest rates is recognizable
before this greater stability could have been recognized. So I now
think that Marshall, after all, does win that trick in the game,
Keynes, in the 1930s, had all Marshall's experience behind him;
and that of another forty years to add. The immediate problem with
which he was confronted was more like Ricardo's than like
Marshall's: another great war debt, but this time with international
complications which in Ricardo's case had been absent. We know of
the work he did in the twenties to find ways of escaping from those
entanglements. By the time he came to write his Genera/ Theory, he
could think of that task as being done. So his book, on the whole, is a
theory of a closed economy.
Monetary measures, to match the 'absorption' of 1815-35, had
already when he wrote been taken; the way for them had been
cleared by the floating of the pound in 1931. There had then been a
conversion, reducing the effective rate on (long) war debt from 5 per
cent to 3| or thereabouts—more or less to the rate which had so long
been held steady in the reign of Victoria. The practical problem, as
Keynes saw it at the time he was writing, was to keep the (long) rate
at the level it had so recently reached—to stop it going back to where
it had been not so long ago. That he knew would be easier if the
market could become convinced that 3| (or whatever it was) was the
proper or normal rate—a rate which had some prestige. History
5
Critical Essays, p. 93.
 T H E O R I E S OF INTEREST: K E Y N E S VEKSUS M A H S H A L i 79

could give it some prestige, but that was a long time ago, So to
restore its authority it would have to be maintained, for a consider-
able time; the market would have to get used to it. Then it might
(perhaps) be put back, on to its throne,
Not enough time was in fact allowed for this experiment. The
British economy, in, 1936, was not starting on a long period of peace,
with an old-fashioned Conservative government; quite the reverse.
So it was not until the Second World War was over, when Keynes
was no longer available as an adviser, that there seemed, to some
people, to be another chance. I much doubt if Keynes would have
thought or ought, to have thought it to be another chance. Surely, on
his principles, the 'Daltons' of 1947 came much too soon.
Looking back on all this from our own standpoint, with another
forty years of experience at our disposal, there is surely one thing
about it which stands out. In the days of Marshall, widely considered,
and in those of Keynes, also widely considered, it was rates of interest
between 2 and 5 per cent, at the extremes, which were being con-
sidered. Since 1950 we have become adjusted to rates which are at
the least double. We are getting to think of 8 per cent as a very
moderate rate; not, as it would have been in the old days, cata-
strophic. Why has this happened? This is the very topical question to
which my historical enquiry has been leading up.
It is commonly thought that these high rates of interest are a con-
sequence of inflation; that if prices are rising, at 4 per cent per
annum, a nominal rate of interest of 8 per cent per annum is equiva-
lent to a real rate of 4. It is true that inflation makes these high rates
of interest bearable, so that their consequences are not so desperate
as they would have been in the past. But to make those consequences
into causes surely takes things the wrong way round. Does not
Marshall come to the rescue? He would have attributed the high
rates of interest to the great expansion of government expenditure, in
many countries, for peace-time indeed rather than war-time purposes;
and to the unwillingness of governments to impose the taxes to pay
for them. That, he would have said, was how the phenomenon
started; the high rates and the inlation were consequences of it;
admittedly they facilitated each other. Would he have been wrong?
 10 Markets in Equities:
Ownership and Control

There is a curious correspondence between the matters to be con-

sidered in this chapter and those which were considered in Chapter 4
at the end of the irst part of this book. We there saw that in the
labour market there would be two sorts of employment to be dis-
tinguished, one of what I called established labour, where employer
and employed expect their relation to continue for some time to
come, and one in which there is no such expectation of continuance.
There are advantages, from the point of view of economic efficiency,
which either may have over the other. There are things which can
be done by a team of people who are accustomed to working to-
gether, quickly understanding one another, which cannot be done
by a collection of people more loosely attached. But the other has the
advantage of being more easily adaptable. We should expect the typi-
cal firm to pay some attention to each of these advantages. It would
then tend to employ a nucleus of established labour, to which a fringe
of less established labour would be attached. Something like that
would serve as a model of a typical firm.
Now surely in such a model the management itself would be in-
cluded in the nucleus. So contracts between the managers, like those
between themselves and their established employees, would be links
which tie the firm together. If the managers are also proprietors, con-
tracts between them in that capacity could also be regarded as link-
ages, of course important linkages, They would be expressed in terms
of shares which particular proprietors hold.
A firm in which proprietors are also managers, and there are no
other proprietors, can hardly be other than quite a small firm; what-
ever be the legal form that is given to it, it is economically a partner-
ship. Shares in a partnership can, on occasion, be bought and sold,
For one of the partners may retire from the business; he is 'bought
out'; he sells his share to the others. But this will surely be a complic-
ated transaction, requiring much negotiation; there can be no
market, certainly no competitive market, for shares of that type. In
order that there be such a market, the link between proprietorship
 M A R K E T S I N E Q U I T I E S : O W N E E S H I P A N D CONTROL 8l

and management must be broken, or at least diluted. It was the insti-

tution of Limited Liability which brought this about.
In the absence of this facility, each partner would be personally
liable for the debts of the partnership; thus if a creditor could not
recover from it as a body, he could endeavour to recover from each of
them individually, against the whole of his assets, whether those
assets were engaged in the business or not. For the law would treat
his dealings through the firm as being done in his private capacity. It
was this which fortified the connection between ownership and
management. For it would be most unwise to come in as a 'sleeping
partner1, exposing oneself to responsibility for the debts of the firm
without taking part in its management. For one could then be
ruined, as sleeping partners have indeed been ruined, by the mis-
takes the management had made.
If the firm can be made into a company (or corporation) with
limited liability, the worst that can happen to a shareholder, as a
result of his participation, is that his shares become valueless. A cred-
itor can proceed against the company for payment, but he cannot
proceed against shareholders individually. It must of course be
announced to those who trade with the company that it has this
privilege; but this has proved in practice to be no serious obstacle to
trading. It does however remove an obstacle to the admission of out-
side shareholders, shareholders who do not themselves take part in
the running of the firm. It therefore removes an obstacle to the
growth of the capital invested in the firm.
It was clearly for the purpose of removing this obstacle that limited
liability was invented. In its absence the capital that was invested In
a firm would be restricted by what the partners (both owners and
managers) could themselves put up, and by what they could borrow.
But what they could borrow would itself be limited by the security
they could offer, and that again would depend upon their personal
wealth. A small group of partners might not be able in this way to
offer much security; but if outsiders will take up shares, carrying
limited liability, the total capital invested in the firm can be much in-
creased. A creditor therefore has better security, since he is more
likely to be able to recover from the. firm even though he loses the
right to recover from shareholders individually.
It is true that even in the absence of limited liability a firm could
grow, by reinvestment of profits; so it could become large, but it
could not start large. The development of such industries as need to
82 MONEY AND F I N A N C E

make large Investments of capital in order to produce at all would be

severely hampered. Thus it is not surprising that the building of rail-
ways is historically connected with the corning of limited liability.
At this stage, when the purpose is just to enable entrepreneurs to
get control of more capital than they could put up themselves, or
borrow on bonds or debentures, it would seem that the 'ownership'
of the company, by the shareholders who are thus introduced, can be
no more than nominal. The shareholders, by law, have the right to
elect the directors; but the directors are in place before a first issue of
shares is made, and the shareholders show, by subscribing, that they
have no desire to turn those directors out. Thus, whatever be the
legal position, the fact is that capital is being provided by share-
holders, just as it would be by bondholders, only on different terms.
While the bondholder is promised a fixed rate of interest, the share-
holder is promised a share in profits. The shareholder is much more
like a bondholder than he appears to be in legal theory. That this is
the right way of looking at the matter is shown by the fact that the
terms of an equity issue can be varied, just as security offered to
bondholders can be varied, by adjusting the capital that the propri-
etors, becoming directors, are supposed to be putting up and indeed
in other ways. That is how the limited liability company begins.
Now how has it been possible to raise capital on these peculiar
terms—capital which, it must be supposed, could not be raised, or
not so advantageously raised, by borrowing at fixed interest? This is
a key question; it will be convenient, when attempting to answer it,
to take it in two stages, beginning with the case in which the shares
are not freely transferable, and going on to the more important case
when they are. Only in the latter, of course, can there be a market in
the shares. But the former is quite a real case. It extends beyond the
case of the private company, or non-quoted company, where there
are legal constraints on transferability; for there may be de facto re-
strictions, as when the new shareholders are no more than a moder-
ate extension of the small group who have direct knowledge of the
business, to people who have personal knowledge of the directors, or
some personal relation with them. It will be useful to begin with this
private company case, since it brings out by contrast how important
is transferability,
Already in the case of non-transferable shares, we have to consider
the matter from two points of view: that of the investor who is to buy
the shares, and that of the company, as represented by the director.
 M A R K E T S I N E Q U I T I E S : O W N E R S H I P A N D CONTROL 83

which is to issue them. Take first the investor's position. Why should
he buy shares, when he could have invested to bonds? That the
equity investment may be regarded as a hedge against inflation is an
argument in its favour which in some recent years may have been of
importance; but it may here be disregarded, for at the time when
limited liability first prospered, it cannot have been significant. There
must have been another advantage which was the attraction,
One may look at the matter in terms of the probabilities of return1
which are offered on the two alternatives. All that has to be con-
sidered, when the share is to be non-transferable, Is the annual return,
which is to be expected in some representative future year. If the
capital had been raised on the bond, it could only have been so raised
if there were a strong probability that the interest agreed would be
paid; there would nevertheless be a (usually small) probability that
the borrower would be unable to pay, so that a part, or the whole, of
what had been invested would be lost. In the case of the share, St is
useful to suppose that there is some most probable return, which plays
much the same part as the agreed return on the bond. The maximum
that can be lost, when there is limited liability, is the same as with the
bond—total loss of all that has been invested. The probability of total
loss is however greater, since the claims of bondholders must be met
before there is anything to distribute on the shares. The probability of
partial loss—a return which is significantly less than that which
appeared to be the most probable—is clearly much greater in the
case of the share. Both of these are disincentives to investing in
shares. What is there to be set against them? It might seem at first
sight that all that there is to set against them is the possibility of a
return which is greater than the most probable return; this is zero in
the case of the bond, but in the case of the share can be substantial.
One of the attractions of shareholding, as against bondholding, may
be the possibility of what one might call extra gains,
But it is unlikely that it is the only attraction. If it were to be the
main attraction, the investor would have to be a bit of a gambler—
who may be defined as a person who is willing to swap a high prob-
ability of a moderate loss against a low probability of a gain that is
large in comparison. It is not to be denied that a propensity to gamble
is one of the things which explains the growth of equity investment.
* Probability in the non-technical sense as might be used in business transactions.
There is no scope in this discussion for the concepts of mathematical probability theory
(see Appendix).
84 MONEY AND F I N A N C E
But the extra gains that are possible from investing in a share with
restricted transferability can rarely be large enough to arouse the
propensity to gamble. There must be something else.
What that is emerges as soon as we consider the alternatives from
the company's point of view. When saying that the agreed return on
the bond and the most probable return on the share were compar-
able, it was not implied that they would be equal. There is a good
reason why the return on the share should be higher. It is in
the interest of the company—as represented here by the former
proprietors—to offer a prospect of a higher return on the share,
because (so long as the question of control does not arise) finance by
share issue diminishes risk.
Interest would have to be paid in good and bad times alike; but in
bad times the dividends could be reduced, so the burden of finance by
shares would be less. It is true that it would be expected that in good
times the dividend would be increased; but it is just in such times that
an increased out-payment can most easily be borne. The firm would
be insuring itself, against a strain which in difficult times could be
serious, at the cost of an increased payment in conditions when it
would be easy to meet it. It is in this sense that the riskiness of its
position would be diminished.
Thus it would be worth its while to offer a premium, in terms of
most probable dividend, to offset the other disadvantages, to the in-
vestor, of investing in shares, which of course would remain.2
All this applies to the private company, where the circle of share-
holders is restricted, and (usually in consequence) the funds that can
be raised from them are restricted. Much more can be done if trans-
ferability is permitted. The positions of all parties are then trans-
formed.
There can in the first place be a much stronger appeal to the pro-
pensity to gamble. When transferability is restricted, the investor
must expect to wait quite a time for the possibility of extra gains; and

* t think this analysis does get to the root of the matter; but we may also have to
distinguish between the probabilities, as they appear to the 'insiders' who are-raising
the capital and as they appear to 'outside' investors. The 'insiders' may be expected to
have better information on which to base their estimate, A major objective of
company law has been to diminish this inequality of information. If that endeavour
has some success—It can never be wholly successful—it diminishes the risks to the in-
vestor of investing in shares. And that should be favourable to growth of equity invest-
ment, for if the risk to the investor is diminished, it should be easier to persuade him to
'bite'.
 M A R K E T S IN EQUITIES: O W N E R S H I P AND CONTROL 85

when they appear (if they do appear) they can hardly be 'glittering
prizes'. But when the share is transferable, there is a chance that it
will be possible to sell it at a profit, even in the quite short run. I put
this first because in the case of a new issue, it is the motive which
comes in first Big issues of shares are most easily placed when there
is a bull market, when the market most closely resembles what
Keynes called it, a 'casino'.5 And the association of share issue with a
bull market is one that goes both ways.
This is because the transferable share, carrying limited liability, is
a liquid asset. It is not as liquid as cash, or as the closer substitutes for
cash; but it is liquid in the looser sense that it can be turned into
cash, at short notice, whenever cash is required. So the successful
speculator can take his profit, whenever it seems good to him to take
it. Even if he does not sell, he has a security against which he can
borrow, then perhaps using the proceeds of the loan to buy more
shares. Thus a speculative boom is built up. Where does the money
come from to support it? The answer, as economists now are well
aware {though it took them much trouble to find it), is that the
market as a whole, meaning by that the total of all those who are
dealing on the market, because it is regarding the share as a liquid
asset, is willing to hold less of its in monetary form than it was
before the boom started. New shares, the dasling prospects of which
have been well advertised, are ideal as a speculative counter.
But though it is in the bull market that shares are most easily
placed (and therefore from the point of view of the issuing firm most
cheaply placed) they will have in the end to be held; and at that stage
those who hold them must surely have a normal risk-aversion. Even.
so there are advantages, to the investor, of transfersbility. One of
them, still is the liquidity advantage. The share can still be disposed
of, at a price, if the shareholder changes his mind about the prospects
of the company or if he wants cash for any other reason. Another,
which comes from the combination of limited liability with transfer-

