The General Theory of Employment Interest and Money by John Maynard Keynes

The General Theory of Employment, Interest,and Money

John Maynard Keynes

1935

CONTENTS

Contents i

Preface v

Preface to the German edition ix

Preface to the Japanese edition xiii

Preface to the French edition xv

I Introduction 1

1 The General Theory 3

2 The Postulates of the Classical Economics 5

3 The Principle of Effective Demand 23

II Definitions and Ideas 35

4 The Choice of Units 37

5 Expectation as Determining Output and Employ-ment 45

6 The Definition of Income, Saving and Investment 51

i

ii CONTENTS

7 The Meaning of Saving and Investment Further Con-sidered 73

III The Propensity to Consume 85

8 The Propensity to Consume: I. The Objective Factors 87

9 The Propensity to Consume: II. The Subjective Fac-tors 105

10 The Marginal Propensity to Consume and the Mul-tiplier 111

IV The Inducement to Invest 129

11 The Marginal Efficiency of Capital 131

12 The State of Long-Term Expectation 143

13 The General Theory of the Rate of Interest 159

14 The Classical Theory of the Rate of Interest 169

15 The Psychological and Business Incentives To Liq-uidity 189

16 Sundry Observations on the Nature of Capital 205

17 The Essential Properties of Interest and Money 217

18 The General Theory of Employment Re-Stated 239

V Money-Wages and Prices 249

19 Changes in Money-Wages 251

CONTENTS

20 The Employment Function 275

21 The Theory of Prices 287

VI Short Notes Suggested by the General Theory 305

22 Notes on the Trade Cycle 307

23 Notes on Mercantilism, The Usury Laws, StampedMoney and Theories of Under-Consumption 327

24 Concluding Notes on the Social Philosophy towardswhich the General Theory might Lead 369

iii

PREFACE

This book is chiefly addressed to my fellow economists. Ihope that it will be intelligible to others. But its main purposeis to deal with difficult questions of theory, and only in thesecond place with the applications of this theory to practice.For if orthodox economics is at fault, the error is to be foundnot in the superstructure, which has been erected with greatcare for logical consistency, but in a lack of clearness and ofgenerality in the pre misses. Thus I cannot achieve my objectof persuading economists to re-examine critically certain oftheir basic assumptions except by a highly abstract argumentand also by much controversy. I wish there could have beenless of the latter. But I have thought it important, not only toexplain my own point of view, but also to show in what re-spects it departs from the prevailing theory. Those, who arestrongly wedded to what I shall call ‘the classical theory’, willfluctuate, I expect, between a belief that I am quite wrong anda belief that I am saying nothing new. It is for others to de-termine if either of these or the third alternative is right. Mycontroversial passages are aimed at providing some materialfor an answer; and I must ask forgiveness If, in the pursuit ofsharp distinctions, my controversy is itself too keen. I myselfheld with conviction for many years the theories which I nowattack, and I am not, I think, ignorant of their strong points.

The matters at issue are of an importance which cannot beexaggerated. But, if my explanations are right, it is my fel-low economists, not the general public, whom I must firstconvince. At this stage of the argument the general public,though welcome at the debate, are only eavesdroppers at anattempt by an economist to bring to an issue the deep diver-gences of opinion between fellow economists which have for

v

PREFACE

the time being almost destroyed the practical influence of eco-nomic theory, and will, until they are resolved, continue to doso.

The relation between this book and my Treatise on Money[JMK vols. v and vi], which I published five years ago, isprobably clearer to myself than it will be to others; and whatin my own mind is a natural evolution in a line of thoughtwhich I have been pursuing for several years, may sometimesstrike the reader as a confusing change of view. This difficultyis not made less by certain changes in terminology which Ihave felt compelled to make. These changes of language Ihave pointed out in the course of the following pages; butthe general relationship between the two books can be ex-pressed briefly as follows. When I began to write my Treatiseon Money I was still moving along the traditional lines of re-garding the influence of money as something so to speak sep-arate from the general theory of supply and demand. When Ifinished it, I had made some progress towards pushing mon-etary theory back to becoming a theory of output as a whole.But my lack of emancipation from preconceived ideas showeditself in what now seems to me to be the outstanding faultof the theoretical parts of that work (namely, Books III andIV), that I failed to deal thoroughly with the effects of changesin the level of output. My so-called ’fundamental equationswere an instantaneous picture taken on the assumption of agiven output. They attempted to show how, assuming thegiven output, forces could develop which involved a profit-disequilibrium, and thus required a change in the level of out-put. But the dynamic development, as distinct from the in-stantaneous picture, was left incomplete and extremely con-fused. This book, on the other hand, has evolved into whatis primarily a study of the forces which determine changes inthe scale of output and employment as a whole; and, whilstit is found that money enters into the economic scheme in anessential and peculiar manner, technical monetary detail fallsinto the background. A monetary economy, we shall find, isessentially one in which changing views about the future arecapable of influencing the quantity of employment and not

vi

merely its direction. But our method of analysing the eco-nomic behaviour of the present under the influence of chang-ing ideas about the future is one which depends on the inter-action of supply and demand, and is in this way linked upwith our fundamental theory of value. We are thus led to amore general theory, which includes the classical theory withwhich we are familiar, as a special case.

The writer of a book such as this, treading along unfamiliarpaths, is extremely dependent on criticism and conversationif he is to avoid an undue proportion of mistakes. It is aston-ishing what foolish things one can temporarily believe if onethinks too long alone, particularly in economics (along withthe other moral sciences), where it is often impossible to bringone’s ideas to a conclusive test either formal or experimental.In this book, even more perhaps than in writing my Treatiseon Money, I have depended on the constant advice and con-structive criticism of Mr R.F. Kahn. There is a great deal inthis book which would not have taken the shape it has ex-cept at his suggestion. I have also had much help from MrsJoan Robinson, Mr R.G. Hawtrey and Mr R.F. Harrod, whohave read the whole of the proof-sheets. The index has beencompiled by Mr D. M. Bensusan-Butt of King’s College, Cam-bridge.

The composition of this book has been for the author along struggle of escape, and so must the reading of it be formost readers if the author’s assault upon them is to be suc-cessful,—a struggle of escape from habitual modes of thoughtand expression. The ideas which are here expressed so labo-riously are extremely simple and should be obvious. The dif-ficulty lies, not in the new ideas, but in escaping from the oldones, which ramify, for those brought up as most of us havebeen, into every corner of our minds.

J. M. KEYNES13 December 1935

vii

PREFACE TO THE GERMAN EDITION

Alfred Marshall, on whose Principles of Economics all contem-porary English economists have been brought up, was at par-ticular pains to emphasise the continuity of his thought withRicardo’s. His work largely consisted in grafting the marginalprinciple and the principle of substitution on to the Ricar-dian tradition; and his theory of output and consumption as awhole, as distinct from his theory of the production and dis-tribution of a given output, was never separately expounded.Whether he himself felt the need of such a theory, I am notsure. But his immediate successors and followers have cer-tainly dispensed with it and have not, apparently, felt the lackof it. It was in this atmosphere that I was brought up. I taughtthese doctrines myself and it is only within the last decadethat I have been conscious of their insufficiency. In my ownthought and development, therefore, this book represents areaction, a transition away from the English classical (or or-thodox) tradition. My emphasis upon this in the followingpages and upon the points of my divergence from receiveddoctrine has been regarded in some quarters in England asunduly controversial. But how can one brought up a Catholicin English economics, indeed a priest of that faith, avoid somecontroversial emphasis, when he first becomes a Protestant?

But I fancy that all this may impress German readers some-what differently. The orthodox tradition, which ruled in nine-teenth century England, never took so firm a hold of Ger-man thought. There have always existed important schoolsof economists in Germany who have strongly disputed theadequacy of the classical theory for the analysis of contempo-rary events. The Manchester School and Marxism both deriveultimately from Ricardo,—a conclusion which is only superfi-

ix

PREFACE TO THE GERMAN EDITION

cially surprising. But in Germany there has always existed alarge section of opinion which has adhered neither to the onenor to the other.

It can scarcely be claimed, however, that this school of thoughthas erected a rival theoretical construction; or has even at-tempted to do so. It has been sceptical, realistic, content withhistorical and empirical methods and results, which discardformal analysis. The most important unorthodox discussionon theoretical lines was that of Wicksell. His books were avail-able in German (as they were not, until lately, in English); in-deed one of the most important of them was written in Ger-man. But his followers were chiefly Swedes and Austrians,the latter of.whom combined his ideas with specifically Aus-trian theory so as to bring them in effect, back again towardsthe classical tradition. Thus Germany, quite contrary to herhabit in most of the sciences, has been content for a wholecentury to do without any formal theory of economics whichwas predominant and generally accepted.

Perhaps, therefore, I may expect less resistance from Ger-man, than from English, readers in offering a theory of em-ployment and output as a whole, which departs in importantrespects from the orthodox tradition. But can I hope to over-come Germany’s economic agnosticism? Can I persuade Ger-man economists that methods of formal analysis have some-thing important to contribute to the interpretation of contem-porary events and to the moulding of contemporary policy?After all, it is German to like a theory. How hungry and thirstyGerman economists must feel after having lived all these yearswithout one! Certainly, it is worth while for me to make theattempt. And if I can contribute some stray morsels towardsthe preparation by German economists of a full repast of the-ory designed to meet specifically German conditions, I shallbe content. For I confess that much of the following book isillustrated and expounded mainly with reference to the con-ditions existing in the Anglo-Saxon countries.

Nevertheless the theory of output as a whole, which is whatthe following book purports to provide, is much more easilyadapted to the conditions of a totalitarian state, than is the

x

theory of the production and distribution of a given outputproduced under conditions of free competition and a largemeasure of laissez-faire. The theory of the psychological lawsrelating consumption and saving, the influence of loan expen-diture on prices and real wages, the part played by the rate ofinterest—these remain as necessary ingredients in our schemeof thought.

I take this opportunity to acknowledge my indebtednessto the excellent work of my translator Herr Waeger (I hopehis vocabulary at the end of this volume may prove usefulbeyond its immediate purpose) and to my publishers, MessrsDuncker and Humblot, whose enterprise, from the days nowsixteen years ago when they published my Economic Conse-quences of the Peace, has enabled me to maintain contact withGerman readers.

J. M. KEYNES7 September 1936

xi

PREFACE TO THE JAPANESE EDITION

Alfred Marshall, on whose Principles of Economics all contem-porary English economists have been brought up, was at par-ticular pains to emphasise the continuity of his thought withRicardo’s. His work largely consisted in grafting the marginalprinciple and the principle of substitution on to the Ricar-dian tradition; and his theory of output and consumption as awhole, as distinct from his theory of the production and dis-tribution of a given output, was never separately expounded.Whether he himself felt the need of such a theory, I am notsure. But his immediate successors and followers have cer-tainly dispensed with it and have not, apparently, felt the lackof it. It was in this atmosphere that I was brought up. I taughtthese doctrines myself and it is only within the last decadethat I have been conscious of their insufficiency. In my ownthought and development, therefore, this book represents areaction, a transition away from the English classical (or or-thodox) tradition. My emphasis upon this in the followingpages and upon the points of my divergence from receiveddoctrine has been regarded in some quarters in England asunduly controversial. But how can one brought up in Englisheconomic orthodoxy, indeed a priest of that faith at one time,avoid some controversial emphasis, when he first becomes aProtestant?

Perhaps Japanese readers, however, will neither require norresist my assaults against the English tradition. We are wellaware of the large scale on which English economic writingsare read in Japan, but we are not so well informed as to howJapanese opinions regard them. The recent praiseworthy en-terprise on the part of the International Economic Circle ofTokyo in reprinting Malthus’s ‘Principles of Political Econ-

xiii

PREFACE TO THE JAPANESE EDITION

omy’ as the first volume in the Tokyo Series of Reprints en-courages me to think that a book which traces its descent fromMalthus rather than Ricardo may be received with sympathyin some quarters at least.

At any rate I am grateful to the Oriental Economist formaking it possible for me to approach Japanese readers with-out the extra handicap of a foreign language.

J. M. KEYNES4 December 1936

xiv

PREFACE TO THE FRENCH EDITION

For a hundred years or longer, English Political Economy hasbeen dominated by an orthodoxy. That is not to say that anunchanging doctrine has prevailed. On the contrary. Therehas been a progressive evolution of the doctrine. But its pre-suppositions, its atmosphere, its method have remained sur-prisingly the same, and a remarkable continuity has been ob-servable through all the changes. In that orthodoxy, in thatcontinuous transition, I was brought up. I learnt it, I taught it,I wrote it. To those looking from outside I probably still be-long to it. Subsequent historians of doctrine will regard thisbook as in essentially the same tradition. But I myself in writ-ing it, and in other recent work which has led up to it, havefelt myself to be breaking away from this orthodoxy, to be instrong reaction against it, to be escaping from something, to begaining an emancipation. And this state of mind on my partis the explanation of certain faults in the book, in particular itscontroversial note in some passages, and its air of being ad-dressed too much to the holders of a particular point of viewand too little ad urbem et orbem. I was wanting to convincemy own environment and did not address myself with suf-ficient directness to outside opinion. Now three years later,having grown accustomed to my new skin and having almostforgotten the smell of my old one, I should, if I were writingafresh, endeavour to free myself from this fault and state myown position in a more clear-cut manner.

I say all this, partly to explain and partly to excuse, myselfto French readers. For in France there has been no orthodoxtradition with the same authority over contemporary opinionas in my own country. In the United States the position hasbeen much the same as in England. But in France, as in the rest

xv


of Europe, there has been no such dominant school since theexpiry of the school of French Liberal economists who were intheir prime twenty years ago (though they lived to so great anage, long after their influence had passed away, that it fell tomy duty, when I first became a youthful editor of the EconomicJournal to write the obituaries of many of them—Levasseur,Molinari, Leroy-Beaulieu). If Charles Gide had attained to thesame influence and authority as Alfred Marshall, your posi-tion would have borne more resemblance to ours. As it is,your economists are eclectic, too much (we sometimes think)without deep roots in systematic thought. Perhaps this maymake them more easily accessible to what I have to say. But itmay also have the result that my readers will sometimes won-der what I am talking about when I speak, with what someof my English critics consider a misuse of language, of the‘classical’ school of thought and ’classical’ economists. It may,therefore, be helpful to my French readers if I attempt to in-dicate very briefly what I regard as the main differentiae of myapproach.

I have called my theory a general theory. I mean by thisthat I am chiefly concerned with the behaviour of the eco-nomic system as a whole,—with aggregate incomes, aggre-gate profits, aggregate output, aggregate employment, aggre-gate investment, aggregate saving rather than with the in-comes, profits, output, employment, investment and savingof particular industries, firms or individuals. And I argue thatimportant mistakes have been made through extending to thesystem as a whole conclusions which have been correctly ar-rived at in respect of a part of it taken in isolation.

Let me give examples of what I mean. My contention thatfor the system as a whole the amount of income which issaved, in the sense that it is not spent on current consump-tion, is and must necessarily be exactly equal to the amountof net new investment has been considered a paradox and hasbeen the occasion of widespread controversy. The explana-tion of this is undoubtedly to be found in the fact that this re-lationship of equality between saving and investment, whichnecessarily holds good for the system as a whole, does not

xvi

hold good at all for a particular individual. There is no reasonwhatever why the new investment for which I am responsibleshould bear any relation whatever to the amount of my ownsavings. Qute legitimately we regard an individual’s incomeas independent of what he himself consumes and invests. Butthis, I have to point out, should not have led us to overlookthe fact that the demand arising out of the consumption andinvestment of one individual is the source of the incomes ofother individuals, so that incomes in general are not indepen-dent, quite the contrary, of the disposition of individuals tospend and invest; and since in turn the readiness of individ-uals to spend and invest depends on their incomes, a rela-tionship is set up between aggregate savings and aggregateinvestment which can be very easily shown, beyond any pos-sibility of reasonable dispute, to be one of exact and necessaryequality. Rightly regarded this is a banale conclusion. But itsets in motion a train of thought from which more substantialmatters follow. It is shown that, generally speaking, the actuallevel of output and employment depends, not on the capacityto produce or on the pre-existing level of incomes, but on thecurrent decisions to produce which depend in turn on cur-rent decisions to invest and on present expectations of currentand prospective consumption. Moreover, as soon as we knowthe propensity to consume and to save (as I call it), that is tosay the result for the community as a whole of the individualpsychological inclinations as to how to dispose of given in-comes, we can calculate what level of incomes, and thereforewhat level of output and employment, is in profit-equilibriumwith a given level of new investment; out of which developsthe doctrine of the Multiplier. Or again, it becomes evidentthat an increased propensity to save will ceteris paribus con-tract incomes and output; whilst an increased inducement toinvest will expand them. We are thus able to analyse the fac-tors which determine the income and output of the systemas a whole;—we have, in the most exact sense, a theory ofemployment. Conclusions emerge from this reasoning whichare particularly relevant to the problems of public finance andpublic policy generally and of the trade cycle.

xvii


Another feature, specially characteristic of this book, is thetheory of the rate of interest. In recent times it has been heldby many economists that the rate of current saving determinedthe supply of free capital, that the rate of current investmentgoverned the demand for it, and that the rate of interest was,so to speak, the equilibrating price-factor determined by thepoint of intersection of the supply curve of savings and the de-mand curve of investment. But if aggregate saving is necessar-ily and in all circumstances exactly equal to aggregate invest-ment, it is evident that this explanation collapses. We have tosearch elsewhere for the solution. I find it in the idea that it isthe function of the rate of interest to preserve equilibrium, notbetween the demand and the supply of new capital goods, butbetween the demand and the supply of money, that is to saybetween the demand for liquidity and the means of satisfyingthis demand. I am here returning to the doctrine of the older,pre-nineteenth century economists. Montesquieu, for exam-ple, saw this truth with considerable clarity,—Montesquieuwho was the real French equivalent of Adam Smith, the great-est of your economists, head and shoulders above the phys-iocrats in penetration, clear-headedness and good sense (whichare the qualities an economist should have). But I must leaveit to the text of this book to show how in detail all this worksout.

I have called this book the General Theory of Employment,Interest and Money; and the third feature to which I may callattention is the treatment of money and prices. The follow-ing analysis registers my final escape from the confusions ofthe Quantity Theory, which once entangled me. I regard theprice level as a whole as being determined in precisely thesame way as individual prices; that is to say, under the influ-ence of supply and demand. Technical conditions, the levelof wages, the extent of unused capacity of plant and labour,and the state of markets and competition determine the sup-ply conditions of individual products and of products as awhole. The decisions of entrepreneurs, which provide the in-comes of individual producers and the decisions of those in-dividuals as to the disposition of such incomes determine the

xviii

demand conditions. And prices—both individual prices andthe price-level—emerge as the resultant of these two factors.Money, and the quantity of money, are not direct influencesat this stage of the proceedings. They have done their workat an earlier stage of the analysis. The quantity of money de-termines the supply of liquid resources, and hence the rateof interest, and in conjunction with other factors (particularlythat of confidence) the inducement to invest, which in turnfixes the equilibrium level of incomes, output and employ-ment and (at each stage in conjunction with other factors) theprice-level as a whole through the influences of supply anddemand thus established.

I believe that economics everywhere up to recent times hasbeen dominated, much more than has been understood, bythe doctrines associated with the name of J.-B. Say. It is truethat his ’law of markets’ has been long abandoned by mosteconomists; but they have not extricated themselves from hisbasic assumptions and particularly from his fallacy that de-mand is created by supply. Say was implicitly assuming thatthe economic system was always operating up to its full ca-pacity, so that a new activity was always in substitution for,and never in addition to, some other activity. Nearly all sub-sequent economic theory has depended on, in the sense thatit has required, this same assumption. Yet a theory so based isclearly incompetent to tackle the problems of unemploymentand of the trade cycle. Perhaps I can best express to Frenchreaders what I claim for this book by saying that in the the-ory of production it is a final break-away from the doctrinesof J.-B. Say and that in the theory of interest it is a return to thedoctrines of Montesquieu.

J. M. KEYNES20 February 1939King’s CollegeCambridge

xix

Book I

Introduction

CHAPTER 1

THE GENERAL THEORY

I have called this book the General Theory of Employment, In-terest and Money, placing the emphasis on the prefix general.The object of such a title is to contrast the character of my ar-guments and conclusions with those of the classical1 theory ofthe subject, upon which I was brought up and which dom-inates the economic thought, both practical and theoretical,of the governing and academic classes of this generation, asit has for a hundred years past. I shall argue that the postu-lates of the classical theory are applicable to a special case onlyand not to the general case, the situation which it assumes be-ing a limiting point of the possible positions of equilibrium.Moreover, the characteristics of the special case assumed bythe classical theory happen not to be those of the economicsociety in which we actually live, with the result that its teach-ing is misleading and disastrous if we attempt to apply it tothe facts of experience.

1“The classical economists” was a name invented by Marx to cover Ricardoand James Mill and their predecessors, that is to say for the founders of the the-ory which culminated in the Ricardian economics. I have become accustomed,perhaps perpetrating a solecism, to include in “the classical school” the followersof Ricardo, those, that is to say, who adopted and perfected the theory of the Ri-cardian economics, including (for example) J. S. Mill, Marshall, Edgeworth andProf. Pigou.