* I myself rather think that the resemblance to a casino can be overdene. It is well
known that the player at Monte Carlo is bound 'in the end' to be a loser, but hopes,
irrationally, that this time he will win. (Or perhaps, a little more rationally, that he
himself will have the strength of mind to retire from the game at a good moment.) An
amateur stock exchange speculator may indeed behave like that but a boom could get
built up entirely by professionals. Their conduct would not be irrational; it would be
based on information, but this could be very restricted information, information that
was available quickly. Perhaps no more than what the person at the next desk was
doing.
86 M O N E Y AND F I N A N C E

ability, but which is not fully effective until the market in shares is
well developed, is that the investor can reduce his risks by diversify-
ing his portfolio. Under limited liability, as was shown, the share-
holder is made more like a bondholder, in that he cannot lose more
than he has put in; so he can imitate the age-old practice of lenders at
fixed interest, diminishing his risks by spreading them. It is only the
risk of loss on his portfolio as a whole which matters to him. If the
securities in which he invests are different, being subject, at least to
some extent, to different risks, the risk of loss over the portfolio as a
whole is likely to be proportionally less than on any single invest-
ment. Without limited liability this could not be done, since invest-
ment even in a second enterprise would increase the investor's risks,
but with limited liability it can be. It is this, combined with the liquid-
ity advantage, which explains why it is possible to issue shares at no
more than a modest 'premium', even though investors are risk-
averters.
That is how it happens; but consider the consequences. It is inher-
ent in these advantages, both the liquidity advantage and the prac-
tice of risk-spreading, that the individual shareholder, in a fully
developed limited liability system, must tend to lose contact, other
than a purely financial contact, with the companies of which he is
legally part-owner. His right to elect directors must then tend to
become a mere formality. There is a way in which this may not
happen; but I begin by considering the common case when it does.
The firm is now (we will take it) well established; it has been, well
established for some considerable time. Who controls it? And on
what principles should we expect them to control it?
The first of these questions takes time to come up, for (as we have
seen) the original directors are in place at the beginning. But the time
will come when they have to be replaced; successors, on occasion,
will have to be found. When provision for the election of directors by
shareholders has become ineffective, the directors themselves must
make provision for succession. It is however by no means evident
that the method of providing successors by co-option is any less effi-
cient than that which has to be used in the private business, or
partnership. It should indeed be more effective, since in the latter
case the choice is likely to be confined to members of a few families,
while the directors of a company are not so confined. The company is
ruled by what in political terms would be called an oligarchy; but ex-
perience seems to show that 'open' oligarchies, which recruit them-
 MARKETS IN EQUITIES: O W N E R S H I P A N D CONTROL 87

selves from wider circles, are more efficient and more durable than
those where the field for recruitment is narrower.4
One may approach the other question by thinking of that which is
familiar in examination papers in economics; what does the entre-
preneur maximize? So long as one can think of a stogie entrepreneur
in command of a business that belongs to him, there is no doubt
what he should be expected to maximize. It is the discounted value of
expected net returns—discounted at his individual rate of time-
preference. Allowance will have to be made for uncertainty, and per-
haps for the toil and trouble involved; but all that is well understood.
In the original formation of the limited liability company the same
principle, as we have been seeing, will apply. The entrepreneur, at
this stage, is the former proprietor, or proprietors, becoming direc-
tors; it is the prospective net return, accruing to them, which they
will naturally seek to maximize. But after formation it is surely the
prospective net return to the shareholders as a whole which they
should seek to maximize. That is the legal theory; but when one has
gone to our late stage, with a new generation of directors, and share-
holders who have lost touch with the company, can that legal theory
be realistic? One can more easily see the directors thinking it to be
their duty to do their best, not for the shareholders, but for the com-
pany. But what, in this sense, is the company?
Before the appearance of transferable shares there would have
been no doubt about this; it would have been the proprietors. After-
wards we can get a hint from what was said at the beginning of this
chapter about the labour market, about the structure of the firm in
terms of the labour which it employs. Here also, on the side of capital
structure, a distinction between nucleus and fringe appears to be
required. In the case of the private company, where shareholders are
closely attached, there will be no question that on the side of capital
they constitute a nucleus. Capital that is borrowed from outside
lenders (most definitely when it is expected that the loan will be
repaid) will be fringe. In the case of the public company, with shares
that are easily transferable, there is no distinction that can be put
* Corresponding experience in the political field is instructive. Was there any Euro-
pean monarchy, in the days when kings really ruled, which did not suffer disasters
from adhering to the rule of hereditary succession? The papacy, where in effect succes-
sors are chosen from a wider circle (for though the college of cardinals is restricted in
number, they themselves can be chosen much more widely >, has over many centuries
done much better. It is interesting that, the Russian communists appear to be following
the papal example.
88 MONEY AND FINANCE

into so objective a form. There must however usually be a practical

distinction between Inside' shareholders, who feel themselves to be
closely associated with the company, so that (like established labour)
they expect to go on holding for considerable periods, and the fleeting
population of shareholders who are more loosely attached. All share-
holders alike will have to be paid the dividend, but while the out-
siders are concerned with no more than the current dividend and
with the market value of the shares, the insiders are concerned with
the future of the company, and so with the dividends they expect, on
their own information, to receive at future dates.
This is enough to give rise to some conflict of interest. The outsider
may be expected to desire a dividend as large as possible, but it is, at
least to some extent, in the interest of the insider that profits should
be retained. Why then should it be the practice of companies, which
the system leads to be controlled by insiders, to pay anything more
than a nominal dividend? The answer seems to be that it is in the
interest of the company, even when that is interpreted as we have
been coming to interpret it, to keep up the market value of its shares.
And this, being formed by outsiders, is likely to be strongly affected
by the current dividend.
The traditional reason for desiring a strong value of shares is that
it makes it easier for the company to borrow; but this, since large
retentions make borrowings less necessary, cuts both ways. A more
important reason, in recent times, but not only in recent times, con-
cerns take-overs and mergers.
A take-over bid, in the light of what has been said, may be
regarded as a counter-attack by shareholders who, in the course of
the process described, have been so nearly pushed out. Although a
'change of government' as a result of a discussion at an annual meet-
ing is not practicable, the market has provided an alternative. The
prices at which a company's shares are normally valued are based on
the expectation that the company will continue under the same
management; for the sale of a few shares by one individual and their
purchase by another is unlikely to affect any voting. If however a
company's policy is such that an outsider can feel sure that it is not
doing its best for its shareholders, it will be worth his while to offer
higher prices for large blocks of shares, in order to acquire enough
voting power to change the management, either in person or in
policy. Having brought about this revolution, the shares can again
be sold; and if the prospects of the company, as assessed by the
 MARKETS IN EQUITIES: O W N E R S H I P AND CONTROL 89

market, then seem to be sufficiently improved, they can be sold at a

profit. When directors have got themselves into the habit of mind
(the origins of which have been explained) of thinking that they are
working for the interest of the 'company', or even of the 'public', or
for anything other than the interests of their shareholders, they may,
in accordance with the legal position, lay themselves open to a flank
attack on these lines.
The outsider who makes the bid will itself In most cases be a com-
pany; and it mall be easier for it to make the bid if its own shares are
strong. For on the one hand it will then be possible to offer share-
holders in the weak company shares in the strong company in
exchange; and on the other, if cash has to be offered, the cash can
more easily be borrowed. Thus it is not only weak companies which
have an interest to see that the value of their shares is maintained,
for defensive reasons; it is also strong companies who have a corres-
ponding incentive, so as to make it easier for them to expand by ac-
quisitions. When the two incentives are taken together, they cover a
good part of the ground; so the position of the shareholder is stronger
than might appear at first sight,
The take-over sanction has on the whole a good reputation among
economists; but we should distinguish. There is no doubt that in the
case of a management that has gone sleepy, or has forgotten to keep
its eye on profitability, it has an economic function; the quality of
management should be improved. But this is by no means the only
thing that can happen. Take the case, at the other extreme, of two
firms which are competing in the sale of their products. The stronger
may buy up the weaker just to get rid of a competitor. It can be a
cheaper way of doing this than by 'cut-throat competition', forcing it
into bankruptcy. So the availability of the take-over may in the end
facilitate monopolistic practices, thus protecting inefficiencies which
it might have been expected to obviate,
Yet there is more to be said for an alternative which may be hard
to distinguish from this last. If the firm taken over is absorbed into a
'group', that also may be conducive to monopolistic behaviour but
need not be. This will happen if the members are left with powers of
initiative, the 'centre' retaining little more than some functions
which are clearly appropriate to it. Two of these have emerged from
what has here been said. One is providing help over succession of
managers; the other is financial, The 'centre' naturally acts as a bank
for the group—a bank which, because of its closer and probably
9O MONEY AND F I N A N C E

more permanent association, can keep in closer touch with the needs
of members than an outside bank could do.
We saw in Chapter 7 that it is to an arrangement of this sort that a
banking system itself is likely to tend. There is no reason why it
should not show its strength also outside the banking field. The ulti-
mate reason for it is that the conventional unit of production is sub-
ject in different ways to scale economies; the 'optimum' in one
direction may be different from that in another. It makes for effi-
ciency to separate them out; when that can be done consistently
with co-operation.
 PART III

PROBLEMS AND POLICIES

This page intentionally left blank
 11 The Old Trade Cycle

That the work of Keynes was a turning-point, between one epoch of

economic thinking and another, has been shown in several ways in
previous chapters of this book. That it was associated with another
turning-point, in what economists call the 'real world', has not been
given the same attention. He was writing during and just after the
world economic crisis of 1929-34; this is background to what he is
saying. At that time It was obvious to his readers that there was this
association: not only his occasional pieces, to be collected as Essays in
Persuasion, but also his major works, had a topical side to them. They
reflected his perception of what was going on around him—the
passage from one way of organizing economic affairs to what would
have to be another.
Other economists—1 think one may safely say, most other
economists then engaged in writing and teaching—did not see it like
that. They had their own idea of what was happening; they looked
upon, it as an example, admittedly a major example, of a familiar type
of disturbance which was called the trade cycle or business cycle. It
was already many years since that had first been recognized. We
may perhaps date the beginning of its recognition (in Britain at least)
to the work of Jevons in the 18 70s. This is important, not in itself (for
his sunspot theory of cycles did not stand up to investigation) but
because of the belief, on which it was founded, that something was
happening which was sufficiently repetitive to warrant analysis by
some sort of scientific method. There could be no doubt after Jevons
that a question at least had been asked.
We need however to understand that the cycles, which Jevons and
his immediate successors had in mind, were not the statistical cycles
so well known to the modern economist. These can be found in any
time-series, of any variable, not at all necessarily economic. The
current observation is greater or less than its predecessor: the series
is going up or going down. A single rise or fall, we admit, is not signi-
ficant; but the statisticians have been busy in giving us tests of signi-
ficance, which largely consist in the fitting of the observations into a
cyclical pattern. Mere continuance of the motion in the same direc-
tion gives a crude test of significance; as when the market analysts
94 PROBLEMS AND POLICIES

say that a fall in an index, continuing for three quarters, may be

called a recession. But we do not want a rise that starts from a low
point, and never gets far from that low point, to have the same signi-
ficance as one that starts high. What however is 'high' or 'low'? It
must be with reference to some norm, or standard. One can con-
struct a norm by purely statistical methods, fitting a trend, and
judging the actuals by reference to the trend. Most economic series
do not jump about very much from one observation to the next (for
the observations are snapshots of what in fact is a continuing pro-
cess); so when there is a deviation from trend which is at all consider-
able it will usually last for a number of observations. There will be
sequences of pluses and sequences of minuses; and it is easy for these
to begin to look like a cycle.
I want to insist that this is not what Jevons and his contemporaries
can have had in mind; for the statistical apparatus of trend-fitting
had not in their time come into use. They were thinking of the se-
quence of trade crises which had marked the preceding half-century,
occurring in 1825, 1837 (possibly1), 1847, 1857, and 1867; that
looked very much like a ten-year cycle (which would match the
sunspots!). These crises were striking events; even the earlier ones
had received a good deal of attention from economists,* though it
1
For it was more important in the US than in Europe.
1
And others. Thomas Love Peacock, whose satirical novels are still worth reading,
wrote a piece entitled 'Pan in town" on the 1825 crisis. (Pan Is the god of pan-ic.) I
cannot resist quoting a few lines of it,
Pan: The country banks are breaking
The London banks are shaking
Suspicion is awaking
E'en quakers now are quaking:
Experience seems to settle
That paper is not metal.
And promises of payment
Are neither food nor raiment...
Chorus: Our balances, our balances
Our balances, our balances
Pay—pay—-pay—pay
Without delay
Our balances, our balances.
(Enter Scottish economist)
Economist; A weel sirs, what's the matter?
An' hegh sirs, what's the clatter?
Ye dinna ken
Ye seely men,
Your fortunes ne'er were better,
 THE OLD T R A D E C Y C L E 95
was not until the time of Jevoos that the series as listed was complete.
So it was not until then that the periodicity could be recognized (of
course it is clear from later experiences that too much attention was
given to it, not only by Jevons). Nevertheless, though the periodicity
may have been dubious, there was not much doubt about the repeti-
tion.
Not enough attention has later been paid to the useful discussion
of the earlier crises which is to be found in the Principles of John
Stuart Mill (1848), surely the standard work on principles of eco-
nomics belonging to that central Victorian period. Mill did not have
the whole story before him; but what he says about the part on
which he did have information is good. The cycle was a financial
cycle. There was a boom, with rising prices and then rising interest
rates; it led to a crisis, with a wave of bankruptcies. The unemploy-
ment which followed was a consequence of the bankruptcies, or of
attempts to avoid them. After the crisis prices fell: rates of interest
then came down. The latter was a first step on the road to recovery.
Mill was already able to show how this could happen.1 A general
rise in prices—many prices rising, few falling—would require to be
financed. It could at first be financed by trade credit (as described in
Chapters 5 and 6 in this book). The buyer got the goods at the
expense of an increase in his debts. But trade credit though expans-
ible would not be indefinitely expansible; the time would come when
some of the extended trade had to be financed by something com-
manding wider confidence. Recourse would then be had to the bank-
ing system, and there would be an expansion of bank money. But
even bank money (then) might not be thought of as being reliable
enough; so there would come to be an increased demand for the
really solid money, the gold coins which were then in common circu-
lation. The custodian of the central gold reserve was the Bank of
England; it had a reserve but it was very limited. So when the Bank
was in sight of the point at which it would be unable to exchange its
notes for gold, it had to take action to enforce a restriction on