3

CHAPTER 2

THE POSTULATES OF THE CLASSICALECONOMICS

I

Most treatises on the theory of value and production are pri-marily concerned with the distribution of a given volume ofemployed resources between different uses and with the con-ditions which, assuming the employment of this quantity ofresources, determine their relative rewards and the relativevalues of their products.1

The question, also, of the volume of the available resources,in the sense of the size of the employable population, the ex-tent of natural wealth and the accumulated capital equipment,has often been treated descriptively. But the pure theory ofwhat determines the actual employment of the available resourceshas seldom been examined in great detail. To say that it hasnot been examined at all would, of course, be absurd. Forevery discussion concerning fluctuations of employment, ofwhich there have been many, has been concerned with it. Imean, not that the topic has been overlooked, but that the fun-damental theory underlying it has been deemed so simple and

1This is in the Ricardian tradition. For Ricardo expressly repudiated any in-terest in the amount of the national dividend, as distinct from its distribution. Inthis he was assessing correctly the character of his own theory. But his succes-sors, less clear-sighted, have used the classical theory in discussions concerningthe causes of wealth. Vide Ricardo’s letter to Malthus of October 9, 1820: “Polit-ical Economy you think is an enquiry into the nature and causes of wealth — Ithink it should be called an enquiry into the laws which determine the divisionof the produce of industry amongst the classes who concur in its formation. Nolaw can be laid down respecting quantity, but a tolerably correct one can be laiddown respecting proportions. Every day I am more satisfied that the formerenquiry is vain and delusive, and the latter only the true objects of the science.”

5

2. THE POSTULATES OF THE CLASSICAL ECONOMICS

obvious that it has received, at the most, a bare mention.2

The classical theory of employment—supposedly simpleand obvious—has been based, I think, on two fundamentalpostulates, though practically without discussion, namely:

I. The wage is equal to the marginal product of labourThat is to say, the wage of an employed person is equal

to the value which would be lost if employment were to bereduced by one unit (after deducting any other costs whichthis reduction of output would avoid); subject, however, tothe qualification that the equality may be disturbed, in accor-dance with certain principles, if competition and markets areimperfect.

II. The utility of the wage when a given volume of labour is em-ployed is equal to the marginal disutility of that amount of employ-ment.

That is to say, the real wage of an employed person is thatwhich is just sufficient (in the estimation of the employed per-sons themselves) to induce the volume of labour actually em-ployed to be forthcoming; subject to the qualification that theequality for each individual unit of labour may be disturbedby combination between employable units analogous to theimperfections of competition which qualify the first postulate.Disutility must be here understood to cover every kind of rea-son which might lead a man, or a body of men, to withholdtheir labour rather than accept a wage which had to them autility below a certain minimum.

This postulate is compatible with what may be called ‘fric-tional’ unemployment. For a realistic interpretation of it legit-imately allows for various inexactnesses of adjustment which

2For example, Prof. Pigou in the Economics of Welfare (4th ed. p. 127) writes(my italics): “Throughout this discussion, except when the contrary is expresslystated, the fact that some resources are generally unemployed against the willof the owners is ignored. This does not affect the substance of the argument, whileit simplifies its exposition.”. Thus, whilst Ricardo expressly disclaimed any at-tempt to deal with the amount of the national dividend as a whole, Prof. Pigou,in a book which is specifically directed to the problem of the national dividend,maintains that the same theory holds when there is some involuntary unem-ployment as in the case of full employment.

6

I.

stand in the way of continuous full employment: for example,unemployment due to a temporary want of balance betweenthe relative quantities of specialised resources as a result ofmiscalculation or intermittent demand; or to time-lags con-sequent on unforeseen changes; or to the fact that the change-over from one employment to another cannot be effected with-out a certain delay, so that there will always exist in a non-static society a proportion of resources unemployed ‘betweenjobs’. In addition to ‘frictional’ unemployment, the postulateis also compatible with voluntary’ unemployment due to therefusal or inability of a unit of labour, as a result of legislationor social practices or of combination for collective bargainingor of slow response to change or of mere human obstinacy,to accept a reward corresponding to the value of the prod-uct attributable to its marginal productivity. But these twocategories of ‘frictional’ unemployment and ‘voluntary’ un-employment are comprehensive. The classical postulates donot admit of the possibility of the third category, which I shalldefine below as ‘involuntary’ unemployment.

Subject to these qualifications, the volume of employed re-sources is duly determined, according to the classical theory,by the two postulates. The first gives us the demand sched-ule for employment, the second gives us the supply schedule;and the amount of employment is fixed at the point where theutility of the marginal product balances the disutility of themarginal employment. It would follow from this that thereare only four possible means of increasing employment:

(a) An improvement in organisation or in foresight which di-minishes ’frictional’ unemployment;

(b) a decrease in the marginal disutility of labour, as expressedby the real wage for which additional labour is available,so as to diminish ’voluntary’ unemployment;

(c) an increase in the marginal physical productivity of labourin the wage-goods industries (to use Professor Pigou’s con-venient term for goods upon the price of which the utilityof the money-wage depends); or

7


(d) an increase in the price of non-wage-goods compared withthe price of wage-goods, associated with a shift in the ex-penditure of non-wage-earners from wage-goods to non-wage-goods.

This, to the best of my understanding, is the substance ofProfessor Pigou’s Theory of Unemployment—the only detailedaccount of the classical theory of employment which exists.3

II

Is it true that the above categories are comprehensive in viewof the fact that the population generally is seldom doing asmuch work as it would like to do on the basis of the currentwage? For, admittedly, more labour would, as a rule, be forth-coming at the existing money-wage if it were demanded.4 Theclassical school reconcile this phenomenon with their secondpostulate by arguing that, while the demand for labour at theexisting money-wage may be satisfied before everyone will-ing to work at this wage is employed, this situation is due toan open or tacit agreement amongst workers not to work forless, and that if labour as a whole would agree to a reductionof money-wages more employment would be forthcoming. Ifthis is the case, such unemployment, though apparently in-voluntary, is not strictly so, and ought to be included underthe above category of ‘voluntary’ unemployment due to theeffects of collective bargaining, etc.

This calls for two observations, the first of which relates tothe actual attitude of workers towards real wages and money-wages respectively and is not theoretically fundamental, butthe second of which is fundamental.

Let us assume, for the moment, that labour is not preparedto work for a lower money-wage and that a reduction in the

3Prof. Pigou’s Theory of Unemployment is examined in more detail in theAppendix to Chapter 19 below.

4Cf. the quotation from Prof. Pigou above, p. 5, footnote.

8

II.

existing level of money-wages would lead, through strikes orotherwise, to a withdrawal from the labour market of labourwhich is now employed. Does it follow from this that the ex-isting level of real wages accurately measures the marginaldisutility of labour? Not necessarily. For, although a reduc-tion in the existing money-wage would lead to a withdrawalof labour, it does not follow that a fall in the value of the ex-isting money-wage in terms of wage-goods would do so, if itwere due to a rise in the price of the latter. In other words,it may be the case that within a certain range the demand oflabour is for a minimum money-wage and not for a minimumreal wage. The classical school have tacitly assumed that thiswould involve no significant change in their theory. But thisis not so. For if the supply of labour is not a function of realwages as its sole variable, their argument breaks down en-tirely and leaves the question of what the actual employmentwill be quite indeterminate.5 They do not seem to have re-alised that, unless the supply of labour is a function of realwages alone, their supply curve for labour will shift bodilywith every movement of prices. Thus their method is tied upwith their very special assumptions, and cannot be adapted todeal with the more general case.

Now ordinary experience tells us, beyond doubt, that a sit-uation where labour stipulates (within limits) for a money-wage rather than a real wage, so far from being a mere pos-sibility, is the normal case. Whilst workers will usually resista reduction of money-wages, it is not their practice to with-draw their labour whenever there is a rise in the price of wage-goods. It is sometimes said that it would be illogical for labourto resist a reduction of money-wages but not to resist a reduc-tion of real wages. For reasons given below (p. 14), this mightnot be so illogical as it appears at first; and, as we shall seelater, fortunately so. But, whether logical or illogical, experi-ence shows that this is how labour in fact behaves.

Moreover, the contention that the unemployment whichcharacterises a depression is due to a refusal by labour to ac-

5This point is dealt with in detail in the Appendix to Chapter 19 below.

9


cept a reduction of money-wages is not clearly supported bythe facts. It is not very plausible to assert that unemploymentin the United States in 1932 was due either to labour obsti-nately refusing to accept a reduction of money-wages or toits obstinately demanding a real wage beyond what the pro-ductivity of the economic machine was capable of furnishing.Wide variations are experienced in the volume of employmentwithout any apparent change either in the minimum real de-mands of labour or in its productivity. Labour is not more tru-culent in the depression than in the boom—far from it. Noris its physical productivity less. These facts from experienceare a prima facie ground for questioning the adequacy of theclassical analysis.

It would be interesting to see the results of a statistical en-quiry into the actual relationship between changes in money-wages and changes in real wages. In the case of a change pe-culiar to a particular industry one would expect the change inreal wages to be in the same direction as the change in money-wages. But in the case of changes in the general level of wages,it will be found, I think, that the change in real wages associ-ated with a change in money-wages, so far from being usu-ally in the same direction, is almost always in the oppositedirection. When money-wages are rising, that is to say, it willbe found that real wages are falling; and when money-wagesare falling, real wages are rising. This is because, in the shortperiod, falling money-wages and rising real wages are each,for independent reasons, likely to accompany decreasing em-ployment; labour being readier to accept wage-cuts when em-ployment is falling off, yet real wages inevitably rising in thesame circumstances on account of the increasing marginal re-turn to a given capital equipment when output is diminished.

If, indeed, it were true that the existing real wage is a min-imum below which more labour than is now employed willnot be forthcoming in any circumstances, involuntary unem-ployment, apart from frictional unemployment, would be non-existent. But to suppose that this is invariably the case wouldbe absurd. For more labour than is at present employed isusually available at the existing money-wage, even though

10

II.

the price of wage-goods is rising and, consequently, the realwage falling. If this is true, the wage-goods equivalent ofthe existing money-wage is not an accurate indication of themarginal disutility of labour, and the second postulate doesnot hold good.

But there is a more fundamental objection. The secondpostulate flows from the idea that the real wages of labourdepend on the wage bargains which labour makes with theentrepreneurs. It is admitted, of course, that the bargains areactually made in terms of money, and even that the real wagesacceptable to labour are not altogether independent of whatthe corresponding money-wage happens to be. Neverthelessit is the money-wage thus arrived at which is held to deter-mine the real wage. Thus the classical theory assumes that itis always open to labour to reduce its real wage by accepting areduction in its money-wage. The postulate that there is a ten-dency for the real wage to come to equality with the marginaldisutility of labour clearly presumes that labour itself is in aposition to decide the real wage for which it works, thoughnot the quantity of employment forthcoming at this wage.

The traditional theory maintains, in short, that the wage bar-gains between the entrepreneurs and the workers determine the realwage; so that, assuming free competition amongst employersand no restrictive combination amongst workers, the lattercan, if they wish, bring their real wages into conformity withthe marginal disutility of the amount of employment offeredby the employers at that wage. If this is not true, then thereis no longer any reason to expect a tendency towards equalitybetween the real wage and the marginal disutility of labour.

The classical conclusions are intended, it must be remem-bered, to apply to the whole body of labour and do not meanmerely that a single individual can get employment by ac-cepting a cut in money-wages which his fellows refuse. Theyare supposed to be equally applicable to a closed system asto an open system, and are not dependent on the character-istics of an open system or on the effects of a reduction ofmoney-wages in a single country on its foreign trade, whichlie, of course, entirely outside the field of this discussion. Nor

11


are they based on indirect effects due to a lower wages-bill interms of money having certain reactions on the banking sys-tem and the state of credit, effects which we shall examinein detail in chapter 19. They are based on the belief that in aclosed system a reduction in the general level of money-wageswill be accompanied, at any rate in the short period and sub-ject only to minor qualifications, by some, though not alwaysa proportionate, reduction in real wages.

Now the assumption that the general level of real wagesdepends on the money-wage bargains between the employ-ers and the workers is not obviously true. Indeed it is strangethat so little attempt should have been made to prove or torefute it. For it is far from being consistent with the generaltenor of the classical theory, which has taught us to believethat prices are governed by marginal prime cost in terms ofmoney and that money-wages largely govern marginal primecost. Thus if money-wages change, one would have expectedthe classical school to argue that prices would change in al-most the same proportion, leaving the real wage and the levelof unemployment practically the same as before, any smallgain or loss to labour being at the expense or profit of otherelements of marginal cost which have been left unaltered.6

They seem, however, to have been diverted from this line ofthought, partly by the settled conviction that labour is in a po-sition to determine its own real wage and partly, perhaps, bypreoccupation with the idea that prices depend on the quan-tity of money. And the belief in the proposition that labouris always in a position to determine its own real wage, onceadopted, has been maintained by its being confused with theproposition that labour is always in a position to determinewhat real wage shall correspond to full employment, i.e. themaximum quantity of employment which is compatible with agiven real wage.

To sum up: there are two objections to the second postulate

6This argument would, indeed, contain, to my thinking, a large elementof truth, though the complete results of a change in money-wages are morecomplex, as we shall show in Chapter 19 below.

12

III.

of the classical theory. The first relates to the actual behaviourof labour. A fall in real wages due to a rise in prices, withmoney-wages unaltered, does not, as a rule, cause the supplyof available labour on offer at the current wage to fall belowthe amount actually employed prior to the rise of prices. Tosthat it does is to suppose that all those who are now unem-ployed though willing to work at the current wage will with-draw the offer of their labour in the event of even a small risein the cost of living. Yet this strange supposition apparentlyunderlies Professor Pigou’s Theory of Unemployment7, and it iswhat all members of the orthodox school are tacitly assuming.

But the other, more fundamental, objection, which we shalldevelop in the ensuing chapters, flows from our disputing theassumption that the general level of real wages is directly de-termined by the character of the wage bargain. In assumingthat the wage bargain determines the real wage the classicalschool have slipt in an illicit assumption. For there may be nomethod available to labour as a whole whereby it can bringthe wage-goods equivalent of the general level of money wagesinto conformity with the marginal disutility of the current vol-ume of employment. There may exist no expedient by whichlabour as a whole can reduce its real wage to a given figure bymaking revised money bargains with the entrepreneurs. Thiswill be our contention. We shall endeavour to show that pri-marily it is certain other forces which determine the generallevel of real wages. The attempt to elucidate this problem willbe one of our main themes. We shall argue that there has beena fundamental misunderstanding of how in this respect theeconomy in which we live actually works.

III

Though the struggle over money-wages between individualsand groups is often believed to determine the general levelof real-wages, it is, in fact, concerned with a different object.

7Cf. Chapter 19, Appendix.

13


Since there is imperfect mobility of labour, and wages do nottend to an exact equality of net advantage in different occu-pations, any individual or group of individuals, who consentto a reduction of money-wages relatively to others, will suffera relative reduction in real wages, which is a sufficient jus-tification for them to resist it. On the other hand it wouldbe impracticable to resist every reduction of real wages, dueto a change in the purchasing-power of money which affectsall workers alike; and in fact reductions of real wages arisingin this way are not, as a rule, resisted unless they proceed toan extreme degree. Moreover, a resistance to reductions inmoney-wages applying to particular industries does not raisethe same insuperable bar to an increase in aggregate employ-ment which would result from a similar resistance to everyreduction in real wages.

In other words, the struggle about money-wages primar-ily affects the distribution of the aggregate real wage betweendifferent labour-groups, and not its average amount per unitof employment, which depends, as we shall see, on a differ-ent set of forces. The effect of combination on the part of agroup of workers is to protect their relative real wage. Thegeneral level of real wages depends on the other forces of theeconomic system.

Thus it is fortunate that the workers, though unconsciously,are instinctively more reasonable economists than the classicalschool, inasmuch as they resist reductions of money-wages,which are seldom or never of an all-round character, eventhough the existing real equivalent of these wages exceedsthe marginal disutility of the existing employment; whereasthey do not resist reductions of real wages, which are associ-ated with increases in aggregate employment and leave rela-tive money-wages unchanged, unless the reduction proceedsso far as to threaten a reduction of the real wage below themarginal disutility of the existing volume of employment. Ev-ery trade union will put up some resistance to a cut in money-wages, however small. But since no trade union would dreamof striking on every occasion of a rise in the cost of living, theydo not raise the obstacle to any increase in aggregate employ-

14

IV.

ment which is attributed to them by the classical school.

IV

We must now define the third category of unemployment,namely ‘involuntary’ unemployment in the strict sense, thepossibility of which the classical theory does not admit.

Clearly we do not mean by ‘involuntary’ unemploymentthe mere existence of an unexhausted capacity to work. Aneight-hour day does not constitute unemployment because itis not beyond human capacity to work ten hours. Nor shouldwe regard as ‘involuntary’ unemployment the withdrawal oftheir labour by a body of workers because they do not chooseto work for less than a certain real reward. Furthermore, itwill be convenient to exclude ’frictional’ unemployment fromour definition of ‘involuntary’ unemployment. My definitionis, therefore, as follows: Men are involuntarily unemployed If, inthe event of a small rise in the price of wage-goods relatively to themoney-wage, both the aggregate supply of labour willing to workfor the current money-wage and the aggregate demand for it at thatwage would be greater than the existing volume of employment. Analternative definition, which amounts, however, to the samething, will be given in the next chapter (Chapter 3).

It follows from this definition that the equality of the realwage to the marginal disutility of employment presupposedby the second postulate, realistically interpreted, correspondsto the absence of ‘involuntary’ unemployment. This state ofaffairs we shall describe as ‘full’ employment, both ‘frictional’and ‘voluntary’ unemployment being consistent with ‘full’ em-ployment thus defined. This fits in, we shall find, with othercharacteristics of the classical theory, which is best regardedas a theory of distribution in conditions of full employment.So long as the classical postulates hold good, unemployment,which is in the above sense involuntary, cannot occur. Ap-parent unemployment must, therefore, be the result either oftemporary loss of work of the ‘between jobs’ type or of inter-mittent demand for highly specialised resources or of the ef-

15


fect of a trade union ‘closed shop’ on the employment of freelabour. Thus writers in the classical tradition, overlooking thespecial assumption underlying their theory, have been driveninevitably to the conclusion, perfectly logical on their assump-tion, that apparent unemployment (apart from the admittedexceptions) must be due at bottom to a refusal by the unem-ployed factors to accept a reward which corresponds to theirmarginal productivity. A classical economist may sympathisewith labour in refusing to accept a cut in its money-wage, andhe will admit that it may not be wise to make it to meet condi-tions which are temporary; but scientific integrity forces himto declare that this refusal is, nevertheless, at the bottom of thetrouble.

Obviously, however, if the classical theory is only applica-ble to the case of full employment, it is fallacious to apply it tothe problems of involuntary unemployment—if there be sucha thing (and who will deny it?). The classical theorists resem-ble Euclidean geometers in a non-Euclidean world who, dis-covering that in experience straight lines apparently paralleloften meet, rebuke the lines for not keeping straight—as theonly remedy for the unfortunate collisions which are occur-ring. Yet, in truth, there is no remedy except to throw overthe axiom of parallels and to work out a non-Euclidean ge-ometry. Something similar is required to-day in economics.We need to throw over the second postulate of the classicaldoctrine and to work out the behaviour of a system in whichinvoluntary unemployment in the strict sense is possible.

V

In emphasising our point of departure from the classical sys-tem, we must not overlook an important point of agreement.For we shall maintain the first postulate as heretofore, subjectonly to the same qualifications as in the classical theory; andwe must pause, for a moment, to consider what this involves.

It means that, with a given organisation, equipment andtechnique, real wages and the volume of output (and hence

16

V.

of employment) are uniquely correlated, so that, in general,an increase in employment can only occur to the accompani-ment of a decline in the rate of real wages. Thus I am notdisputing this vital fact which the classical economists have(rightly) asserted as indefeasible. In a given state of organi-sation, equipment and technique, the real wage earned by aunit of labour has a unique (inverse) correlation with the vol-ume of employment. Thus if employment increases, then, inthe short period, the reward per unit of labour in terms ofwage-goods must, in general, decline and profits increase.8

This is simply the obverse of the familiar proposition that in-dustry is normally working subject to decreasing returns inthe short period during which equipment etc. is assumed tobe constant; so that the marginal product in the wage-good in-dustries (which governs real wages) necessarily diminishes asemployment is increased. So long, indeed, as this propositionholds, any means of increasing employment must lead at thesame time to a diminution of the marginal product and henceof the rate of wages measured in terms of this product.

But when we have thrown over the second postulate, adecline in employment, although necessarily associated withlabour’s receiving a wage equal in value to a larger quantity ofwage-goods, is not necessarily due to labour’s demanding alarger quantity of wage-goods; and a willingness on the partof labour to accept lower money-wages is not necessarily aremedy for unemployment. The theory of wages in relationto employment, to which we are here leading up, cannot befully elucidated, however, until chapter 19 and its Appendixhave been reached.