There's too much population

And too much cultivation
That's a' that alls the nation.
So he sends them off to a lecture on the principles of banking. But the bank has
nothing but Chilean bonds to pay them with.
1
Principles. Book HI. Chapter 12. His essay 'On the influence of consumption on
production' in Essays on Unsettled Questions (1844) needs to be read with it.
96 PROBLEMS AND POLICIES

borrowing. That was the crisis. The rising interest rates which pre-
ceded the crisis showed that some people saw it approaching.
In the crisis, weak positions were uncovered, and there were
failures. But the Baak itself survived, and most of the banks survived,
The pressure then relaxed; interest rates, being symptoms of the pres-
sure, came down. When the debris had been cleared up, so that
nearly all firms which survived were of unquestioned solvency, an
'equilibrium', as it might be called, would be restored. But it would be
an unstable equilibrium, since it was just from such that the former
boom had started. Sooner or later the cycle would be re-enacted.
There are two particular characteristics of this classical (or
jevonian) cycle which need to be emphasized. One is that at the time
in question Britain, beyond doubt, was the economic centre of the
world. So it is perfectly proper, for that time, to treat the British ex-
perience, as the experience; much of what happened in other coun-
tries just followed from it. The other is that in this period the Gold
Standard was sacred; in Britain it was sacred. There was no question
(it was known that there was no question) that the convertibility of
Bank of England notes, and deposits, into gold at a fixed parity must
be maintained. It was this which in the end provided a firm ceiling on
expansion, a monetary ceiling. It was known that the expansion
could not go on indefinitely; so, as the boom developed, people began
to take precautions. Wise men had battened down the hatches before
the storm broke.
There was also the question, known to be the more difficult ques-
tion, of providing a floor. Prices, particularly of primary commodities,
would fall at the crisis; but if the cycle was to continue (or if there
was to be a return to equilibrium) they must be stopped from going
on falling. How should that be? They had fallen in the crisis, not
because they had been thought to be 'too high', but because the
money that was needed to support them had been lacking; bear
speculation, selling to buy back later, would nevertheless have been
a feature of the fall. One can see that a point would be reached (there
is plenty of experience of its being reached) when the balance of
opinion among speculators would turn in favour of the fall having
gone too far, so they would begin to speculate for a rise, at first very
tentatively. Such speculation is stabilizing; it needs to be encouraged.
It does not appear that Mill had got so far as to see this, though it is
stated in a much earlier work of which he made use—the Paper
Credit of Henry Thornton (1802). There it appears in a context
 THE OLD T R A D E C Y C L E 97

which by the middle of the century had become very historical; so

the whole of its relevance to current concerns may not have been
obvious, even to Mill. The Thornton precept (as I shall call it, for be de-
serves to have his name attached to it) was in two parts. The first
necessity, when the crisis has arisen, is for the centre of the system
(in his case, or Mill's, the Bank of England) to ensure its own security;
for that purpose it must maintain high rates of interest, so as to draw
funds to itself, to replenish its reserves. However, when that has been
done, it should turn over decisively to the other tack, with the aim of
spreading security from itself to the rest of the banking system, and
then outside. The two belong together.4
The Thornton precept does not only mean, and perhaps does not
even principally mean, that rates of interest which have risen in the
crisis, and indeed on the way to the crisis, should come down as soon
as it is safe for the Bank to reduce them. For it should not be taken for
granted that offers to lend, at low rates of interest, to suitable
borrowers, will in such conditions be easily taken up. Active lending
presupposes confidence, on the part both of the lender and of the
borrower; arid of the borrower, because of the lender in the borrower,
for how should the (now) selective lender have confidence in the
borrower if the borrower cannot give reason for feeling that con-
fidence himself? That loans are available at low rates, from one part of
the central nucleus of the banking system to another—between well-
established banks, or from them to other well-established financiers—
does not necessarily imply that they are so easily available outside.
The essential point of the Thornton precept is that after a crisis, when
that has been a considerable crisis, the inancial system, and the rest
4
The name which later writers have commonly associated with what 1 call the
Thornton precept is that of Walter Bagehot {Lombard Street, 1,875). Bagehot does say
explicitly, on his page 209: The best palliative to a panic is a confidence in the
adequate amount of the bank reserve, and in the efficient use of that reserve.' This no
doubt was meant to mean much the same thing as was meant by Thornton. It may
fust be that because the banking system with which Thornton is concerned is simpler,
the principles on which he is basing the precept seem to me to come out more clearly
than they do in Bagehot. It was not until 19 39, on the publication of a reprint of Paper
Credit, that Dennis Robertson, who needed it, read the Thornton book. But I remember
what he said to me years later, when I tola him that I was giving a tecture on Thorn-
ton, to be reprinted in my Critical Essays: 'Oh, Thornton,, he knew everything.' He did
indeed have all the essentials of the theory of monetary policy, of which Robertson had
made himself the champion. He was nearer to Robertson than he was to Hayek, who
wrote the introduction to the reprint. I doubt if Keynes ever had an opportunity to
study Thornton, or indeed on these matters, even Mill. The 'classics' he did study were
less help to him than they would have been.
98 PROBLEMS AND POLICIES

of the economy, so far as that Is dependent on the financial system,

needs to be nursed back to health.
If this had been accepted—and it may be that by the 1870s it was
by the more far-seeing bankers beginning to be accepted—it could be
maintained that though the financial system had no sure means of
maintaining an 'equilibrium', it did have means of correcting excess-
ive departure from it. There was no doubt that it was possible for
leading banks by acting together to prevent 'overheating'; and if the
precept were followed they would have at least some means of com-
bating a divergence in the other direction. Violent swings, as experi-
ence showed, were damaging; but if the remedial measures were
promptly and adequately taken, the swings would be damped and so
made less harmful. That however implied that there was someone, or
some body, that was in a position to take the action required. In that
crucial sense the system needed to have a centre.
It would not necessarily be implied that this should be what we
now think of as a Central Bank. A Central Bank, we have in later
days come to take for granted, is a bank which-—whether or not it is
actually set up by government—has a close relation to government;
it is a part of what in a wide sense may be called the political struc-
ture of the country in which it is operating. But a centre need not
begin in that way. As we saw in the previous chapter on banking
(Chapter 7), because of the risks to which any bank is in principle
subject, a banking system as such has a tendency to develop a centre,
being a bank, or group of banks, on which other banks come to rely.
This can come about without any action by government. The rela-
tion so established can indeed transcend the boundaries of nations. It
is not inevitable that the centre and the circle of banks dependent
upon it should be situated in the same country. This was already
apparent in the British case in what we may call the post-Jevonian
period, London was acting as a centre, not only for the British bank-
ing system in a narrow sense, but also for banks in the then far-flung
British Empire, and even for banks in countries which were in no
political sense British, some South American countries being at that
time leading examples. The circle for which the Bank of England was
centre was a very extensive circle. That may well have made it easier
for the Bank to pursue a stabilising policy. We do in fact find that in
the days of Marshall and Edgeworth financial crises were mild; the
financial cycle was almost disappearing.
But for this it is likely that there will have been another reason.
 THE OLD T R A D E CYCLE 99

Though Britain was retaining much of her financial leadership, the

economic leadership which she had possessed in the middle of the
century was going. Most of the 1880s were years of depression in
England: but they were not years of depression in Germany or in the
United States. It is indeed inevitable, when countries are on a
common standard (in this case a gold standard) and are committed
to the maintenance of that standard, that those which are lagging
(as we should now say, in terms of productivity) will not be able to
let themselves go' in the way the stronger countries can do. So the
booms which they can allow themselves will have to be weak.
Though their 'growth' over the cycle as a whole will be less than that
of the others, nevertheless within, the cycle they are likely to be more
stable,
So I do not need to say more about the British story, in the sequel
to the Jevons period. The British had more or less solved their prob-
lem: but by the end of the century the scene was shifting to the
United States. The US, at the time in question, had no Central Bank;
to have one was thought to be inconsistent with their political prin-
ciples; it was too central an institution to fit a federation. Neverthe-
less when the need arose a substitute was found, one which at first
did not do so badly. The American banking system, as we should
expect (see again Chapter 7), was finding a centre for itself; it con-
sisted of the big New York banks, on which the rest had come to
depend. In the (American) crisis of 1893 that worked. The New York
banks acting together, however informally, were strong enough to
carry through a Thornton-type policy.
Nevertheless on the next occasion (1907) when there was a
strain, it was a greater strain, and it became doubtful if what had
been done on the former occasion was sufficient. Some (at least tem-
porary) more formal arrangement appeared to be required. But these
were the days of President 'Teddy' Roosevelt and his war on 'Trusts',
industrial combinations, centres of economic power that were
thought to be dangerous. Presidential approval of the bankers'
arrangement had thus to be sought; and though in the end it was
got, it was hard to get it.' So, after 1907, it became accepted, (I) that
some sort of a formal centre was needed and (2) that it must be
federal, explicitly federal. This was the origin of the Federal Eeserve
System (the 'Fed'), legislation for which was passed in 1912, though
' I have been helped by the lively account of this In the essay on Pierpont Morgan
in J. R. T. Hughes, The Vital Few.
IOO PROBLEMS AND POLICIES

it did not begin to operate until November 1914, after the beginning
of war in Europe. Thus for no fault of its own, it had a difficult start.
Its behaviour between 1914 and 1921 has been much criticized, but
it is hard to see that the criticisms go very deep. War-time inflation
could hardly have been avoided by any policy; and, what may be the
most important test, recovery from the slump of 1921 was remark-
ably quick.
The events of 1929-3 3 were very different. We now know that the
policy of the 'Fed' at that time had much to do with the disaster
which occurred. It has been shown* that in its early years it had been
much under the influence of experienced New York bankers, per-
sonified by the New York governor, Benjamin Strong; but he died in
1928, Thereafter policy was determined by a majority of governors
of 'member banks', many of them coming from the fringes of the
banking system. They had not learned how to act as central bankers.
They did not realize that the reserves, that could be available to the
'Fed', were immensely strong; they did not need a great deal of fortifi-
cation. They were thinking of their own regional banks, which did
need fortifying, but could have been fortified from the centre. So as
this was neglected, they went on clinging to the first part of the
Thornton precept, and forgot the second.
The result was just what a Thorntonian could have predicted.
About a year after the Wall Street crash, when (if his policy had been
followed) there should have been signs of recovery, there was a
secondary crisis, still largely confined to the US, when many banks,
and not only banks, closed their doors. But the pressure was kept up,
and in 1931 it happened again. This time Europe also was involved.
Leading European countries at that date had returned (after the
War) to a gold standard, but a very fragile gold standard, largely sup-
ported by American credit. So we may think of the 'Fed' as having a
circle of banks dependent upon it, not only within but also outside
America. Foreigners could however protect themselves, in the emer-
gency, by abandoning their fixed link with the dollar, 'going off
gold'.7 This is what they did, a good many of them did, beginning
with the floating of the British pound in September 1931.
* I hare been deeply influenced, In this interpretation, by what I have leaned from
the massive researches of Friedman and Schwartz, Monetary History of the United
Stote(196?). What happened in 1929-33 has been made much clearer by them than
it was before. The monetarist form in which their analysis is put has little effect on the
substance of their argument,
' \ cannot resist inserting a personal recollection. In those years I was a junior lec-
turer at LSE, I had been working on labour problems; I knew very little about money.
 THE OLD T R A D E C Y C L E IOI

From the point of view of a country which took this course, it was
in itself a release. For the position of its Central Bank, in terms of its
own currency, was now secure; it was free, as soon as it was clear that
internal confidence in the currency had not been damaged, to take
reniedial action. So it was that already in the summer of 1932. rates
of interest in London began to come down. It took time for this to
exercise much effect on activity, but it was a first step on the road to
recover}',
It must however be emphasized that the condition laid down, of in-
ternal confidence in the currency not being damaged, was essential
for this (relatively) favourable outcome. One needs to recognize that
for a country such as Germany (still at that time a constitutional
republic) the British way out was probably not open. For the Great
Inflation, which had destroyed the internal value of the German
Mark, was then less than a decade away: confidence in its successor
could not be counted on; a flight from it could be easily aroused. So it
may well be that the only course which was open was rigorous
exchange control—which prepared the way for Hitler."
Whichever way it was that other countries reacted, the effect on
the US itself was much the same. The overseas assets of American
banks, which they had been encouraged to build up (to help in post-
war reconstruction) were suddenly devalued in terms of dollars. So
there was another round of American bank failures, continuing
through 1932—in America the worst year of the Depression—and
culminating in the general closure of banks, as Franklin Roosevelt
took office, in March 1933. That was the end of the old 'Fed', The
American Government had to take power to support the banking
system, details of which I do not have to discuss. Other countries, in
one way or another, followed this example. The old style of financial
cycle, which has been the subject of this chapter, thus came to an
end,
In July 19J1, at the time when the monetary crisis erupted in Germany, i met our
monetary expert, Theodore Gregory, who had just been working with Keynes on the
Macmillan Committee, I asked him, in my innocence: what does this mean? The
whole of Europe.' he said, 'will be off the gold standard within a week,' He was very
well informed: if he had said two months he would have been about right. That has
always been a lesson tt> tne on economists* predictions.
* ! may support this with another anecdote. In September 19 J I , Ursula Webb, who
was to become my wife, was studying in Vienna. She used to relate how the famous
Austrian economist, L. von Mises, proclaimed at his seminar: 'In one week, England
will be in a hyper-inflation!' That was ridiculous, applied to England; I used, for many
years, to think it WBS ridiculous. But would it have been so if applied to A ustria™-or
Germany? Recent experience in South America seems to leave this an open question.
 12 The Credit Economy: Wicksell

The issue between the two theories of interest, which was discussed
in Chapter 9, could, In the light of what has now been said, have
been looked at in another way. Each made interest determined at a
particular margin, but the choice margins to which they directed
attention were different. For Marshall it was a choice between bonds
and goods—consumption goods on the savers' side, investment
goods on the other. For Keynes it was a choice between holding
bonds and holding money, a non-interest-bearing money. Keynes-
ians often insisted that since it was a money rate of interest that was
in question, it was the latter margin which must be of decisive im-
portance. If Marshall was to determine a money rate of interest in his
manner, he must be assuming that the price-level of (either sort of)
goods was fairly stable, or at least that by his 'savers' and 'investors'
no persisting change in it, over a relevant future, was contemplated.
At the time when he was writing, would that ha\'e been such a bad
assumption? He would nevertheless have been obliged to admit, if he
had lived to face the issue, that it had become a much worse assump-
tion by the time of Keynes. And, in spite of the invention of index-
linked bonds, it is not a very appealing assumption now.1
Nevertheless, if Marshall was put out of date by Keynes, has not
much the same thing happened to Keynes's own theory, at least in
the form he gave it in his most famous book? It is just that form of his
theory which is affected; the more general theory of choice among
assets, as set out in Chapter 8 (deeply influenced by his work)
remains. What has happened is that the non-interest-bearing
money, on which his argument depended, has changed its character.
Of course it still exists; but it has largely lost its function as a reserve
1
If it were the case that all (or most) medium-term lending was index-linked,
Marshall might come hack into his own. It would certainly be a real rate of interest
which would be determined in his manner; is there any other way it could be deter-
mined? I suspect that it is a model of this type which is often in the minds of economic
commentators; and in their context I am prepared to give it some countenance myself.
It Is to their world that It seems to belong, I cannot however believe that there are
many actual financial transactions which proceed in terms of (even subjective) real
rates. If they did, then surely index-linking would have made more impact than it
seems to have done.
 THE C R E D I T E C O N O M Y : W I C K S E L L IOJ

asset, at the most liquid end of a spectrum of assets, the position

which at the time of Keynes it was natural to think of it as occupying.
When short rates of interest were as low as they were at that time,
there seemed to be continuity between cash, which did not yield
interest, and bills (and time-deposits) which did, but so very little. So,
as we have seen, it was natural for Keynes and his immediate fol-
lowers to take them together, neglecting the fact that some deposits
did bear interest, treating them all as money.
The much higher (short) rates of Interest to which, in so many
countries, we have become accustomed in the seventies and eighties,
have made this no longer possible. It is no longer so nearly a matter
of indifference whether liquid reserves are held in cash or in 'shorts'.
Any rational operator will get out of holding cash, as a reserve, as
completely as he can. The cash that he still goes on holding is now
not a reserve asset; it is a running asset, like the goods in process in the
balance-sheet of a manufacturer. It is dependent on the money
values of the business he is doing; but there is not much room for
choice about the ratio between them, A gulf has opened between
that circulating money and the most liquid of his reserves.
It was tempting, once this had been perceived, to suppose that the
way was open to a much simpler system of control than either
Keynes or Marshall had contemplated, that to which the name of
'monetarism' has been given. If the circulating money could be iden-
tified, and if the velocity with which it entered into the purchase of
output was independent of economic incentives (so that it could be
taken to be fairly constant), control of the quantity of that circulating
money should enable the value of output (Keynes's Y) to be con-
trolled. But how should this quantity be controlled? There would be
no direct way that was open to the banking system for controlling it.
All it could do would be to adjust its lending policy, changing the
rate, or rates, of interest that it was charging to borrowers (possibly
also changing a system of 'credit rationing' that it was practising),
watching the quite roundabout effects of these on F, then accepting
the effects of these on circulating money. Monetarism gets cause and
effect the wrong way round. So it offers no short-cut.
At modern rates of interest, to hold barren money even as a run-
ning asset has become costly; so it must be expected that means for
economizing in it will be looked for and will be found—as surely they
have! We are on the way to a credit economy, in which any money
that does not bear interest has become no more than small change,
IO4 PROBLEMS AND POLICIES