8The argument runs as follows: n men are employed, the nth man adds abushel a day to the harvest, and wages have a buying power of a bushel a day.The n + 1th man, however, would only add .9 bushel a day, and employmentcannot, therefore, rise to n + 1 men unless the price of corn rises relatively towages until daily wages have a buying power of .9 bushel. Aggregate wageswould then amount to 9

10 (n + 1) bushels as compared with n bushels previ-ously. Thus the employment of an additional man will, if it occurs, necessarilyinvolve a transfer of income from those previously in work to the entrepreneurs.

17


VI

From the time of Say and Ricardo the classical economistshave taught that supply creates its own demand;—meaningby this in some significant, but not clearly defined, sense thatthe whole of the costs of production must necessarily be spentin the aggregate, directly or indirectly, on purchasing the prod-uct.

In J.S. Mill’s Principles of Political Economy the doctrine isexpressly set forth:

What constitutes the means of payment for com-modities is simply commodities. Each person’smeans of paying for the productions of other peo-ple consist of those which he himself possesses.All sellers are inevitably, and by the meaning ofthe word, buyers. Could we suddenly double theproductive powers of the country, we should dou-ble the supply of commodities in every market;but we should, by the same stroke, double the pur-chasing power. Everybody would bring a dou-ble demand as well as supply; everybody wouldbe able to buy twice as much, because every onewould have twice as much to offer in exchange.

As a corollary of the same doctrine, it has been supposedthat any individual act of abstaining from consumption neces-sarily leads to, and amounts to the same thing as, causing thelabour and commodities thus released from supplying con-sumption to be invested in the production of capital wealth.The following passage from Marshall’s Pure Theory of DomesticValues9 illustrates the traditional approach:

The whole of a man’s income is expended inthe purchase of services and of commodities. Itis indeed commonly said that a man spends some

9p. 34

18

VI.

portion of his income and saves another. But it isa familiar economic axiom that a man purchaseslabour and commodities with that portion of hisincome which he saves just as much as he doeswith that he is said to spend. He is said to spendwhen he seeks to obtain present enjoyment fromthe services and commodities which he purchases.He is said to save when he causes the labour andthe commodities which he purchases to be devotedto the production of wealth from which he expectsto derive the means of enjoyment in the future.

It is true that it would not be easy to quote comparable pas-sages from Marshall’s later work10 or from Edgeworth or Pro-fessor Pigou. The doctrine is never stated to-day in this crudeform. Nevertheless it still underlies the whole classical theory,which would collapse without it. Contemporary economists,who might hesitate to agree with Mill, do not hesitate to ac-cept conclusions which require Mill’s doctrine as their pre-miss. The conviction, which runs, for example, through al-most all Professor Pigou’s work, that money makes no realdifference except frictionally and that the theory of produc-tion and employment can be worked out (like Mill’s) as beingbased on ’real’ exchanges with money introduced perfuncto-rily in a later chapter, is the modern version of the classicaltradition. Contemporary thought is still deeply steeped in thenotion that if people do not spend their money in one waythey will spend it in another.11 Post-war economists seldom,

10Mr. J. A. Hobson, after quoting in his Physiology of Industry (p. 102) theabove passage from Mill, points out that Marshall commented as follows onthis passage as early as his Economics of Industry, p. 154. “But though menhave the power to purchase, they may not choose to use it.” “But”, Mr Hobsoncontinues, “he fails to grasp the critical importance of this fact, and appears tolimit its action to periods of ‘crisis’.” This has remained fair comment, I think,in the light of Marshall’s later work.

11Cf. Alfred and Mary Marshall, Economics of Industry, p. 17: “It is not goodfor trade to have dresses made of material which wears out quickly. For if peo-ple did not spend their means on buying new dresses they would spend them

19


indeed, succeed in maintaining this standpoint consistently;for their thought to-day is too much permeated with the con-trary tendency and with facts of experience too obviously in-consistent with their former view.12 But they have not drawnsufficiently far-reaching consequences; and have not revisedtheir fundamental theory.

In the first instance, these conclusions may have been ap-plied to the kind of economy in which we actually live by falseanalogy from some kind of non-exchange Robinson Crusoeeconomy, in which the income which individuals consume orretain as a result of their productive activity is, actually andexclusively, the output in specie of that activity. But, apartfrom this, the conclusion that the costs of output are alwayscovered in the aggregate by the sale-proceeds resulting fromdemand, has great plausibility, because it is difficult to dis-tinguish it from another, similar-looking proposition which isindubitable, namely that the income derived in the aggregateby all the elements in the community concerned in a produc-tive activity necessarily has a value exactly equal to the valueof the output.

Similarly it is natural to suppose that the act of an individ-ual, by which he enriches himself without apparently takinganything from anyone else, must also enrich the communityas a whole; so that (as in the passage just quoted from Mar-shall) an act of individual saving inevitably leads to a parallelact of investment. For, once more, it is indubitable that thesum of the net increments of the wealth of individuals mustbe exactly equal to the aggregate net increment of the wealthof the community.

Those who think in this way are deceived, nevertheless,by an optical illusion, which makes two essentially different

on giving employment to labour in some other way.” The reader will notice thatI am again quoting from the earlier Marshall. The Marshall of the Principles hadbecome sufficiently doubtful to be very cautious and evasive. But the old ideaswere never repudiated or rooted out of the basic assumptions of his thought.

12It is this distinction of Prof. Robbins that he, almost alone, continues tomaintain a consistent scheme of thought, his practical recommendations be-longing to the same system as his theory.

20

VII.

activities appear to be the same. They are fallaciously sup-posing that there is a nexus which unites decisions to abstainfrom present consumption with decisions to provide for fu-ture consumption; whereas the motives which determine thelatter are not linked in any simple way with the motives whichdetermine the former.

It is, then, the assumption of equality between the demandprice of output as a whole and its supply price which is to beregarded as the classical theory’s ‘axiom of parallels’. Grantedthis, all the rest follows—the social advantages of private andnational thrift, the traditional attitude towards the rate of in-terest, the classical theory of unemployment, the quantity the-ory of money, the unqualified advantages of laissez-faire in re-spect of foreign trade and much else which we shall have toquestion.

VII

At different points in this chapter we have made the classicaltheory to depend in succession on the assumptions:

1. that the real wage is equal to the marginal disutility ofthe existing employment;

2. that there is no such thing as involuntary unemploy-ment in the strict sense;

3. that supply creates its own demand in the sense that theaggregate demand price is equal to the aggregate supplyprice for all levels of output and employment.

These three assumptions, however, all amount to the samething in the sense that they all stand and fall together, any oneof them logically involving the other two.

21

CHAPTER 3

THE PRINCIPLE OF EFFECTIVE DEMAND

I

We need, to start with, a few terms which will be definedprecisely later. In a given state of technique, resources andcosts, the employment of a given volume of labour by an en-trepreneur involves him in two kinds of expense: first of all,the amounts which he pays out to the factors of production(exclusive of other entrepreneurs) for their current services,which we shall call the factor cost of the employment in ques-tion; and secondly, the amounts which he pays out to other en-trepreneurs for what he has to purchase from them togetherwith the sacrifice which he incurs by employing the equip-ment instead of leaving it idle, which we shall call the usercost of the employment in question.1 The excess of the valueof the resulting output over the sum of its factor cost and itsuser cost is the profit or, as we shall call it, the income of theentrepreneur. The factor cost is, of course, the same thing,looked at from the point of view of the entrepreneur, as whatthe factors of production regard as their income. Thus the fac-tor cost and the entrepreneur’s profit make up, between them,what we shall define as the total income resulting from theemployment given by the entrepreneur. The entrepreneur’sprofit thus defined is, as it should be, the quantity which heendeavours to maximise when he is deciding what amount ofemployment to offer. It is sometimes convenient, when we arelooking at it from the entrepreneur’s standpoint, to call theaggregate income (i.e. factor cost plus profit) resulting froma given amount of employment the proceeds of that employ-

1A precise definition of user cost will be given in Chapter 6.

23

3. THE PRINCIPLE OF EFFECTIVE DEMAND

ment. On the other hand, the aggregate supply price2 of theoutput of a given amount of employment is the expectationof proceeds which will just make it worth the while of the en-trepreneurs to give that employment.3

It follows that in a given situation of technique, resourcesand factor cost per unit of employment, the amount of em-ployment, both in each individual firm and industry and inthe aggregate, depends on the amount of the proceeds whichthe entrepreneurs expect to receive from the correspondingoutput.4 For entrepreneurs will endeavour to fix the amountof employment at the level which they expect to maximise theexcess of the proceeds over the factor cost.

2Not to be confused (vide infra) with the supply price of a unit of output inthe ordinary sense of this term.

3The reader will observe that I am deducting the user cost both from theproceeds and from the aggregate supply price of a given volume of output, so thatboth these terms are to be interpreted net of user cost; whereas the aggregatesums paid by the purchasers are, of course, gross of user cost. The reasons whythis is convenient will be given in Chapter 6. The essential point is that theaggregate proceeds and aggregate supply price net of user cost can be defineduniquely and unambiguously; whereas, since user cost is obviously dependentboth on the degree of integration of industry and on the extent to which en-trepreneurs buy from one another, there can be no definition of the aggregatesums paid by purchasers, inclusive of user cost, which is independent of thesefactors. There is a similar difficulty even in defining supply price in the ordinarysense for an individual producer; and in the case of the aggregate supply priceof output as a whole serious difficulties of duplication are involved, which havenot always been faced. If the term is to be interpreted gross of user cost, theycan only be overcome by making special assumptions relating to the integra-tion of entrepreneurs in groups according as they produce consumption-goodsor capital-goods which are obscure and complicated in themselves and do notcorrespond to the facts. If, however, aggregate supply price is defined as abovenet of user cost, thew difficulties do not arise. The reader is advised, however,to await the fuller discussion in Chapter 6 and its appendix.

4An entrepreneur, who has to reach a practical decision as to his scale ofproduction, does not, of course, entertain a single undoubting expectation ofwhat the sale-proceeds of a given output will be, but several hypothetical ex-pectations held with varying degrees of probability and definiteness. By hisexpectation of proceeds I mean, therefore, that expectation of proceeds which,if it were held with certainty, would lead to the same behaviour as does the bun-dle of vague and more various possibilities which actually makes up his stateof expectation when he reaches his decision.

24

I.

Let Z be the aggregate supply price of the output from em-ployingN men, the relationship between Z andN being writ-ten Z = φ(N), which can be called the aggregate supply func-tion.5 Similarly, let D be the proceeds which entrepreneursexpect to receive from the employment of N men, the rela-tionship between D and N being written D = f(N), whichcan be called the aggregate demand function.