or petty cash. It is surely as least as tolerable a simplification as

others to which an economic theorist is accustomed to take it that
this has already happened. That is what I shall do in the rest of this
chapter.
Money remains of course a standard of value, in terms of which
people do their calculations, and in terms of which debts are
expressed. But money as means of payment is just a debt. The pay-
ment of a debt is an exchange of debts. We regard it as payment
because the debts have different quality. It is quality from the point of
view of the creditor that matters. I pay my creditor with a cheque on
a bank; he accepts it, because he has more confidence in a debt from
the bank than he has in a debt from me. Even if he kept his own
account at a different bank from mine, he will usually accept it as a
money payment; debts from the one bank and from the other have,
from the points of view of all concerned, equal reliability. If however I
had drawn my cheque on a bank of which he had never heard, and
which (so far as he could tell) might be purely imaginary, he would
not have accepted it as a payment.
One can construct, in the light of this consideration, two pure
models of a credit economy, each of which has its uses. I call them
the monocentric and polycentric respectively.
It is characteristic of a monocentric model that it has just one
central entity, promises to pay by which have superior quality, or
reputation, than those of any other entity. The continuance of that
superiority is taken for granted, so that there is just this one 'monet-
ary authority'. The polycentric model has no such single centre. But
there will still be differences in the qualities of promises by different
entities: so there will be some, at any particular moment, which have
highest quality. If there is one which at the moment has the highest
quality, it acts, for the moment, like a monocentre; but there is no
certainty that it will retain that position, ft may be, on the other
hand, that there are several that have established an equal reliabil-
ity, each maintaining a willingness to convert its promises, at a fixed
rate, into those of the others. If there is perfect convertibility, each
has some of the properties of a monocentre, but there is no single
'monetary authority'.
It will surely be noticed that a monocentric model is likely to have
most relevance to the problems of a national economy, especially
when the international aspects of that economy—external trade and
capital movements-—are being left out of account. The polycentric
 THE CREDIT KCONOMY: WICKSK1.I. IO>

model has most relevance to International problems, ! shall have

enough to do in the present chapter If I confine myself to the former.
The latter is postponed to Chapter 14, since it needs more prepara-
tion.
The simplest example of a monocentric model is that constructed,
not, long after the time that Marshall was writing, by Knut Wicksell.2
There is just one bank, providing credit money, the only money there
is, and there are no other financial bodies. The only form, in which
savings can be held is as deposits in the bank, and it is only from the
bank that those undertaking investment can borrow. The only
means of control that is available to the bank is the rate of interest it
pays (and charges). This is the basic money rate of interest.
It is tempting to say that a system such as this would be in equilib-
rium over a period, if the net increase in the volume of loans being
made by the bank was equal to the net increase in the volume of
savings deposits. For this would mean that the volume of money that
was circulating outside the bank would be remaining unchanged
over the period. Wicksell himself was sometimes inclined to this
'monetarist' interpretation of equilibrium, but the model is by no
means committed to it. There is another criterion, which also makes
its appearance in Wicksell's work, according to which the system is
in equilibrium if the price-level is remaining constant. The equilib-
rium rate of interest is then that which niaintains constant prices.
But there is more in that direction that can be said,
There are after all many indices of prices, which do not always
move together; so there should be an equilibrium rate of interest cor-
responding to each. One of them is a wage index, so there should be
an equilibrium rate which would keep the wage-level constant. Or if
wages are sticky, it could be interpreted as that which would give a
desired level of employment (in the manner of Keynes). And if it is
not the wage-level which has become sticky, but a conventional rate
of rise in the wage-level which has become sticky, there could be an
equilibrium rate of interest which (in the same sense) would it that
rate of rise in the wage-level. To keep the money rate of interest at
that level would not cure inflation; all it could do would be to ensure.
by being high enough, that monetary policy was not in itself aggrav-
ating the inflation; and, by being low enough, that the activity of" the
1
Geldzins und Guterpreise (1898); the English translation. Interest and Prices, was
not available until 1939. Its influence on English economics goes further back, for to
the thirties there were some of us who did read German!
IO6 PROBLEMS AND POLICIES

economy was not being depressed by monetary policy. All these

things are in sight of what I may call the original Wicksell model,
But this is not the only direction in which it needs to be developed.
For what about uncertainty, and costs of making transactions,
which I have repeatedly emphasized to be at the heart of monetary
theory? Let us try to put them back and see what happens,
Not much needs to happen on the side of the savers. Their deposits
in the bank are certain; they can deposit what they like and will get
the rate of interest which the bank is offering, to all alike. But on the
side of the bank's advances it is a different story. We must no longer
think of the bank just fixing a rate of interest at which it is willing to
lend, letting anyone who is prepared to pay that rate have the
money. The bank will have to attend to the prospects of particular
borrowers and to their character; so whether, in its judgement, they
can be relied on to repay. This is a question of the information xvhich
is available to the bank.
One can conceive of a situation when the bank was receiving
deposits but was finding it difficult to find suitable borrowers. It
would thus be withdrawing money from circulation; this could go on
to such an extent that in none of the senses that have been dis-
tinguished could there be an equilibrium. Should one then say that
the money rate is too high? That would only make sense if a way out
could be found by lowering it. But to lower the rate which the bank
was charging to borrowers would not automatically help it to find
new trustworthy borrowers; it does not itself provide an automatic
way of finding them, (How far it provides a sufficient incentive for
existing borrowers to borrow more is a matter I shall be considering
in the chapter that follows.) To lower the rate it pays on deposits may
do something to discourage deposits; but since savings deposits will
for the most part be held as reserves, even a zero rate on deposits may
fail to choke them off to the extent which is required.
We are thus already in sight of the famous crux—an excess of
"saving" over Investment' that by interest policy cannot be righted.
But it is too soon to look for a Deus ex machines. We should first con-
sider whether we have not been placing too great a responsibility
upon our myopic bank.
To insist, as in this First Revision of the Wicksell model we have
been doing, that all borrowing and lending must take place through
the bank, is quite unnecessary. If the bank is unable to find suitable
borrowers, why should not some of the savers search them, out for
 THE CREDIT ECONOMY: WICKSELL K>7

themselves? There is no reason why we should deny them the pos-

sibility of making more direct contacts. In so far as they do this, they
will have less to deposit with the bank; so the surplus of funds, which
we have been supposing to be going into the bank, will be moder-
ated. It should however be noticed that since the alternative of
depositing with the bank remains open to the savers, and the liquid-
ity of their tendings to the firms (as it will now be convenient to call
them) must be less than that of their deposits in the bank, the rate of
interest which is offered by the inns* must be greater than the
deposit rate of the bank. This will act as a minimum to the system of
interest rates which in this Second Revision begins to develop.
It is indeed this deposit rate which acts as the king-pin of the
system, playing much the same part as was played by the money rate
in the original Wicksell model. For the alternative of depositing, at
the fixed deposit rate of the bank, is always open to the savers, while
the alternative of borrowing directly from the bank is not so regu-
larly open to the firms. We need not exclude the possibility of the
bank doing some direct lending to the firms—at a rate which (we
should now say) must be somewhat higher than the rate which it
was paying on deposits. For the mode! is now to allow for transaction
costs; the bank will have costs of administration and these must be
covered. The firms will get what they can at this rate, which we may
fairly suppose will be lower than what they pay on direct loans from
the savers: but they want more than this, and it can be got from the
savers.
It will be noticed that in this Second Revision the bank is less
'central* than it was in the First. But it still retains some power of
control, through its deposit rate.
The solution that has so far been found for the information prob-
lem is very imperfect; it has really added nothing more than an
opportunity for the firms to raise some part of the funds they require
by borrowing from their friends. A modern economy does not rely
much on that, though there have been times in history when it has
been important. Its place has largely been taken by an alternative of
much greater potency—the introduction of financial intermediaries,
which make a direct attack on the information problem.

1
Or its equivalent, in terms of prospect offered, if the direct lending to the firms
takes the form of subscription to equities.
IO8 PROBLEMS AND POLICIES

The financial intermediary can prosper* if it can make use of

specialized knowledge about the prospects of particular kinds of real
investment, so that it can make advances to firms, or Investments in
the securities of firms, which the bank would not know were sound
investments; and if it can acquire resources which enable it to make
those financial investments at less loss of liquidity than they would
entail upon the private saver. But it cannot prosper unless it makes a
profit; this implies that it must borrow at a lower rate than that at
which it lends, there being a sufficient difference to cover its adminis-
trative costs, and to compensate it for the additional risk with which
it, in its turn, is involved in every extension of its operations.
Thus its in-rate (as we may call it—the rate at which it borrows)
and also its out-rate will have to be fitted into the structure of rates,
which had already appeared in our Second Revision. It is not neces-
sary that the in-rate of the financial intermediary should be higher
than the rate which is paid by the firms (directly) to the bank, or by
the firms (directly) to the savers. For it may be expected that the
intermediary may be able to attract funds from the bank and from
the savers, by offering a greater degree of security (by pooling of
risks) than the firms could do directly. And it need not necessarily
charge a lower out-rate than is charged by the bank, since it will be
willing to do business with the firms which the bank would not do. It
is however clear that its in-rate must be appreciably higher than the
bank's deposit rate, and its out-rate must be higher than its in-rate, if
it is to function at all.
It will surely be granted that this Third Revision has brought our
post-Wicksell model quite a step nearer reality. The particular insti-
tutions which play the parts I have assigned to 'bank' and 'inter-
mediaries' (even to 'firms' and 'savers') will no doubt differ from
country to country, and from one time to another. But that some will
be found who play something like the parts in question can hardly be
doubted, though it may take some thought to find them.
For instance, I accept that the bank, as it appears in a Wicksell or
post-Wicksell model, differs from a Central Bank, as that appears in
practice, in a way which would appear to be essential. We would not
expect a Central Bank to accept deposits from all and sundry; so it
does not have a deposit rate which it can use as a means of control. If
* J here borrow a passage which first appeared in my Capital and Growth (1965,
pp. 266-7) and was also used in the paper 'The credit economy', written for the
second volume of my Collected Essays, on which the whole of this chapter is based.
 THE CREDIT ECONOMY: WICKSELL IO9

it accepts deposits (or what amount to deposits) it is only from a

limited circle that it is willing to accept them. (So there are inter-
mediaries on this side also.) Nevertheless though its power as a
borrower is hardly ever exercised, it remains as a sanction which
could be brought into play.
It is evident from British experience (which I cannot help having in
mind) that Minimum landing Rate, or Bank Rate as it used to be called,
does not have to be 'effective'; actual market rates, for loans on sim-
ilar conditions, would often be below it. In order to make it effective,
if it desired to do so, the Bank would have to find some means of
withdrawing funds from the market—that is to say, in ray termino-
logy, of coming in as a borrower.* One must suppose that it was
taken for granted that some way of doing this would be discoverable.
For once that was accepted, a rise in Bank Rate, taken under con-
ditions and in such a manner* as to indicate that the Bank 'meant
business', would be followed by commercial banks and by others
within the 'circle', without it being necessary for the sanction to be
actually used. Once there was this understanding, it would be the
whole 'centre' of the banking system (consisting not only of what
was called the Central Bank but also of what had become its satel-
lites) which would be acting like the single bank of the Wicksell
model.
Indeed, as this shows, the model when taken strictly is one of a
system in which all transactions are 'at arm's length', all participants
acting independently. It is useful to have such a model, not only
because there are people—important people!—who think it to be an
ideal arrangement, but also because it is a useful basis with which
other arrangements can conveniently be compared. It is, I think,
shown up when it is used in that manner that it may be far from
being an ideal arrangement.
For let us consider further how an arm's length system could
work. I will simplify a little ([ think it is only a little) by supposing
that marginal funds—those needed for a marginal expansion in an
investment programme—will always be raised by the firms from the
intermediaries, and by the intermediaries from the bank. There is

! it is not evident that mere abstention from renewal of short-dated lending could
be relied on to be sufficient.
* The old Bank of England convention, that a half-point rise in bank rate showed no
more than that the Bank was just following the market, white a whole-point rise was
meant to give a lead, had its uses as a simple signal.
110 P R O B L E M S AND P O L I C I E S

then a straightforward sequence. The savers deposit in the bank at

an interest rate r0, the bank lends to the intermediaries at an interest
rate r l t and intermediaries lend to the Irms at r2. Rate r, must exceed
r0 to cover the administrative costs of the bank; r2 must exceed r t not
only because of administrative costs, but also to provide a liquidity
premium. So r2 exceeds r0 by two margins, those of the bank and of
the intermediaries.
And having gone so far, we may allow ourselves to go further. Let
us suppose that there is a rate R, which will stand for the return
which the firms expect to get from marginal investment; this will
have to exceed rt by another liquidity premium. We could then say
that it would be R which, from the point of view of the Wicksell con-
struction, would need to be kept at an equilibrium level, the level
which was appropriate to the kind of equilibrium which it was
desired to attain.
All this is involved in an arm's length system. It is inherent in it
that there should be a gap, depending on liquidity preferences of the
various parties and also on administrative costs, between anything
that can be directly controlled by an interest policy of the bank and
the yield on investment (R} which should be the key to equilibrium. If
the gap is narrow, and can be relied on to be narrow, interest policy
can be quite effective; but if the gap is wide and undependable, there
is a formidable obstacle in its way,
We should however remember that the problem is an information
problem. In order to make wise decisions on the big issues which
may here be in question, many sorts of information need to be gath-
ered; they can hardly be gathered without having many listening
points. The financial intermediaries are listening points; the problem
is one of transmission, from them to the centre. It is not obvious that
an arm's length system, using the 'price-mechanism*, must always
be the most efficient, or most 'economic', way of doing this job. It is
not always the only way that is open for doing it. When the inter-
mediaries are well established, the relations between them and the
centre can be much closer. It seems to follow from what has been
said that it is desirable that they should be closer. Control must have
power if it is to be effective. Whatever we think about monopoly and
competition, in the rest of the economic system, there are good
reasons in this monetary sphere for not being afraid of some concen-
tration.
Whatever the links between the centre and the intermediaries, the
 THE CREDIT ECONOMY: WICKSELL I I I1

collection of information, the information required, can ne¥er be per-

fect. The system will still be subject to shocks; things will happen
which no one, who was in a position to take action, had foreseen, It
is a system which is based upon arrangements for risk-taking in the
face of an uncertain future; as such it is bound to be fragile. An arm's
length system is particularly fragile. A serious blow to one part of it
can have wide repercussions. Closer association, by making it easier
to find a lender of last resort', reduces the fragility.
 13 Interest and Investment

The post-Wicksellian models, with which we have been dealing in

the preceding chapter, have been concerned with the transmission of
signals from the centre to the 'circumference' of the monetary or
financial system; their effects on the 'real' economy, outside that
system, have been left unexplored, WlckselTs preoccupation with
long-run equilibrium made it hard for him to have anything useful to
say on this matter; Keynes was not troubled by that obstacle, but
there was another which, as we shall see, got in his way. There are
ideas which have come up since the time of Keynes which give more
help.
They can best be introduced by some consideration of a theory
which was available to Keynes and which he is well known to have
rejected—rightly if it is taken as he understood it, but if he had looked
round to see what could be made of it, not so rightly. This is that
which is associated with the name of his contemporary, R. G.
Hawtrey.' Hawtrey's Currency and Credit appeared in 1919; so it may
be reckoned to have been a standard work on monetary theory in the
decade before Keynes's Treatise. It was Hawtrey's doctrine that the
principal way in which interest affects trade activity is through its
effects on speculative markets in commodities. I have given a sketch
of the working of such a market in Chapter 2. I have no reason to
suppose that either Hawtrey or Keynes would have disagreed with it.
It appeared that in such a market, positive quantities of stock being
held, current price would tend to equal expected price for some
future date, minus cost of holding to that date; and in that cost, inter-
est would be a constituent part. Thus if the expected price was given,
a rise in rate of interest (here, presumably, a short rate of interest
closely tied to central bank rate) would cause a fall in current price in
correspondence. We may call this the arithmetical Hawtrey effect; it
was this that most of his readers in the twenties, Keynes in particu-
lar, thought him to have had in mind. If that had been all, it looked
easy to crush it. No more need be said than what Keynes said in the
Treatise:*
1
I shall be drawing on my paper on Hawtrey in Economic Perspectives (1977).
2
Treatise, Volume I,pp, 194-5.
 INTEREST AND INVESTMENT II.-j

(Hawtrey relies] exclusively on the increased costs of business resulting from

dearer money, [He] admits that these additional costs will be too small
materially to affect the manufacturer, but assumes without investigation
that they do materially affect the trader. He does not base his argument on
the arousing of an expectation of falling price-levels in the minds of the
dealers. ... Yet probably the question whether he is paying 5 or 6 per cent
for the accommodation he receives from his banker influences the mind of
the dealer very little more than it influences the mind of the manufacturer,
as compared with the current and prospective rate of off-take for the goods
he deals in and his expectations as to their prospective price-movements.
So, for Keynes and subsequently for Keynesians, that was that!
Hawtrey however did not admit that he was defeated; in sub-
sequent works5 he returned to the charge. Chiefly he maintained
that there was much historical evidence, which he assembled in
detail, to show that changes in bank rate had, or often had, much
more influence on markets than Keynes would allow. (It should be
noticed that this evidence largely related to the period of what was in
Chapter 11 called the old trade cycle.) For this he had an explana-
tion, in what he called 'psychological factors*. That was a bad term,
but a better has since been discovered. TLe issue was one of the
rationality of expectations.
Neither he nor Keynes disputed that 'expectation of prospective
price-movements' is a dominant influence on speculative markets.
But how are such expectations formed? There were at first sight two
alternatives. Either they are based on information, available in the
present, or they are quite random. Close-up observation of markets
led Keynes to prefer the latter alternative; markets so often appeared
to move in ways that no one could explain. Speculation, on this view,
is just a gamble. A 'herd instinct' would be invoked to account for the
phenomenon of bull and bear markets. But, Hawtrey might rejoin
(though I don't think he did so explicitly), are these necessarily indi-
cations of irrationality?
For suppose one regards perfect rationality and perfect irrationa-
lity as two extremes, leaving plenty of room for something in
between. Rational expectations must be based on information; but
the information may be good, or much less than is required. Or it
may be good in some directions but deficient in others. (In terms of
my 'harvest' example in Chapter 2, the date at which better informa-
tion about a forthcoming harvest would be available could be known
s
Chiefly in A Century of Bank Rate (1939).
114 P R O B L E M S AND P O L I C I E S

fairly exactly; that would be a help to the stockholder, but by no

means all he would like to have.) That is the kind of thing I mean by
rational choice on imperfect information,
We can thus admit that if relevant information is hard to come by,
the market may well approximate to what Keynes called it—a
'casino'. If the most that is available to the Individual operator is to
notice what others are doing, the 'herd' phenomenon could be
explained, without abandoning rationality altogether. Short-run
expectations would then be formed from extrapolation of current
price-movements; in a fluid market, from which it is easy to extract
oneself if one makes a mistake, it may be quite rational to proceed in
that way. But even so, it would be rational to change one's expecta-
tion on the receipt of new information; and Keynes did elsewhere
admit that this might happen. A change in bank rate, he said him-
self, may 'constitute a new fact—by throwing new light on the policy
and intentions of the monetary authority'.4 That really gives
Hawtrey what he needs; as he put it,*
When the use of bank rate to restrict credit became an established practice,
traders, being aware of the intentions of the Bank, were inclined to anticip-
ate them. When bank rate went up from 3 to 4 per cent, a trader would
reason that this was intended to have a restrictive effect on markets, and
that, if the effect was not brought about, the rate would simply go higher and
higher until it was.
So, if a rise in bank rate was taken in such a way, and in such con-
ditions, that it would be taken to be intended to have a decisive effect,
it would have that effect, even though the arithmetical effect which
could be calculated to follow from it was small.
I therefore conclude, in spite of Keynes, that Hawtrey's channel is
one of the channels along which a change in monetary policy,
expressed as a change in central bank rate, may take effect. In the
days of the old cycle, to which most of Hawtrey's examples refer, it
could well have been a major channel; in latter days, when so many
commodity markets are managed in the interests of producers, it is
less impressive, at least if taken literally. It should however be noticed
(as Hawtrey himself observed) that when currencies are floating,
speculation on foreign exchanges works in a similar manner. A rise
in central bank rate in a single country, if it is taken decisively, has its
* Treatise, Volume 1, p. 202.
' Century, p. 279.
 INTEREST AND INVESTMENT 115

most unmistakeable effect on the external value of that country's

currency. Effects on commodity markets are mixed up with that. And
there is no question that a definite change in the exchange rate has
an effect, at least on industries that are exposed to foreign competi-
tion,
Hawtrey's own precept was not for unlimited floating, but for
floating about a parity, which was to be kept at least relatively firm.
That in his view would retain a good part of the controllability of the
old cycle. It is effectively the arrangement which has been formalized
in the constitution of the European system (EMS), So, granted this
interpretation, the Hawtrey channel would appear to have survived.
I turn to consider the other channel, which plays a corresponding
part in the work of Keynes. This was a matter of the impact of the
(long) rate of interest on fixed capital investment. It will prove to
have been useful to approach it, as I have been doing, from con-
sideration of his controversy with Hawtrey, since it can hardly be
doubted that this was the route along which Keynes was led to it. He
had shown to his own satisfaction, that the Hawtrey effect was not
strong enough to bear the weight that Hawtrey had laid upon it. As I
showed, this objection ran in terms of its calculated arithmetical
effect. Keynes had no difficulty in showing that in the case of invest-
ment in fixed capital the arithmetical effect would be larger.
It will however be useful to have some figures in mind. Take a rise
in interest from 5 to 6 per cent, such as both of our authors appear to
have been considering. A Hawtreyan merchant, planning to sell at
the end of six months, and perfectly confident of the price at which
he would be selling, would reduce his current demand price in the
ratio of 97| to 97, approximately, that is, by f per cent. A Keynesian
manufacturer supposing (a very extreme case!) that he thought his
machine would last for ever, and was confident that it would go on
yielding the same return, would reduce his demand price in the ratio
of 1 to 6, nearly 17 per cent. But that sets weight on what is to
happen in the distant future, to an extent which is quite ridiculous. A
similar calculation, with the machine expected to last for no more
than ten years, seems to result in a fall of about 5 per cent. That is
more reasonable, and it could be sufficient for Keynes.
It would nevertheless have been desirable to ask; how much of
industrial investment would fit into anything like that pattern?
Perhaps the most promising case is that of the 'speculative* builder of
dwelling-houses, houses which he will let out for rent. The value to
Il6 PROBLEMS AND POLICIES

him of a completed house could then be taken to be the capitalized

value of the stream of rents which he would charge. Houses being
quite durable, this would be fairly sensitive to changes in the .(long)
rate of interest. The same would hold, in a slightly more complicated
manner, if the house is to be sold to its prospective occupier, who has
raised the money to pay for it on a mortgage. A reduction in the
interest rate, making a mortgage loan easier to get and to pay for,
would surely stimulate the demand for new houses. (This was in fact
the most obvious effect, in the 1930s in England, of the reduction in
interest which actually occurred at the time when Keynes was
writing. There was a boom in the construction of private residential
houses.)
There is however a general principle which in this latter case
already comes up, A person who raises the money to buy a house
which he is himself to live in, whether by borrowing on mortgage or
by selling securities he had been owning, puts himself in a less liquid
position; so a question of liquidity comes in here, no less than in the
financial markets. The toss of liquidity is moderated if the house that
has been bought can readily be resold. It is thus of much importance
that facility of borrowing does not only increase the demand for new
houses; it has a similar effect on already existing houses; but greater
activity in the market, for such houses makes a house a less illiquid
asset, so as well as the direct there is this indirect effect.
Though there are some durable instruments employed in industry
for which much the same is true as is true in the case of dwelling-
houses (transport vehicles are perhaps a leading example), it is surely
more common for the instrument to be associated with an existing,
or planned, process of production. It is expected to play a part in that
process of production; it is much more valuable in that place than it
could be in any alternative use. The price that the firm is prepared to
pay for it will then be hardly affected by what could be got for it on
resale. The 'machine* is a single member of a bundle of complements.
Two quite different problems of choice then make their appearance,
according to whether the investment is defensive or innovative.
We shall need a definition of this distinction. Every investment in
capital goods has to be paid for, so its effect on the balance-sheet of
the firm undertaking it has two sides. On the one hand there is the
acquisition of the good, a new asset; on the other a new liability, or
the giving-up of a financial asset. Whichever way the expenditure is
financed, there is on that side a loss of liquidity, an increase in ex-
 INTEREST AND INVESTMENT IIJ
6
posure to risk; this will naturally be less formidable if the funds can
be got on easy terms. But the corresponding effect on the other side
may go either way. There are some kinds of investment in fixed
capital which diminish in this direction the riskiness of the firm's
position; these I call defensive investment. There are some which in-
crease it; these I call innovative,
The simplest kind of defensive investment is mere replacement.
The new 'machine* takes the place of a predecessor that is worn out,
or is wearing out. Whether or not to replace will not often be affected
by interest, or by the terms on which funds can be raised; the gain
from having the whole plant in proper working order must usually
outweigh any interest cost on a particular machine. A change in
interest may indeed affect the date at which the new machine is
ordered. At a high rate of interest it may well pay to hang on longer
with an appliance that, is bound in the end to be condemned. But to
attach much importance to that sort of saving is hardly symptomatic
of good business management.
That interest does not have much to do with pure problems of re-
placement is I think widely accepted; but it is not so often noticed
that the same holds much more widely. Any existing process of pro-
duction may be in need of reinforcement, for other reasons than
physical wearing-out. So there are many kinds of defensive invest-
ment. Some may arise from a desire to improve relations with labour
employed, which have been deteriorating; facilities provided to im-
prove 'working conditions' are quite likely to require some capital ex-
penditure. Most important are those which arise from competition.
The introduction of a new technique we should reckon as innovat-
ive, in the case of the firm which first introduces it; but it is defensive,
in the firms which find themselves threatened by its competition, and
must follow' in order not to be left behind. They must follow, what-
ever the cost, or drop out of the race.
The effect of interest on each kind of defensive investment thus
seems unlikely to be considerable,
I accordingly turn to innovative investment,7 where there are
three types to be distinguished. In all of them, quite apart from the
loss of liquidity involved in the capital expenditure, there is increased
exposure to risk. It is difficult to think of any such increase in ex-
* See Appendix on risk and uncertainty,
' f have been helped in writing this section by the book of Areendola and Caffard,
The Innovative Choice (19881. which came into my hands as I was starting to write it.
Il8 PEOBLEMS AND POLICIES

posure which does not involve the construction of a new plant (or
department of plant); so I shall take it that that is always involved.
The plant has to be constructed (and that takes time) before it can be
used. So there is a construction period and a utilization, or running,
period which need to be distinguished. It is in the running period that
the gain, which is the incentive to make the innovation, must ordin-
arily be expected to accrue. My three types are classified according to
the form in which it is expected (or hoped) that it will accrue.
In Type A, which may perhaps be regarded as the most innovative,
there is not only a new plant but a new product to be produced from
it—a product that has not been produced by anyone before. Here
there are not only risks on the production side, of neither construc-
tion nor running coming out as intended; there are also risks on the
side of selling the product, of finding a demand for it.
In Type B there is no change in the character of the product; the
gain is expected to come from an increase8 in capacity to produce it.
In Type C there is again no change in character of product but also
no change in planned output of it. The gain from the installation of the
new plant must be entirely a matter of reduction in running costs.
With these three types in our minds, each may be looked at
further, in comparison with the others.
The risks involved in investment of Type A, considered in general,
are very formidable. It faces not only the production risks, which are
common to all our Types, but also selling risks arising in a form
which makes them particularly hard to judge. The best information
about saleability must often come from experience (of somebody)
with a similar product. The more similar, the better the information;
but also the more similar, the harder it must be to break into the
market. Thus to start a project of Type A, on at all a large scale, must
usually be intolerably risky. What happens in practice is to start with
a small-scale pilot plant, which is not expected to be profitable, taken
by itself; the object in constructing it is not direct pursuit of profit, it is
'testing the water', gaining information. It is like an experiment in
natural science.
Success of the experiment gives evidence (not necessarily conclu-
sive evidence) that it can be profitable to repeat it, on a larger scale.
That is to say, it prepares for an investment of Type B. So, from the
point of view of the economist, the two may just as well be taken to-
" It must be an increase in capacity, for if the new plant is to have a smaller capacity,
its output could be produced from the old plant, so the gain falls under C.
 I N T E R E S T AND I N V E S T M E N T 119