Now if for a given value of N the expected proceeds aregreater than the aggregate supply price, i.e. if D is greaterthan Z, there will be an incentive to entrepreneurs to increaseemployment beyond N and, if necessary, to raise costs bycompeting with one another for the factors of production, upto the value ofN for whichZ has become equal toD. Thus thevolume of employment is given by the point of intersectionbetween the aggregate demand function and the aggregatesupply function; for it is at this point that the entrepreneurs’expectation of profits will be maximised. The value of D atthe point of the aggregate demand function, where it is in-tersected by the aggregate supply function, will be called theeffective demand. Since this is the substance of the General The-ory of Employment, which it will be our object to expound,the succeeding chapters will be largely occupied with exam-ining the various factors upon which these two functions de-pend.

The classical doctrine, on the other hand, which used to beexpressed categorically in the statement that ‘Supply createsits own Demand’ and continues to underlie all orthodox eco-nomic theory, involves a special assumption as to the relation-ship between these two functions. For ‘Supply creates its ownDemand’ must mean that f(N) and φ(N) are equal for all val-ues ofN , i.e. for all levels of output and employment; and thatwhen there is an increase in Z ( = φ(N)) corresponding to anincrease in N , D ( = f(N)) necessarily increases by the sameamount as Z. The classical theory assumes, in other words,that the aggregate demand price (or proceeds) always accom-

5In Chapter 20 a function closely related to the above will be called theemployment function.

25


modates itself to the aggregate supply price; so that, whateverthe value of N may be, the proceeds D assume a value equalto the aggregate supply price Z which corresponds toN . Thatis to say, effective demand, instead of having a unique equilib-rium value, is an infinite range of values all equally admissi-ble; and the amount of employment is indeterminate except inso far as the marginal disutility of labour sets an upper limit.

If this were true, competition between entrepreneurs wouldalways lead to an expansion of employment up to the point atwhich the supply of output as a whole ceases to be elastic, i.e.where a further increase in the value of the effective demandwill no longer be accompanied by any increase in output. Ev-idently this amounts to the same thing as full employment.In the previous chapter we have given a definition of full em-ployment in terms of the behaviour of labour. An alternative,though equivalent, criterion is that at which we have now ar-rived, namely a situation in which aggregate employment isinelastic in response to an increase in the effective demand forits output. Thus Say’s law, that the aggregate demand price ofoutput as a whole is equal to its aggregate supply price for allvolumes of output, is equivalent to the proposition that thereis no obstacle to full employment. If, however, this is not thetrue law relating the aggregate demand and supply functions,there is a vitally important chapter of economic theory whichremains to be written and without which all discussions con-cerning the volume of aggregate employment are futile.

II

A brief summary of the theory of employment to be workedout in the course of the following chapters may, perhaps, helpthe reader at this stage, even though it may not be fully intel-ligible. The terms involved will be more carefully defined indue course. In this summary we shall assume that the money-wage and other factor costs are constant per unit of labouremployed. But this simplification, with which we shall dis-pense later, is introduced solely to facilitate the exposition.

26

II.

The essential character of the argument is precisely the samewhether or not money-wages, etc., are liable to change.

The outline of our theory can be expressed as follows. Whenemployment increases, aggregate real income is increased. Thepsychology of the community is such that when aggregatereal income is increased aggregate consumption is increased,but not by so much as income. Hence employers would makea loss if the whole of the increased employment were to be de-voted to satisfying the increased demand for immediate con-sumption. Thus, to justify any given amount of employmentthere must be an amount of current investment sufficient toabsorb the excess of total output over what the communitychooses to consume when employment is at the given level.For unless there is this amount of investment, the receipts ofthe entrepreneurs will be less than is required to induce themto offer the given amount of employment. It follows, there-fore, that, given what we shall call the community’s propen-sity to consume, the equilibrium level of employment, i.e. thelevel at which there is no inducement to employers as a wholeeither to expand or to contract employment, will depend onthe amount of current investment. The amount of currentinvestment will depend, in turn, on what we shall call theinducement to invest; and the inducement to invest will befound to depend on the relation between the schedule of themarginal efficiency of capital and the complex of rates of in-terest on loans of various maturities and risks.

Thus, given the propensity to consume and the rate of newinvestment, there will be only one level of employment con-sistent with equilibrium; since any other level will lead to in-equality between the aggregate supply price of output as awhole and its aggregate demand price. This level cannot begreater than full employment, i.e. the real wage cannot be lessthan the marginal disutility of labour. But there is no reasonin general for expecting it to be equal to full employment. Theeffective demand associated with full employment is a spe-cial case, only realised when the propensity to consume andthe inducement to invest stand in a particular relationship toone another. This particular relationship, which corresponds

27


to the assumptions of the classical theory, is in a sense an op-timum relationship. But it can only exist when, by accidentor design, current investment provides an amount of demandjust equal to the excess of the aggregate supply price of theoutput resulting from full employment over what the com-munity will choose to spend on consurnption when it is fullyemployed.

This theory can be summed up in the following proposi-tions:

(1) In a given situation of technique, resources and costs, in-come (both money-income and real income) depends onthe volume of employment N .

(2) The relationship between the community’s income andwhat it can be expected to spend on consumption, desig-nated by D1, will depend on the psychological character-istic of the community, which we shall call its propensity toconsume. That is to say, consumption will depend on thelevel of aggregate income and, therefore, on the level ofemployment N , except when there is some change in thepropensity to consume.

(3) The amount of labour N which the entrepreneurs decideto employ depends on the sum (D) of two quantities, namelyD1, the amount which the community is expected to spendon consumption, and D2, the amount which it is expectedto devote to new investment. D is what we have calledabove the effective demand.

(4) Since D1 + D2 = D = φ(N), where is the aggregate sup-ply function, and since, as we have seen in (2) above,D1 isa function of N , which we may write χ(N), depending onthe propensity to consume, it follows that φ(N)−χ(N) =D2.

(5) Hence the volume of employment in equilibrium dependson (i) the aggregate supply function, (ii) the propensity toconsume, and (iii) the volume of investment, D2. This isthe essence of the General Theory of Employment.

28

II.

(6) For every value of N there is a corresponding marginalproductivity of labour in the wage-goods industries; andit is this which determines the real wage. (5) is, therefore,subject to the condition that N cannot exceed the valuewhich reduces the real wage to equality with the marginaldisutility of labour. This means that not all changes in Dare compatible with our temporary assumption that money-wages are constant. Thus it will be essential to a full state-ment of our theory to dispense with this assumption.

(7) On the classical theory, according to which D = φ(N)for all values of N , the volume of employment is in neu-tral equilibrium for all values of N less than its maximumvalue; so that the forces of competition between entrepreneursmay be expected to push it to this maximum value. Onlyat this point, on the classical theory, can there be stableequilibrium.

(8) When employment increases, D1 will increase, but not byso much as D; since when our income increases our con-sumption increases also, but not by so much. The keyto our practical problem is to be found in this psycho-logical law. For it follows from this that the greater thevolume of employment the greater will be the gap be-tween the aggregate supply price (Z) of the correspond-ing output and the sum (D1) which the entrepreneurs canexpect to get back out of the expenditure of consumers.Hence, if there is no change in the propensity to consume,employment cannot increase, unless at the same time D2

is increasing so as to fill the increasing gap between Zand D1. Thus—except on the special assumptions of theclassical theory according to which there is some forcein operation which, when employment increases, alwayscauses D2 to increase sufficiently to fill the widening gapbetween Z and D1—the economic system may find itselfin stable equilibrium with N at a level below full employ-ment, namely at the level given by the intersection of theaggregate demand function with the aggregate supply func-tion.

29


Thus the volume of employment is not determined by themarginal disutility of labour measured in terms of real wages,except in so far as the supply of labour available at a given realwage sets a maximum level to employment. The propensity toconsume and the rate of new investment determine betweenthem the volume of employment, and the volume of employ-ment is uniquely related to a given level of real wages—notthe other way round. If the propensity to consume and therate of new investment result in a deficient effective demand,the actual level of employment will fall short of the supplyof labour potentially available at the existing real wage, andthe equilibrium real wage will be greater than the marginaldisutility of the equilibrium level of employment.

This analysis supplies us with an explanation of the para-dox of poverty in the midst of plenty. For the mere existenceof an insufficiency of effective demand may, and often will,bring the increase of employment to a standstill before a levelof full employment has been reached. The insufficiency of ef-fective demand will inhibit the process of production in spiteof the fact that the marginal product of labour still exceeds invalue the marginal disutility of employment.

Moreover the richer the community, the wider will tendto be the gap between its actual and its potential production;and therefore the more obvious and outrageous the defects ofthe economic system. For a poor community will be prone toconsume by far the greater part of its output, so that a verymodest measure of investment will be sufficient to providefull employment; whereas a wealthy community will have todiscover much ampler opportunities for investment if the sav-ing propensities of its wealthier members are to be compatiblewith the employment of its poorer members. If in a poten-tially wealthy community the inducement to invest is weak,then, in spite of its potential wealth, the working of the prin-ciple of effective demand will compel it to reduce its actualoutput, until, in spite of its potential wealth, it has become sopoor that its surplus over its consumption is sufficiently di-minished to correspond to the weakness of the inducement toinvest.

30

III.

But worse still. Not only is the marginal propensity to con-sume6 weaker in a wealthy community, but, owing to its ac-cumulation of capital being already larger, the opportunitiesfor further investment are less attractive unless the rate of in-terest falls at a sufficiently rapid rate; which ‘brings us to thetheory of the rate of interest and to the reasons why it does notautomatically fall to the appropriate level, which will occupyBook IV.

Thus the analysis of the propensity to consume, the defini-tion of the marginal efficiency of capital and the theory of therate of interest are the three main gaps in our existing knowl-edge which it will be necessary to fill. When this has beenaccomplished, we shall find that the theory of prices falls intoits proper place as a matter which is subsidiary to our generaltheory. We shall discover, however, that money plays an es-sential part in our theory of the rate of interest; and we shallattempt to disentangle the peculiar characteristics of moneywhich distinguish it from other things.

III

The idea that we can safely neglect the aggregate demandfunction is fundamental to the Ricardian economics, whichunderlie what we have been taught for more than a century.Malthus, indeed, had vehemently opposed Ricardo’s doctrinethat it was impossible for effective demand to be deficient;but vainly. For, since Malthus was unable to explain clearly(apart from an appeal to the facts of common observation)how and why effective demand could be deficient or exces-sive, he failed to furnish an alternative construction; and Ri-cardo conquered England as completely as the Holy Inquisi-tion conquered Spain. Not only was his theory accepted bythe city, by statesmen and by the academic world. But contro-versy ceased; the other point of view completely disappeared;it ceased to be discussed. The great puzzle of effective de-

6Defined in Chapter 10, below.