getfaer. The experiment may be reckoned as part of the construc-

tional stage of the A + B investment. It lengthens out this
constructional stage, leaving the utilizationai stage of B as that at
which it is hoped to make a profit.
The whole project (A + B) will have a time-scale that is quite con-
siderable. Our entrepreneur, as we may now call him, will not want
to be obliged to carry his calculations forward into the distant
future—to the distant date at which the equipment may still be ex-
pected to be in physical working order—for by that time, he will say,
'anything may happen.' It is the nearer future to which he will want
to look. During that time his capital expenditure, beginning low, will
build up to a peak and then decline. While it is declining, a stream of
receipts, on the other side of his account, will start to come in and
then (it is hoped) build up to a steady level. As I have said, he will not
want to lay much stress on that last stage; he is likely to be more con-
cerned with the prior question—can he get by so as to reach those
calmer waters? The peak of his exposure may well come after the
peak of the capital expenditure; for though that is declining, the
receipts which are to offset it have hardly started. So the great ques-
tion for him concerns his liquidity, at the point when he is in sight of
his big output, and can begin to have confidence (of which he can
persuade his banker) in his ability to go on selling it.
It is easy to see, in the light of this approach, why testing a project
in terms of the length of time which will have to be taken for it to
'pay for itself* is popular in business. It is a way of approximating to
the true test—the calculation of the balance-sheet, as it is likely to be
at various dates after the project is undertaken, and considering
whether the inance involved is likely to be forthcoming. I have per-
sonal experience of a firm which did make its decisions in that latter
manner, and did so successfully.
A project cannot be started in the middle; and it would rarely be
advantageous to stop it in the middle, however high the rate of inter-
est rises. So it is the effect of interest rate on the starting of projects
about which we have mainly to think. If the entrepreneur is thinking
in the manner just described, what would chiefly matter to him
would be the state of his balance-sheet at the time of maximum ex-
posure; this at the start may be quite a way ahead. He can be surer of
being protected on that occasion if he finances the plan by borrowing
long, or by issue of shares; but that means raising funds which will
not for some time be spent. Thus, although it may be accepted that in
I2O PROBLEMS AND POLICIES

this case a fall in Interest, provided it gets through to easier con-

ditions for industrial borrowing, is conducive to capital investment,
the effect on that investment may be considerably delayed.
I turn in conclusion to my Type C. This may be more straight-
forward. For if no change in final output is being planned, the only
gain that can be expected from having the new plant is a reduction
in running costs. If the running inputs required for the one plant and
for the other are fairly similar in character, the gain should be calcu-
lable by technical experts, perhaps fairly firmly, The flow of returns
which may be expected from the capital expenditure should thus be
calculable, and the effect of a change in interest should be straight-
forward. It should be nearly as clear as in the case of building houses.
There are however two qualifications which have to be made.
One is that if the inputs required on the one plan and on the other
are notably different, a change in their relative prices—at the stage of
decision, a change in their future relative prices—can be upsetting.
The change may appear to be profitable at current relative prices; but
what if, during the lifetime of the project, they should change? Coal-
firing and oil-firing is obviously a leading example. Uncertainty
about such relative prices, as they will be in the future, is obviously a
deterrent to Type C investment, whatever happens to interest rates.
The other, assuming that the former does not arise, is that this is
precisely the case of the 'machinery effeci' of Ricardo. During the
construction period, while the old plant is still in operation, there
must be a net increase in demand for labour, the labour employed in
the construction being added to the unchanged employment on the
old. But afterwards, when the new plant takes the place of the old, its
construction phase being over, there must be a net diminution in
demand, even with respect to what it was before the construction
started. So if progress is to be maintained at all smoothly, and all
innovation is of this character, there must be a succession of such
investments, as (in effect) Ricardo saw.
Though the point was so well made by Ricardo, neither 'neo-
dassics' nor Keynesians gave it much attention. I may claim that to
my own work, especially in my later work, I have gone on attending
to it, in several ways,*
* It is possible to state it ta 8 'neoclassical' manner, but, only in terms of a three-
factor production function (old capital, new capital, and labour). The great neo-
dassics were not very good at that, but it can be done. (See my chapter on the produc-
tion function, at the end of Capital and Time: also the new paper on 'Elasticity of substi-
tution', in the third volume of my Collected Essays.)
 14 An International Economy

As I have throughout insisted, this is a theoretical book; its conclud-

ing chapters will be no less theoretical than the others. There have
however been practical questions to which some of them may have
had relevance; one to which this may have relevance is very topical.
Should Britain enter, or not enter, or partly enter, the European
Monetary System (EMS)? I shall not attempt any direct answer to
that question, I shall confine myself to the construction of a theoret-
ical model, which may have a bearing upon it—and on similar ques-
tions which have come up in the past, and will surely continue to do
so, as far ahead as one can look,
Thus what I am to mean by an international economy is one con-
sisting of a number of national economies, each of which uses its
own money. Within each nation are a number (which we take to be
a large number) of separate decision-makers, persons and firms.
private and public, which have commercial and financial relations
between them. The parties to a transaction may have the same, or
they may have different, national attachments. The significance of
the national money is that it is used (we take it that legal arrange-
ments compel it to be used) as standard of value in internal trans-
actions; but for external transactions, where the parties belong to
different nations, no such rule can be laid down. The contract which,
as we saw in Chapter 5, must in general precede delivery must be
expressed in terms of some money, acceptable to both parties; that
may be the money of one party or of the other, or some third money
on which the two may more easily agree,1 But whichever way it is
adopted, a market between currencies, on which one money is
exchanged for another, has to be brought in before the bargain can
be completed. It will be assumed, in nearly all cases that I shall be
considering, that such a market exists.
It is conceivable that a country might insist that all exports from it
should be invoiced in its own currency; but that does not often
happen, for what would be the point of imposing such a restriction?
The contract sets up a debt from importer to exporter; if there were to
1
It is sufficient in this context to confirm attention to bilateral transactions.
122 PROBLEMS AND POLICIES

be trouble in collecting it, it would be in the law of his own country

that a claim against him would have to be pursued, if it needed to be
pursued. To insist that the debt must be expressed in terms of the
exporter's own currency must make exporting more difficult, to no
advantage. To insist that imports should be invoiced in terms of the
domestic currency is a possible form of protectionism; but from the
point of view of the protectionist, it is not an attractive alternative.
We may therefore assume that the choice of currency in which the
debt is to be expressed is left to be agreed on by the trading parties
themselves. Each will have some objection to the use of the other's
currency, so it is very likely that it will be some third currency, in
which each has some confidence, that they will choose to use. It
might not be the same currency which was used in this way
throughout the whole international economy; sterling area, dollar
area, French franc area, rouble area, have certainly in practice been
heard of. But we may surely allow ourselves, for the purpose of our
model, to accept that it is just one currency which in this way is
generally used. A model that is constructed in this way will surely fit,
not too badly, a good many historical, and contemporary, facts.
Accordingly, in the model, there are two kinds of currency, and
two kinds of country corresponding. All countries but one have two
currencies that concern them, their own and the international; that
one however has only one currency, for the international money is
also its domestic. It will surely be understood, once this is realized,
that monetary problems and policies look quite different according as
they are regarded from the standpoint of the 'central' country, or
from that of one of the others.
Most of what I shall be saying will turn on this distinction, so it will
be a help to give the countries names. I do not wish to rush on to con-
temporary identifications, so 1 must invent the names. I thus propose
to call the central country (the currency of which, in Bretton Woods
jargon, would have been called the 'key* currency) Centralia, and its
money correspondingly the cent. It has not been so easy to find suit-
able names for the other countries and their currencies; but the
following, though at first sight surprising, should be easy to remem-
ber. When I want to turn round and look at the system from the
point of view of one particular non-central country, I shall call its
money, correspondingly, the penny; Penland will then do as the name
of the country where pennies are the domestic money. This has the
advantage that any non-central country can be called a land; other
 AN I N T E R N A T I O N A L L E C O N O M Y 123

lands than Penland will just be Otherlands. Otherlands of course do

not have a common currency.
Every country may be supposed to have a banking system: but
since we shall not be concerned with the internal structure of that
banking system, we may take it to be fully integrated-—all banking
business, within that country, is performed by that single bank. But
we had better not suppose that the bank has a monopoly of dealings
In foreign exchange. We can allow other people and businesses in
Penland to hold cents as well as pennies (including claims in these
denominations) so that a free market can develop between them. The
Penland bank may buy or sell on that market; but it cannot sell cents
unless it has a stock of cents from which to sell. It can 'create'
pennies if it chooses to do so; so it is only by its supply of cents that it
is constricted.
If its supply of cents is ample, it can 'peg* the rate of exchange, sell-
ing them, when their price on the market rises above a desired level,
buying them when it falls lower. If the peg is perfect, the exchange
rate being completely stabilized, the only source from which the Pen-
land bank could acquire cents would be Penland exporters, except in
so far as there was explicit borrowing of cents from non-Penlanders
(either Centralians or Otherlanders). Subject to a corresponding
qualification, the only reason why cents should be withdrawn from
the bank would be to pay for imports—imports from any country,
since all alike would have to be paid for in cents. Thus, subject again
to these qualifications, the bank's reserve of cents will rise, during a
period, by the amount of the country's balance of payments, if that is
favourable; or fall if it is adverse. This can continue unless, as a result
of a continuing adverse balance, the bank's reserve falls so low that
convertibility, at the pegged rate, is threatened.
If this happens there are of course two alternatives which are open
to Penland {its bank or the government which stands behind it) on
its own Initiative to take. On the one hand, it may be possible to take
direct measures to improve the current balance. Whatever form
these take, whether iscal or monetary, they must nearly inevitably
imply a restriction of imports, for it is easier to work on the import
side than on the export.2 On the other hand, the peg may be with-
drawn; the bank ceases to operate on the market for foreign
1
That devaluations are often followed, rather immediately, by an expansion of
exports (as the econometrists observe) only shows that the devaluation had been ex-
pected.
124 PROBLEMS AND POLICIES
exchange; the exchange is allowed to 'find its own level'. There are
some things which need to be said about each of these alternatives.
First, it is unlikely that there will be any action, not completely dis-
ruptive, which will quickly restore the current import-export
balance. Even the most drastic restrictions on imports will take time
to take effect, if only because those coming up to be paid for have
been contracted for in advance; the contract must be carried out, if
there is not to be a default on it. Taxes on imports are bound to take
time to exercise their full effect, since the domestically produced
goods which are (at least in part) to replace them take time to pro-
duce. Thus the authorities will be well advised to take steps before
they are compelled to do so. If they have not done so, a gap is left
which must be filled in some other way.
A depreciation of the exchange will also take time to affect the
current balance; but this is one of the cases where speculation can be
a help (see Chapters 2 and 13). Non-Penlanders, as we have seen,
have no reason for holding pennies as a transaction balance; but
they can have a speculative demand for them, if they expect their
value, in terms of cents, to be going to recover. This should prevent
their value on the market, in terms of cents, from falling indefinitely.
With this assistance, slowly acting remedies can be given time to
take effect.
It is important however to realize that what is in principle the
same alleviation can occur without a depreciation in the external
value of the currency, by a rise in the rate of interest on short loans
that is offered by the bank. This will have internal effects, within the
country, which will react on the current balance; but that effect also,
for similar reasons, is likely to be slow (see again Chapter 13), If how-
ever it is open to foreigners to lend (to the Penland bank or to other
Penlanders) at these high rates of interest, taking them to be higher
than can easily be got in other countries, there can be a (temporary)
movement of funds towards Penland, purchase of pennies by non-
Penlanders, which in favourable conditions can go a long way
towards filling the gap.
It must however be emphasized, emphatically emphasized, that
the effectiveness of these alternatives, either of these alternatives,
depends on a minimum degree of confidence in the stability of the
currency (of the value of the penny in terms of the cent) being main-
tained. If that confidence is lost, neither will work. It is only if it is ex-
pected that the penny will recover to its normal level or near it that
 AN I N T E R N A T I O N A L E C O N O M Y 125

there can be an incentive to speculate on its recovery; and it is only if

there is confidence that the peg wil! be held that it can be defended by
the offer of a high rate of interest in terms of pennies, which is all that
the Penland bank—without external support, not yet mentioned—is
able to offer. Thus while a pegged rate can readily be defended
against moderate and not long-lasting disturbances, its longer-run
stability depends on flexibility of internal prices and wages, or on the
absence of any disturbance which will lead to a continuing adverse
balance, when there is inflexibility.
I do not think that this principle is greatly modified if one brings in
the possibility of external support, as from a world bank or some such
institution. In terms of our model, this would amount to borrowing
from the Centralian bank. The loan would be expressed in cents, as it
would suit each party for it to be; and it would be natural for the
lender to insist that repayment of it, and interest upon it, should also
be expressed in cents. But if the borrower is to make a credible prom-
ise to make such payments, he must show that he is taking steps to
remedy the present weakness of his position, steps in which the
lender can be induced to have some belief. This can easily be repre-
sented as putting himself under the tutelage of the lender. Tutelage
tends to be resented: more bitterly resented, in this international
case, than in the basically corresponding internal case, where (as we
have seen in Chapter 7) a weak bank may find a way out from its
troubles by entering into dependence on a stronger bank. Thus while
as reinforcement to other means of dealing with a temporary emer-
gency, borrowing from an international agency may well have a
useful function, for dealing with more deep-rooted troubles the eco-
nomic dependency which it implies can hardly be prevented from
revealing itself as political.
I turn to consider, in corresponding terms, or in terms that shall
be as near as possible corresponding, the case of Centralia itself. If
Centralia has an adverse balance, it has to pay out cents; but the
supply of cents is under its own control. The value of the cent is
pegged at unity, whatever anyone does about it, So it is only in terms
of something else that it can conceivably be pegged.
Various plans for such pegging will doubtless come to the mind of
the readers; for are they not what discussions of international monet-
ary policy are commonly all about? I shall not venture to say much
about them. It is more to the point in this place to attend to the prior
question; what is to happen if there is no such peg?
126 PROBLEMS AND POLICIES