31


mand with which Malthus had wrestled vanished from eco-nomic literature. You will not find it mentioned even once inthe whole works of Marshall, Edgeworth and Professor Pigou,from whose hands the classical theory has received its mostmature embodiment. It could only live on furtively, below thesurface, in the underworlds of Karl Marx, Silvio Gesell or Ma-jor Douglas.

The completeness of the Ricardian victory is something ofa curiosity and a mystery. It must have been due to a complexof suitabilities in the doctrine to the environment into whichit was projected. That it reached conclusions quite differentfrom what the ordinary uninstructed person would expect,added, I suppose, to its intellectual prestige. That its teach-ing, translated into practice, was austere and often unpalat-able, lent it virtue. That it was adapted to carry a vast andconsistent logical superstructure, gave it beauty. That it couldexplain much social injustice and apparent cruelty as an in-evitable incident in the scheme of progress, and the attemptto change such things as likely on the whole to do more harmthan good, commended it to authority. That it afforded a mea-sure of justification to the free activities of the individual cap-italist, attracted to it the support of the dominant social forcebehind authority.

But although the doctrine itself has remained unquestionedby orthodox economists up to a late date, its signal failure forpurposes of scientific prediction has greatly impaired, in thecourse of time, the prestige of its practitioners. For profes-sional economists, after Malthus, were apparently unmovedby the lack of correspondence between the results of their the-ory and the facts of observation;—a discrepancy which the or-dinary man has not failed to observe, with the result of hisgrowing unwillingness to accord to economists that measureof respect which he gives to other groups of scientists whosetheoretical results are confirmed by observation when they areapplied to the facts.

The celebrated optimism of traditional economic theory,which has led to economists being looked upon as Candides,who, having left this world for the cultivation of their gardens,

32

III.

teach that all is for the best in the best of all possible worldsprovided we will let well alone, is also to be traced, I think,to their having neglected to take account of the drag on pros-perity which can be exercised by an insufficiency of effectivedemand. For there would obviously be a natural tendencytowards the optimum employment of resources in a societywhich was functioning after the manner of the classical pos-tulates. It may well be that the classical theory represents theway in which we should like our economy to behave. But toassume that it actually does so is to assume our difficultiesaway.

33

Book II

Definitions and Ideas

CHAPTER 4

THE CHOICE OF UNITS

I

In this and the next three chapters we shall be occupied withan attempt to clear up certain perplexities which have no pe-culiar or exclusive relevance to the problems which it is ourspecial purpose to examine. Thus these chapters are in thenature of a digression, which will prevent us for a time frompursulng our main theme. Their subject-matter is only dis-cussed here because it does not happen to have been alreadytreated elsewhere in a way which I find adequate to the needsof my own particular enquiry.

The three perplexities which most impeded my progressin writing this book, so that I could not express myself conve-niently until I had found some solution for them, are: firstly,the choice of the units of quantity appropriate to the problemsof the economic system as a whole; secondly, the part playedby expectation in economic analysis; and, thirdly, the defini-tion of income.

II

That the units, in terms of which economists commonly work,are unsatisfactory can be illustrated by the concepts of the na-tional dividend, the stock of real capital and the general price-level:

(i) The national dividend, as defined by Marshall and Pro-fessor Pigou,1 measures the volume of current output or real

1Vide Pigou, Economics of Welfare, passim, and particularly Part I. chap. iii.

37

4. THE CHOICE OF UNITS

income and not the value of output or money-income.2 Fur-thermore, it depends, in some sense, on net output;—on thenet addition, that is to say, to the resources of the communityavailable for consumption or for retention as capital stock,due to the economic activities and sacrifices of the current pe-riod, after allowing for the wastage of the stock of real capitalexisting at the commencement of the period. On this basis anattempt is made to erect a quantitative science. But it is a graveobjection to this definition for such a purpose that the com-munity’s output of goods and services is a non-homogeneouscomplex which cannot be measured, strictly speaking, exceptin certain special cases, as for example when all the items ofone output are included in the same proportions in anotheroutput.

(ii) The difficulty is even greater when, in order to calcu-late net output, we try to measure the net addition to capitalequipment; for we have to find some basis for a quantitativecomparison between the new items of equipment producedduring the period and the old items which have perished bywastage. In order to arrive at the net national dividend, Pro-fessor Pigou3 deducts such obsolescence, etc., ’as may fairly becalled “normal"; and the practical test of normality is that thedepletion is sufficiently regular to be foreseen, if not in detail,at least in the large’. But, since this deduction is not a deduc-tion in terms of money, he is involved in assuming that therecan be a change in physical quantity, although there has beenno physical change; i.e. he is covertly introducing changes invalue.

Moreover, he is unable to devise any satisfactory formula4

to evaluate new equipment against old when, owing to changes

2Though, as a convenient compromise, the real income, which is taken toconstitute the National Dividend, is usually limited to those goods and serviceswhich can be bought for money.

3Economics of Welfare, Part I. chap. v., on “What is meant by maintainingCapital intact”; as amended by a recent article in the Economic Journal, June1935, p. 225.

4Cf. Prof. Hayek’s criticisms, Economica, Aug. 1935, p. 247.

38

II.

in technique, the two are not identical. I believe that the con-cept at which Professor Pigou is aiming is the right and ap-propriate concept for economic analysis. But, until a satisfac-tory system of units has been adopted, its precise definition isan impossible task. The problem of comparing one real out-put with another and of then calculating net output by settingoff new items of equipment against the wastage of old itemspresents conundrums which permit, one can confidently say,of no solution.

(iii) Thirdly, the well-known, but unavoidable, element ofvagueness which admittedly attends the concept of the gen-eral price-level makes this term very unsatisfactory for thepurposes of a causal analysis, which ought to be exact.

Nevertheless these difficulties are rightly regarded as ‘co-nundrums’. They are ‘purely theoretical’ in the sense that theynever perplex, or indeed enter in any way into, business de-cisions and have no relevance to the causal sequence of eco-nomic events, which are clear-cut and determinate in spite ofthe quantitative indeterminacy of these concepts. It is natural,therefore, to conclude that they not only lack precision but areunnecessary. Obviously our quantitative analysis must be ex-pressed without using any quantitatively vague expressions.And, indeed, as soon as one makes the attempt, it becomesclear, as I hope to show, that one can get on much better with-out them.

The fact that two incommensurable collections of miscel-laneous objects cannot in themselves provide the material fora quantitative analysis need not, of course, prevent us frommaking approximate statistical comparisons, depending onsome broad element of judgment rather than of strict calcula-tion, which may possess significance and validity within cer-tain limits.

But the proper place for such things as net real output andthe general level of prices lies within the field of historical andstatistical description, and their purpose should be to satisfyhistorical or social curiosity, a purpose for which perfect preci-sion—such as our causal analysis requires, whether or not our

39


knowledge of the actual values of the relevant quantities iscomplete or exact—is neither usual nor necessary. To say thatnet output to-day is greater, but the price-level lower, than tenyears ago or one year ago, is a proposition of a similar char-acter to the statement that Queen Victoria was a better queenbut not a happier woman than Queen Elizabeth—a proposi-tion not without meaning and not without interest, but un-suitable as material for the differential calculus. Our precisionwill be a mock precision if we try to use such partly vagueand non-quantitative concepts as the basis of a quantitativeanalysis.

III

On every particular occasion, let it be remembered, an en-trepreneur is concerned with decisions as to the scale on whichto work a given capital equipment; and when we say that theexpectation of an increased demand, i.e. a raising of the ag-gregate demand function, will lead to an increase in aggregateoutput, we really mean that the firms, which own the capitalequipment, will be induced to associate with it a greater ag-gregate employment of labour. In the case of an individualfirm or industry producing a homogeneous product we canspeak legitimately, if we wish, of increases or decreases of out-put. But when we are aggregating the activities of all firms,we cannot speak accurately except in terms of quantities ofemployment applied to a given equipment. The concepts ofoutput as a whole and its price-level are not required in thiscontext, since we have no need of an absolute measure of cur-rent aggregate output, such as would enable us to compare itsamount with the amount which would result from the associ-ation of a different capital equipment with a different quantityof employment. When, for purposes of description or roughcomparison, we wish to speak of an increase of output, wemust rely on the general presumption that the amount of em-ployment associated with a given capital equipment will be asatisfactory index of the amount of resultant output;—the two

40

III.

being presumed to increase and decrease together, though notin a definite numerical proportion.

In dealing with the theory of employment I propose, there-fore, to make use of only two fundamental units of quantity,namely, quantities of money-value and quantities of employ-ment. The first of these is strictly homogeneous, and the sec-ond can be made so. For, in so far as different grades andkinds of labour and salaried assistance enjoy a more or lessfixed relative remuneration, the quantity of employment canbe sufficiently defined for our purpose by taking an hour’semployment of ordinary labour as our unit and weighting anhour’s employment of special labour in proportion to its re-muneration; i.e. an hour of special labour remunerated atdouble ordinary rates will count as two units. We shall callthe unit in which the quantity of employment is measured thelabour-unit; and the money-wage of a labour-unit we shallcall the wage-unit.5 Thus, if E is the wages (and salaries)bill, W the wage-unit, and N the quantity of employment,E = N ×W .

This assumption of homogeneity in the supply of labouris not upset by the obvious fact of great differences in thespecialised skill of individual workers and in their suitabil-ity for different occupations. For, if the remuneration of theworkers is proportional to their efficiency, the differences aredealt with by our having regarded individuals as contribut-ing to the supply of labour in proportion to their remunera-tion; whilst if, as output increases, a given firm has to bringin labour which is less and less efficient for its special pur-poses per wage-unit paid to it, this is merely one factor amongothers leading to a diminishing return from the capital equip-ment in terms of output as more labour is employed on it.We subsume, so to speak, the non-homogeneity of equally re-munerated labour units in the equipment, which we regardas less and less adapted to employ the available labour units

5If X stands for any quantity measured in terms of money, it will often beconvenient to write Xw for the same quantity measured in terms of the wage-unit.

41


as output increases, instead of regarding the available labourunits as less and less adapted to use a homogeneous capi-tal equipment. Thus if there is no surplus of specialised orpractised labour and the use of less suitable labour involvesa higher labour cost per unit of output, this means that therate at which the return from the equipment diminishes asemployment increases is more rapid than it would be if therewere such a surplus.6 Even in the limiting case where dif-ferent labour units were so highly specialised as to be alto-gether incapable of being substituted for one another, thereis no awkwardness; for this merely means that the elasticityof supply of output from a particular type of capital equip-ment falls suddenly to zero when all the available labour spe-cialised to its use is already employed.7 Thus our assump-tion of a homogeneous unit of labour involves no difficulties

6This is the main reason why the supply price of output rises with increas-ing demand even when there is still a surplus of equipment identical in typewith the equipment in use. If we suppose that the surplus supply of labourforms a pool equally available to all entrepreneurs and that labour employedfor a given purpose is rewarded, in part at least, per unit of effort and not withstrict regard to its efficiency in its actual particular employment (which is inmost cases the realistic assumption to make), the diminishing efficiency of thelabour employed is an outstanding example of rising supply price with increas-ing output, not due to internal diseconomies.