In its absence, if the monetary policy of Centralia is entirely pas-

sive, adjustment is left to be made by the other countries, the lands in
the terminology we have been using. If Centralia is running a sur-
plus on its current balance (not matched by bans which its people
are deliberately making to landers) some of the lands must be run-
ning deficits; the means that are open to them for dealing with those
deficits have been discussed. Though the problem in this case is more
global than that we were discussing, it does not appear to be of differ-
ent character. It is the other case, of a Centralia deficit, on which
there is more to be said.
If Centralia is running a deficit (again not covered by deliberate
borrowing) there must be at least some lands where bank reserves
are rising; but their banks are not obliged to take steps to deal with
their situation, apart from paying out domestic currency in
exchange for the cents their exporters will have earned. Let us con-
sider the position of one such country, which (since it is the one of
the lands to which we now desire to direct particular attention) we
may call Penland, as before. And we may for the moment confine
ourselves to relations between Penland and Centralia. The Centra-
lian deficit is matched by a Penland surplus.
When the issue is posed in this manner, it should set us asking;
how could the deficit have come about? There would seem to be two
principal possibilities which need to be distinguished, ways in which
a condition of deficit could succeed a former condition of balance. On
the one hand there might have been an improvement in the effi-
ciency of production in Penland relative to Centralia, so that
Penland's exports have become more competitive with Centralia's
exports, or are displacing domestically produced goods on the home
(Centraiian) market. Though if this happened there would be a stage
in the process which would show up as a balance-of-payment deficit,
it would be (at least technically) 'self-correcting'. There would be a
'deflation' of spending in Centralia and an 'inflation' in Penland—an
adjustment which in the Centraiian case might be highly disagree-
able, but which, if neither country was producing to capacity, might
not be reflected in a significant change in consumer prices, other
than of those where the effect was direct (such as exports from
Penland to Centralia),
The other way in which a Centraiian deficit might come about
would be an increased demand by Centralians for Penland goods,
which could happen independently of any change in production
 AN I N T E R N A T I O N A L ECONOMY 127
costs. The most interesting case of this is when it is engineered by an
expansionary policy of the Centralian government, which (since the
supply of cents is under its own control) is not obliged by external
constraint, to balance its budget—in any sense. We have seen that in
the absence of such policy, an Increase in competitive pressure from
Penland would lead to contraction in Centralia; there could then be
pressure for this to be offset by an expansion in public expenditure.
(A 'Keynesian' policy, whatever it was called!) In so far as the ex-
penditure was on Centralian products, it would go to offset the con-
traction in Centralia which would have occurred without it; but
some (directly or indirectly) must spill over into imports, thus result-
ing in expansion or 'inflation' outside. If (as at this point it would be
proper to do) we extend our model to admit the existence of a
number of lands (non-Centralian countries) there could well be some
for which the expansion was very welcome, while there could be
others, who were having difficulty in controlling their own inflation-
ary tendencies, to whom it would not be welcome at all. It is true that
they could in principle protect themselves by revaluing their curren-
cies in terms of cents; but that, for a country which (by assumption)
was already in a difficult position, could well appear to be a danger-
ous course to take.
It thus appears that when a number of countries are trading to-
gether, and the currency of one of them is accepted to be the inter-
national currency—this may happen, as we have seen, through
'market forces', without any treaty having been needed to establish
it—the others may be subjected to "imported' deflation or inflation
coming from the central country, even when in that central country
there is neither (its price-level being fairly stable).
Against this background, we may consider the question: is there
any device by which the central country could be taken down from
its privileged position, so that the choice of policies facing the various
countries could be made more similar, or as one might say equalized?
Surely, as a matter of historical experience, such a device has existed;
it was the Gold Standard. It was then believed that a currency
(whether central or not) was bound to be more secure if its value
could be kept fixed, or nearly fixed, in terms of some 'hard' or non-
financial commodity, acceptable in each country; that selected,
chiefly for historical reasons which need not here concern us, was
gold. It could however have been any other commodity, which satis-
fied the necessary conditions for a speculative market in it to be
128 PROBLEMS AND POLICIES

possible (see Chapter 2). Since gold is not subject to physical

deterioration when it is stocked, it is indeed particularly suitable for
that purpose; another commodity could however have been used,
though with a little more trouble. We should thus he thinking of the
government of each country setting up a department, associated
with its bank, committed to stabilizing the price of gold in the
national currency, buying and selling gold for national currency, so
as to keep the price of gold in that currency at the level desired. If its
supply of gold was threatening to run out, it would take steps to draw
funds from other countries, using them to acquire gold on the
markets of those other countries—the regular Gold Standard pro-
ceeding!
It is useful to notice that such an, arrangement is not inconsistent
with the currencies of one of the countries being central, in the sense
of our previous discussion. International transactions could be
carried on in terras of that currency, not in terms of gold. Neverthe-
less it would be particularly important, for the maintenance of the
Standard, that the gold price, in terms of that central currency,
should be held firm. Other currencies might be allowed to 'find their
levels'; but if the central currency was firmly based, the system
would possess a firm international money.
This was the system which .did indeed operate—broadly, though of
course with complications of detail—not so badly5 for nearly a cen-
tury before 1914. Throughout that epoch, it was sterling which
most nearly played the part of our central currency. That this was so
was dramatically demonstrated by an event which took place at the
end of it, at the moment of declaration of war in 1914. On the day
war was declared, the Bank of England raised its Bank Rate from 4 to
10 per cent, reckoning, it is to be supposed, that what was then a
scarifyingly high rate would be needed to hold the funds of non-
belligerents in London. But before a week had elapsed, it was brought
down to 5 per cent, nearly where it was before. For the expected
drain had not eventuated.41 think this can be interpreted as an indica-
* Its ability to accommodate itself to exogenous change in flow demand and supply
for gold (such as the hoarding by the new German government after 1870 and the
new South African suppliers in the nineties) is surely remarkable; it had more elasti-
city than is commonly supposed.
* See the account by J. H. Clapham, which (originally written for a planned third
volume of his history of the Bank in the eighteenth and nineteenth centuries), is now
available as an appendix to the continuation carried out by R. S. Sayers (Volume 3,
pp. 31-45).
 AN I N T E R N A T I O N A L E C O N O M Y 129

tion that even then sterling was something of a centre. And since it
was expected that the war would not be long continuing, it would
continue so to act,
That expectation of course was falsified. What destroyed the een-
trality of sterling, as the war continued, was the imposition, inevit-
able in the circumstances, of exchange control by the British
Government, which brought the free market, in gold and in foreign
exchange, to an end. So when more or less free trading resumed,
after the war, the markets had to find another centre, and they found
it hard to find one,
A word may perhaps be said, in concluding this historical digres-
sion, about the abortive British 'return to the Gold Standard* of
1925; it has more significance, from a modern perspective, than is
commonly supposed. Keynes taught his contemporaries to look at it
from an internally British point of view; it has since been generally
accepted that from that point of view at the gold value selected the
pound was overvalued. The General Strike of 1926, and the long
coal-miners' strike that was associated with it, are pointed to as evi-
dence. It can however be maintained on the other side that trouble in
coal-mining was inevitable, whatever the rate of exchange that had
been fixed; it was due to the resumption of German exports, in the ab-
sence of which British coal-miners had been able to establish a level
of wages which was now unsustainable. Apart from that special
trouble, the return of 1925 did not, for Ive years or so, do so badly.
Progress was made towards at least a partial resumption of sterling
centrality. Could even so much have been done if a lower parity had
been selected, as Keynes advocated? The prestige of the old parity
must have facilitated resumption, on the international front,
Nevertheless the episode is worth recalling, since it marked a sub-
sequently influential emergence of the clash between monetary poli-
cies, directed on the one hand to internal and on the other to
external stability, a clash that is still with us today.
I do not have the empirical knowledge to continue, even in this
vein, with the later story, so must proceed to sum up. The most
appropriate way of doing so may be to examine what precepts (to suc-
ceed the Thornton Gold Standard precepts: see Chapter 11) appear to
follow from what I have been saying. But even that is no simple
matter.
For it cannot now be expected that the rules will be the same for all
sorts of countries, fitting for any situation to which a particular
I^O PROBLEMS AND POLICIES

country may find itself. We must at the least distinguish between the
central and non-central countries; but that is not enough. We also
need to distinguish between those which have strong and those that
have weak currencies—using strong and weak In senses which cor-
respond to strength and weakness in the case of banks. Thus a
country would have a strong currency if there was confidence in the
unlikelihood of occasions arising when crisis measures would have
to be taken to defend it. It was balancing, and expected to go on
balancing, its external payments: or it might have a regular balance
in its favour, giving it surplus funds it could invest abroad. If for a
while it had an unfavourable balance, it had reserves on which it
could draw, or it could borrow on its excellent credit. A weak cur-
rency would continually need to be supported, and supporting it
would always be a problem.
Centrality and strength do not necessarily go together. Centrality
is not acquired by a decision of the 'central' government, or of its
banking system; it comes from decisions by others, who choose to
make the currency of that country their chief 'international'. No
doubt it is unlikely that such a choice would have been made unless
the central currency, at the time it became central, had been a strong
currency; but it could continue to be central, having no obvious
rival, even when it was losing its strength. So there are several cases
which fall to be considered, the chief division being between those
where the strength of the central currency in unquestioned, and
those where it is in doubt.
In the former cases, while the central currency is (by definition) a
strong currency, those of the others may be strong or may be weak.
It is tempting to say that the monetary problems of these non-central
currencies can then be treated separately, by expansionary policies
in the stronger, contractionary in the weaker; or if the internal con-
sequences (on employment or Inflation") are unwelcome, by an
upward or downward revaluation of their rates of exchange. Most of
the obstacles in the way of such measures are familiar, and need not,
be specified. It should however be emphasized that it is no solution,
not a way in which a country with a weak currency can easily get
out of its difficulty, just to float its exchange. For a floating rate will
just continue to fall unless an expectation can be aroused that from
some point in the fall it will recover. Thus it is safer to devalue, to a
rate which is planned to be held and which is intended to be
defended, than just to let go. There are many countries which have
 AN I N T E R N A T I O N A L ECONOMY 131
had, and have, bitter experiences in that direction. Neverthhheless itis
clear what should be done, though it may be hard to do it.
The problems of the central currency are more peculiar. So long as
it is of unquestioned strength (as the pound sterling was before 1914
and the US dollar was in the 1960s) its country can afford to base its
monetary policy on internal considerations, thinking of itself as if it
were a closed economy, and acting accordingly. Its actions, though
internally oriented, will indeed have external repercussions; in an
extreme case these may react back on the centre, as happened in
1930-3. A Thornton precept, as we have seen in Chapter 11, could
in such a case have stood up well. The 'devaluation* of the dollar in
1933, by the incoming administration of Franklin Roosevelt, can in
these terms be defended as necessary, in those still partially Gold
Standard days, as a preliminary to the expansionary policies that
were called for, both in the US and outside it.
The floating of the dollar in 1973 would appear, from this point of
view, to be a sad contrast. It is most readily to be interpreted as an
attempt by the American authorities to abdicate from the central res-
ponsibilities that had fallen upon them. But, as we have seen, it is not
in the power of a single government to disclaim such responsibilities.
This became clear after a year or two. The abdication was not
accepted. The dollar remained the central currency; it was accepted
as such throughout the trading world; but it no longer commanded
the former confidence. That meant, consequentially, that non-US
countries, seeking to stabilize the external values of their currencies,
had no standard by which they could judge what was stable. They
were themselves thrown back to working, in the first place, for in-
ternal stability; but that is itself hard to attain if some degree of ex-
ternal does not go with it.
It would appear that there followed a stage when the Americans
realized (under President Reagan) that the centrality of the dollar
should by them be accepted; and that therefore an adverse balance of
payments did not for them much matter. They were able to maintain
a fair stability of internal prices, and a high level of employment, con-
ceding that this left the rest of the world 'awash with dollars'. But
then, as this continued, the centrality of the dollar came under suspi-
cion. Can another centre be discovered? Or can means be invented by
which it would be possible to manage in a stable way without one? i
have no means of forming a judgement on such mighty questions.
 15 What is Bad about Inflation ?

This Is a question which at the end of this book, I think I should try to
answer. My answer is implied in what I have been saying, but it
needs to be set out explicitly.
Economists used to think that they knew the answer, but their old
answer will not do. Nevertheless it is convenient to begin with it. I
may take it in the form it was stated by Dennis Robertson, in a
passage I have often quoted.1
Our economic order is largely based upon the institution of contract... on
the fact, that is, that people enter into binding agreements with one another
to perform certain actions at a future date, for a remuneration which is fixed
here and now in terms of money. A violent or prolonged change in the value
of money saps the confidence with which people make or accept under-
takings of this nature.
One can see why Robertson, writing in the 1920s, thought that this
was the point to emphasize. He was thinking of an inflation that had
started up, after a state of affairs in which prices had been stable, or
fairly stable. Contracts had been made on expectation of stable prices;
those expectations were cheated by the inflation, There can be no
doubt that in those conditions his statement is correct. But his point
has less force when inflation has been continuous, so that people
have had time to adjust themselves to it. It will then appear that his
argument is not an argument for constant prices; it is an argument
for reliability. Once inflation has become established, it is indeed an
argument against acceleration of inflation. But cannot it then be
stood on its head, and used as an argument against deceleration? To
impose a condition of non-inflation, upon an economy which has
become adjusted to rising prices, would surely, from this point of
view, be quite as much of an upset.
I think that the Robertson argument, in this inverted form, does
have great weight with contemporary economists. It becomes a doc-
trine that so long as inflation is 'expected' it does not matter. This is
made particularly appealing by the habit many economists have got
into of thinking in terms of 'steady state' or 'growth equilibrium'
1
D. H. Robertson, Money (Cambridge Economic Handbooks, 1928 edition, p, 1}).
 WHAT IS BAD ABOUT INFLATION? 133

models, in which what happens is what was expected—not that it is

claimed that such models are realistic, only that they are manage-
able. Such a model would behave in real terms in just the same way
whether the level of money prices was constant or rising quite
rapidly. Money rates of interest, being money prices, would have to
adjust; but isn't this what happens? The behaviour of the financial
markets is indeed captured, on the approach in question, better than
that of the rest of the economy,1
It was because I wanted to get clear about this, and for my reader
to get clear about it, that I decided to start this book with those
chapters on the working of markets. They distinguished the ways in
which prices are formed on speculative markets—the financial
markets are in this sense speculative markets—from the way in
which they are increasingly formed on other markets, such as the
markets for manufactured goods and the market for labour. On these
latter markets, which are surely most important markets, prices have
to be 'made" or negotiated; they are not just 'determined' by demand
and supply. It is easier to make them, in a way which is acceptable to
the parties concerned, because it seems fair, if substantial use can be
made of precedent, if one can start with the supposition that what
was acceptable before will be acceptable again. When prices in
general are fairly stable, that may often be rather easy. The particu-
lar prices which result from such bargains may not be ideal from the
point of view of the economist; but the time and trouble which would
be involved in improving them is not worth while. To be obliged to
make them anew, and to go on making them anew, as one is obliged
to do in continuous inflation, involves direct economic loss, and very
often loss of temper as well.
Any system of prices (a system of railway fares, like a system of
wages) has to satisfy economic canons of efficiency and social canons
of fairness—canons which it is very difficult to make compatible. So
it is hard to re-negotiate it and it saves time not to do this very often.
But, that means, in an inflationary process, that the prices and wages
that are fixed are lagged. Suppose that a wage-system is negotiated
2
If one sticks to this approach but still believes that inflation is bad, one is bound to
lay stress upon the cost that is involved, when money rates of interest are high, in
holding a balance in the form of non-interest-bearing money; so business will try to
manage with less hoidinp of money even for transaction purposes. There is a toss of
convenience in this but it is questionable if it is a major matter, Moreover, in response
to the rise in nominal rates on their earning assets, banks will to some extent adjust
their deposit rates.
134 PROBLEMS AND POLICIES

in January, to be revised in the following January. A rate of price-

rise of 5 per cent per year is allowed for in fixing it, for that is what
has been experienced in the previous year. This compensates for
past inflation, but in the current year prices go on rising, while the
wage for the moment does not rise. During the year the wage-
earner feels himself to be losing ground. If an arrangement such as
this continues, he spends most of his time losing ground. Only at
the moment of re-fixing does he recover it. No wonder Inflation is
unpopular with him.
One can see in the light of this why indexation is no answer. It
may simplify negotiation at the annual re-fixing, but during the year
there will be just the same lag. A way out might perhaps be found by
the device which is practised in some countries, of paying only partly
in the form of a weekly wage, which is supplemented by a substantial
annual bonus. The bonus could be indexed, though the weekly wage
was not. But for such a system to be introduced, business must be
strong enough, at the point of its introduction, to pay a bonus on top
of what they were paying before.
Any general system of wage-indexation raises at its outset most
serious problems, ft is inevitable, at any particular stage in an infla-
tionary process, that there should be grave disparities in relative
wages, both social and, economic. Are these to be frozen? If not, how
are they to be mended? And, even if that first step can be successfully
taken, an indexed system will show up its fragility in the event of
new shocks, leading to shortages of particular kinds of labour, thus
upsetting established relativities.
Indexation just institutionalizes the wage-prices spiral, which is
basically the result of defending a level of real wages, or a rate of rise
in real wages (or of real incomes, in the broadest sense, which are
associated with it) which has become inconsistent with economic
realities. That is what happened in Britain in the 1970s, when as a
result of the oil shock, and the related rise in the prices of other im-
ports which went along with it, the rise in real wages to which
British workers had become accustomed was brought to a sudden
halt. That led to wage-inflation; it was the cause of the wage-
inflation; monetary policy was just permissive. It is true that it was
brought to an end by monetary restriction, which to deal with so
great an inflation, never before experienced in peace-time to Britain,
had to be savage. Even so, it was slow-acting; so the pain and grief
were felt in the first years of Mrs Thatcher's government though the
 WHAT IS BAD ABOUT INFLATION? 135

action which led to it had been begun under her predecessor, A

democratic Labour government is bound in the end to be anti-
inflation.
Was there any way in which there could have been a 'soft land-
ing'? It could only have been done if the real causes which had led to
the inflation could have been removed. It is hard to see that the state
of mind, the social causes which connect the movement, of wages
with the movement of prices, could have been changed without a
shock. Even the shock which has been given may not have been
enough. More hopeful is to work on the other link, from wages to
prices. There have been several examples in this century of post-war
inflations being brought under control by post-war recovery, re-
covery of productivity in real terms; as the capital stock of the
country recovered, prices would rise more slowly than wages, so the
spiral would ease off. As we have seen, the quickest effect of monet-
ary restriction, in a single country, is an improvement in the external
value of that country's currency; the 'over-valued' exchange does act
as a brake upon the rise in prices, but at the expense of damage to
exporting and import-competing industries, with which we are so
familiar. The long-run answer must be an improvement in internal
productivity, but the damage suffered by these industries makes this
harder to attain.
So perhaps what is bad about inflation is principally not its
effects—the losses of 'convenience and security' to which older
economists gave so much attention—but the weakening of the eco-
nomy, which is the cause of the evil. If that Is cured, inflation, with
only a little help from monetary policy, will cure itself.
Something should be said in conclusion about hyper-inflation, the
phenomenon of which the leading case was for long the great
German inflation of 1923, but of course there are current examples
in many parts of the world. I would distinguish this from the moder-
ate inflations of a few percentage points per year which we are learn-
ing to live with, not simply by the rate of price rise being so much
greater but by the fact that in hyper-inflation no prices can be estab-
lished, for there will be a rise in the price level before any transaction
can be completed, so that money is losing its capacity to act as a
means of payment. There is a complete or almost complete lack of
confidence in the credit of the government.
That confidence cannot be restored by a mere change in
denomination—the introduction of a new money in place of the old
136 PROBLEMS AND POLICIES

that is discredited. When that is done and nothing else is done, the
new money will just go the way of the old. A way must be found of
giving the new money a new credibility.
Here we must distinguish between internal and external credibil-
ity. There is one case when external credibility does not matter—
when the country can cut itself off from external economic relations,
in particular from foreign trade. (Or it may be able to keep some
foreign trade by recourse to barter deals, as previously noticed.*
These must almost inevitably be inter-governmental deals. They re-
quire that there should be a pair of countries in a similar monetary
position and that each is in a position to offer some of the goods
which the other most urgently requires. This is still a modified
'autarky'.)
If the autarky solution in either form is available, it is only internal
credibility that has to be restored. That can be done if there is a
visible change in government policy, nearly always implying a new
government, and a new government that is unlikely to be displaced:
A condition that is unlikely to be satisfied by a constitutional govern-
ment which has to submit itself periodically to re-election.
If that solution—in so far as it is a solution—is rejected, then both
internal and external credibility must be restored, and in practice the
external comes first, for if external credibility is restored, internal will
fairly easily follow. The ways of ensuring that it does are well known.
But internal credibility will always be undermined by absence of ex-
ternal. For external to be restored needs support from countries with
stronger currency—as was done in the classical German case under
the so-called Dawes Plan and Young Plan, without which the new
Mark could not have been held. Internal measures were also neces-
sary, but without that external help they could not have succeeded.
So in the modern cases stabilization must be associated in the first
case with a clearing up of the foreign debts of the countries affected.
And that can hardly be done without putting themselves, for a time,
under the financial control of their creditors.

1
Above, p. 43.
 APPENDIX:
RISK AND U N C E R T A I N T Y

The title of this appendix is meant to echo that of the famous book by
Frank Knight, Risk, Uncertainty and Profit (1920), on which I shall
have something important to say before I have concluded. It is some-
thing which might not have been uncongenial to him. But I begin
with a piece of very formal theory, which would not have appealed to
him at all.
It Is an exercise in what is called the theory of portfolio selection. I
introduce it here, in spite of the very un-Knightian assumptions on
which it is based—assumptions I do not much care for myself—
because it brings out a point, which in the end does not seem to
depend upon them, and which should be quite a help towards under-
standing what I am saying in this book. So I want to emphasize that
my acceptance of these assumptions is only provisional.
We are to consider the behaviour of an operator, who is disposing
of a capital of given money value K; he is confronted with opportun-
ities of investing it in some combination of n securities, the current
prices of which are given to him, being independent of his own be-
haviour. There are no transactions costs, so he is not hampered by
inheritance from the past; we can think of him as having got the
whole of his capital into money form before he decides to invest it. So
the whole of his portfolio, after he has made his decision, is that
which he has chosen on the basis of the current opportunities open
to him.
We think of him having to hold it until a 'week' has elapsed (as in
Chapter 2), so it is the outcome of his decision, at the end of the week,
which he wants to make as favourable as possible to him. But he does
not know what the outcome will be. There are possibilities of differ-
ent outcomes; he has to make his choice with imperfect knowledge of
them.
The conventional way of dealing with this problem is to suppose
that there are m eventualities, or 'states of the world', any of which
may occur. The operator knows what will be the outcome, in each
eventuality, of the investment of one unit of money in each security,
but he does not know which eventuality will occur. (This just
138 RISK AND U N C E R T A I N T Y

amounts to sweeping the knowns and unknowns into separate

boxes.)
Thus if a is this known outcome, in the ith eventuality, of a unit
of money invested in the jth security, the total outcome of amounts
(x. . , ., xa) to be put into the n securities (among which a non-
interest-bearing money may or may not be included) will be

So, corresponding to each vector x of investments in securities, there

is a vector y of outcomes in eventualities; but how is the chooser to
order them so as to distinguish which he prefers? No advantage is
gained from this way of posing the problem unless we give him some
criterion by which to do so,
In order to make the outcomes y|, comparable with one another, a
means of weighting them is needed: so the next thing to do is to sup-
pose him to attach probabilities to the outcomes, or to the eventu-
alities corresponding. This is what Knight would not allow us to do,
on the ground (with which I agree) that the eventualities, important
for a portfolio choice such as this, are usually not classifiable; they
are not like the results of scientific experiments, where in the long
run there will often be a convergence to definite proportions of failure
and success.1 For where, in the field that is here under consideration,
shall we find the long-run repetition that is needed? Let us however
for the moment forget about that, and suppose that there are cardinal
probabilities which our operator attaches to his eventualities (p, with

The simplest scheme of maximization which then presents itself is

to suppose that what is maximized is Piy,-, the 'mathematical ex-
pectation' of outcome. But that will not do. For

and, since x; = K, the operator would just put the whole of his
capital into that security for which £ pt ait (the ps and as being in-
dependent of his choice) was the largest. Thus, as has been long
understood, this solution must be rejected, since it gives no oppor-
tunity for spreading of risks.
It was already suggested by the great mathematician Bernoulli, at
the beginning of the eighteenth century, that what the chooser
should be supposed to maximize is the 'mathematical expectation' of
1
My present view on this matter is set out at length in the chapter entitled "Prob-
ability and judgement' which is appended to my Causality in Economics (1979).
 RISK AND U N C E R T A I N T Y 139

the utility which he is to derive from his choice. He was writing

before there were any utilitarians; but the utility function u(y) which
he required had just the properties which have become conven-
tional: that u'(y), the marginal utility (MU), would be positive but
would diminish as y increased.
If it is U = Xft "(ill) which is to be maximized, both the ps and the
us must be cardinal numbers. There has been much discussion of
cardinal utility, but I do not think that in this place it is relevant. If
we accept cardinality for probability, why not for utility? If we do not
accept it for probability, the utility question does not arise.
All I shall do with the Bernoulli construction is to see what answer
it gives to an interesting question: is there any utility function u(y)
which is consistent with a change in K leaving the proportions in
which the securities are taken unchanged? It has sometimes been
denied that there is any such function; but I think it can be shown
that it does exist and is quite instructive.
If all xj were increased in the same proportion, then all y, would be
increased in the same proportion (assuming of course that the a
were unchanged). But if the operator were in a preferred position at
the old yi, there must have been a particular relation between their
marginal utilities in the old position; it can be shown that this rela-
tion will be unchanged when the yt increase equi-proportionally, if
the marginal utilities u'(yt) also change in equal proportions.2 Thus,
for a small change in K, the elasticities of the MU curves u'(yt) must be
the same. But these are not different curves; the different ij, and their
corresponding u'(yi) are different points on the same curve. So what
is being said is that the (single) MU curve must have constant elasti-
city.
Now it is well known that for any downward-sloping curve to
have constant elasticity, its equation must be of a particular form.
We must have u'(y) = A y~"ht where A is a constant (positive to keep
MU positive) and h is the reciprocal of the MarshaUian elasticity. Now
the total utility corresponding to this is got by integrating with
respect to y; in general, this gives

constant

' For let U, be the expected marginal utility of investment in tfaejth security. Since
each of the coefficients in K = ]£ ar, is unity, it will be necessary that in a preferred posi-
tion, all Ut should be equal. But if V = £ piU ( y ) < Ui = Eft u'to «./. since the ps and the
as are all constants. Thus if the u'tjj,) change to equal proportions, the U, will also
change in equal proportions. If they were equal before, they will still be equal.
I4O RISK AND U N C E R T A I N T Y

Now the variable part of this will only be positive, for positive y, if
h < 1; that is to say the MU curve, in Marshall's sense, must be elas-
tic, If it is inelastic (h > 1), the variable part will be negative; so u(y)
can only be positive if the constant is large and positive; call this B,
There are thus two cases, according as h is < 1 or ^ 1 . In the
former, the variable part of u(y) will be zero when y = 0, so we may
take that as a base from which to measure; as y increases, through
positive values, u(y) will continue to increase, It will increase very
rapidly when y is small, thereafter at a diminishing rate. Though this
must be given a place in the mathematics, * it does not seem to be of
much economic interest. It is otherwise with the latter case, where it
is proper to write

This I shall claim to be more interesting. Here, as y increases indefin-

itely, the variable part approaches zero, so that B is a limit beyond
which u(y) cannot increase.* But if y = 0, u(y) is minus infinity. This
is surprising (in consumer theory it would be nonsense) but I think it
makes sense, in the present context.
There is no reason why we should exclude the possibilty of an out-
come having a utility of minus infinity; for that would just mean that
the operator would always avoid such an outcome, if it were possible
to avoid it. Now if the securities that are on offer are of the usual
kinds (bills, bonds and equities carrying limited liability) the worst
that can. happen from any choice of such securities is total loss of the
whole portfolio; that is to say, y = 0. To put u(0) equal to minus in-
finity amounts to saying that a chance of such loss would always be
avoided. If there was a particular security for which there was a finite
chance of such total loss, to put the whole capital into that security is
a course that would always be avoided. To put a part of one's capital
into such a security is not a choice that is excluded; for so long as the
rest is out into securities where there is no such possibility, there is
no chance of total loss over the whole portfolio. It is 'plunging' with

' In earlier versions of this appendix (as on pp. 2 51-6 of the second volume of my
Collected Essays'} it was neglected. I have however been persuaded by Stefano Zamagni
that I must give it at least this much attention,
* So B is the 'bliss* of the famous article by Frank Ramsey on "Optimum saving'
(Economic Journal, 1928) which made such an impression on Keynes.
 RISK AND U N C E R T A I N T Y 141

all one's resources which is ruled out. The minus infinity of the utility
function just means that this obviously foolish behaviour is excluded.
It has thus been shown that there is a quite intelligible utility func-
tion which (on Bernoutlian principles) is consistent with the distribu-
tion of the portfolio, over all securities, remaining unchanged when
the capital to be disposed of changes. But, having got so far, we can
rather easily get further, indeed much further. We can Indeed largely
dispense with the help we have got from Bernoulli.
To begin with, why should the disaster point, which is always to
be avoided, be set at this purely arithmetical total loss? It only looks
even apparently plausible to set it at that point, because we have
been following the fashion of looking at the money-securities part of
the typical balance-sheet in isolation, without regard to other items,
real goods on the assets side, and liabilities. If the operator has liabil-
ities as well as assets, or if he is carrying on a productive business
which itself creates calls upon him, he may well face disaster, when
the outcome of his investments in securities has fallen very low, even
if it is not zero. His utility function should then go to minus infinity at
a positive value of y, which we will call c.
We can deal with this amendment rather easily. We have only to
put u'(y) = A(y — c)~h, so that the whole MU curve is shifted to the
right, becoming asymptotic to y = c. The effect of this is very simple.
We still have
and

so that

Thus maximization is just the same as in the former case, save that all
unit outcomes—of particular securities in particular eventualities—
are written down by c/K. If K is large compared with c, the write-
down is negligible; so the chosen portfolio will be practically the
same as with the former function. But as K diminishes, relatively to c.
the write-down will take effect. Some of the % — (c/K) will then start
to go negative, so that the operator will avoid the securities in ques-
tion, those which have a very low outcome in some eventualities. He
will avoid such risky securities,
1 am at last in a position to go back to Knight, His major distinc-
tion, between measureable risks, based on cardinal probabilities (for
which there is evidence) and what he calls true uncertainties, which
142. RISK AND U N C E R T A I N T Y

are not so based, 1 fully accept. Indeed I would now attach much
more importance to it than I did in my first contribution to the sub-
ject,5 written under his influence not much more than ten years after
his book appeared. That I hope will have been made clear in what
precedes.
The chief criticism I would now make of him is that his termino-
logy, which has greatly influenced many subsequent writers, is
rather confusing. Our disaster point should help to get it straight.
For it suggests that what is needed is a four-way, not a two-way,
classification. First, there should be a distinction between choices
that involve a risk of some sort of disaster—these being just what the
plain man would call risky choices—and those which do not. This
should be crossed with Knight's distinction between those where the
chances are measurable, and those where they are not.
That would give us (1) measurable risky choices, which are those
it may be possible to mitigate by some form of insurance; (2) non-
measurable risky choices, which probably match the 'true uncertain-
ties' of Knight; (3) measurable non-risky choices, like buying a ticket
for a lottery, where the loss involved in not getting a prize is easily
bearable; and (4) non-measurable non-risky choices, such as one
might think to be involved in the ordinary running of a business.
That would seem to make sense. But what it has to do with the 'justi-
fication' of profit is another question. I would look for that in a differ-
ent direction.
If this arrangement is accepted, we do not have to worry whether
our actors are 'risk-averse' or not. Every business man must be risk-
averse if he is planning to go on with his business. Even the gambler
must be risk-averse if he plans to go on with his game. If he has
ceased to be risk-averse he has just gone crazy. Risk-aversion is a
consequence of rational behaviour, as we have found it to be in the
case of banks—and so on.

' 'Uncertainty and profit' (Ecanomica. 1931); reprinted, slightly abridged, in the
second volume of my Collected Essays.