7How the supply curve in ordinary use is supposed to deal with the abovedifficulty I cannot say, since those who use this curve have not made their as-sumptions very clear. Probably they are assuming that labour employed for agiven purpose is always rewarded with strict regard to its efficiency for thatpurpose. But this is unrealistic. Perhaps the essential reason for treating thevarying efficiency of labour as though it belonged to the equipment lies in thefact that the increasing surpluses, which emerge as output is increased, accruein practice mainly to the owners of the equipment and not to the more efficientworkers (though these may get an advantage through being employed moreregularly and by receiving earlier promotion); that is to say, men of differingefficiency working at the same job are seldom paid at rates closely proportionalto their efficiencies. Where, however, increased pay for higher efficiency occurs,and in so far as it occurs, my method takes account of it; since in calculating thenumber of labour units employed, the individual workers are weighted in pro-portion to their remuneration. On my assumptions interesting complicationsobviously arise where we are dealing with particular supply curves since theirshape will depend on the demand for suitable labour in other directions. Toignore these complications would, as I have said, be unrealistic. But we need

42

III.

unless there is great instability in the relative remunerationof different labour-units; and even this difficulty can be dealtwith, if it arises, by supposing a rapid liability to change inthe supply of labour and the shape of the aggregate supplyfunction.

It is my belief that much unnecessary perplexity can beavoided if we limit ourselves strictly to the two units, moneyand labour, when we are dealing with the behaviour of theeconomic system as a whole; reserving the use of units of par-ticular outputs and equipments to the occasions when we areanalysing the output of individual firms or industries in iso-lation; and the use of vague concepts, such as the quantityof output as a whole, the quantity of capital equipment as awhole and the general level of prices, to the occasions whenwe are attempting some historical comparison which is withincertain (perhaps fairly wide) limits avowedly unprecise andapproximate.

It follows that we shall measure changes in current out-put by reference to the number of hours of labour paid for(whether to satisfy consumers or to produce fresh capital equip-ment) on the existing capital equipment, hours of skilled labourbeing weighted in proportion to their remuneration. We haveno need of a quantitative comparison between this output andthe output which would result from associating a different setof workers with a different capital equipment. To predict howentrepreneurs possessing a given equipment will respond toa shift in the aggregate demand function it is not necessaryto know how the quantity of the resulting output, the stan-dard of life and the general level of prices would compare

not consider them when we are dealing with employment as a whole, providedwe assume that a given volume of effective demand has a particular distribu-tion of this demand between different products uniquely associated with it. Itmay be, however, that this would not hold good irrespective of the particularcause of the change in demand. E.g. an increase in effective demand due to anincreased propensity to consume might find itself faced by a different aggregatesupply function from that which would face an equal increase in demand dueto an increased inducement to invest. All this, however, belongs to the detailedanalysis of the general ideas here set forth, which it is no part of my immediatepurpose to pursue.

43


with what they were at a different date or in another coun-try.

IV

It is easily shown that the conditions of supply, such as areusually expressed in terms of the supply curve, and the elas-ticity of supply relating output to price, can be handled interms of our two chosen units by means of the aggregate sup-ply function, without reference to quantities of output, whetherwe are concerned with a particular firm or industry or witheconomic activity as a whole. For the aggregate supply func-tion for a given firm (and similarly for a given industry or forindustry as a whole) is given by

Zr = φr(Nr),

where Zr is the proceeds (net of user cost) the expectation ofwhich will induce a level of employment Nr . If, therefore,the relation between employment and output is such that anemployment Nr results in an output Or , where Or = ψr(Nr),it follows that

p =Zr + Ur(Nr)

Or=φr(Nr) + Ur(Nr)

ψr(Nr)

is the ordinary supply curve, where Ur(Nr) is the (expected)user cost corresponding to a level of employment Nr .

Thus in the case of each homogeneous commodity, for whichOr = ψr(Nr) has a definite meaning, we can evaluate Zr =φr(Nr) in the ordinary way; but we can then aggregate theNr’s in a way in which we cannot aggregate the Or’s, sinceΣOr is not a numerical quantity. Moreover, if we can assumethat, in a given environment, a given aggregate employmentwill be distributed in a unique way between different indus-tries, so that Nr is a function of N , further simplifications arepossible.

44

CHAPTER 5

EXPECTATION AS DETERMINING OUTPUT ANDEMPLOYMENT

I

All production is for the purpose of ultimately satisfying aconsumer. Time usually elapses, however—and sometimesmuch time—between the incurring of costs by the producer(with the consumer in view) and the purchase of the output bythe ultimate consumer. Meanwhile the entrepreneur (includ-ing both the producer and the investor in this description) hasto form the best expectations1 he can as to what the consumerswill be prepared to pay when he is ready to supply them (di-rectly or indirectly) after the elapse of what may be a lengthyperiod; and he has no choice but to be guided by these expec-tations, if he is to produce at all by processes which occupytime.

These expectations, upon which business decisions depend,fall into two groups, certain individuals or firms being spe-cialised in the business of framing the first type of expecta-tion and others in the business of framing the second. Thefirst type is concerned with the price which a manufacturercan expect to get for his ‘finished’ output at the time when hecommits himself to starting the process which will produce it;output being ‘finished’ (from the point of view of the manu-facturer) when it is ready to be used or to be sold to a sec-ond party. The second type is concerned with what the en-trepreneur can hope to earn in the shape of future returns ifhe purchases (or, perhaps, manufactures) ‘finished’ output as

1For the method of arriving at an equivalent of these expectations in termsof sale-proceeds see footnote (3) to p. 24 above.

45

5. EXPECTATION AS DETERMINING OUTPUT AND EMPLOYMENT

an addition to his capital equipment. We may call the formershort-term expectation and the latter long-term expectation.

Thus the behaviour of each individual firm in decidingits daily2 output will be determined by its short-term expecta-tions—expectations as to the cost of output on various pos-sible scales and expectations as to the sale-proceeds of thisoutput; though, in the case of additions to capital equipmentand even of sales to distributors, these short-term expecta-tions will largely depend on the long-term (or medium-term)expectations of other parties. It is upon these various expec-tations that the amount of employment which the firms offerwill depend. The actually realised results of the production andsale of output will only be relevant to employment in so far asthey cause a modification of subsequent expectations. Nor, onthe other hand, are the original expectations relevant, whichled the firm to acquire the capital equipment and the stock ofintermediate products and half-finished materials with whichit finds itself at the time when it has to decide the next day’soutput. Thus, on each and every occasion of such a decision,the decision will be made, with reference indeed to this equip-ment and stock, but in the light of the current expectations ofprospective costs and sale-proceeds.

Now, in general, a change in expectations (whether short-term or long-term) will only produce its full effect on employ-ment over a considerable period. The change in employmentdue to a change in expectations will not be the same on thesecond day after the change as on the first, or the same on thethird day as on the second, and so on, even though there beno further change in expectations. In the case of short-term ex-pectations this is because changes in expectation are not, as arule, sufficiently violent or rapid, when they are for the worse,to cause the abandonment of work on all the productive pro-cesses which, in the light of the revised expectation, it was amistake to have begun; whilst, when they are for the better,

2Daily here stands for the shortest interval after which the firm is free torevise its decision as to how much employment to offer. It is, so to speak, theminimum effective unit of economic time.

46

I.

some time for preparation must needs elapse before employ-ment can reach the level at which it would have stood if thestate of expectation had been revised sooner. In the case oflong-term expectations, equipment which will not be replacedwill continue to give employment until it is worn out; whilstwhen the change in long-term expectations is for the better,employment may be at a higher level at first, than it will beafter there has been time to adjust the equipment to the newsituation.

If we suppose a state of expectation to continue for a suf-ficient length of time for the effect on employment to haveworked itself out so completely that there is, broadly speak-ing, no piece of employment going on which would not havetaken place if the new state of expectation had always existed,the steady level of employment thus attained may be calledthe long-period employment3 corresponding to that state ofexpectation. It follows that, although expectation may changeso frequently that the actual level of employment has neverhad time to reach the long-period employment correspondingto the existing state of expectation, nevertheless every state ofexpectation has its definite corresponding level of long-periodemployment.

Let us consider, first of all, the process of transition to along-period position due to a change in expectation, which isnot confused or interrupted by any further change in expecta-tion. We will first suppose that the change is of such a charac-ter that the new long-period employment will be greater thanthe old. Now, as a rule, it will only be the rate of input whichwill be much affected at the beginning, that is to say, the vol-ume of work on the earlier stages of new processes of produc-tion, whilst the output of consumption-goods and the amountof employment on the later stages of processes which were

3It is not necessary that the level of long-period employment should beconstant, i.e. long-period conditions are not necessarily static. For example, asteady increase in wealth or population may constitute a part of the unchangingexpectation. The only condition is that the existing expectations should havebeen foreseen sufficiently far ahead.

47


started before the change will remain much the same as be-fore. In so far as there were stocks of partly finished goods,this conclusion may be modified; though it is likely to remaintrue that the initial increase in employment will be modest.As, however, the days pass by, employment will gradually in-crease. Moreover, it is easy to conceive of conditions whichwill cause it to increase at some stage to a higher level than thenew long-period employment. For the process of building upcapital to satisfy the new state of expectation may lead to moreemployment and also to more current consumption than willoccur when the long-period position has been reached. Thusthe change in expectation may lead to a gradual crescendoin the level of employment, rising to a peak and then declin-ing to the new long-period level. The same thing may occureven if the new long-period level is the same as the old, if thechange represents a change in the direction of consumptionwhich renders certain existing processes and their equipmentobsolete. Or again, if the new long-period employment is lessthan the old, the level of employment during the transitionmay fall for a time below what the new long-period level isgoing to be. Thus a mere change in expectation is capable ofproducing an oscillation of the same kind of shape as a cyclicalmovement, in the course of working itself out. It was move-ments of this kind which I discussed in my Treatise on Moneyin connection with the building up or the depletion of stocksof working and liquid capital consequent on change.

An uninterrupted process of transition, such as the above,to a new long-period position can be complicated in detail.But the actual course of events is more complicated still. Forthe state of expectation is liable to constant change, a new ex-pectation being superimposed long before the previous changehas fully worked itself out; so that the economic machine isoccupied at any given time with a number of overlapping ac-tivities, the existence of which is due to various past states ofexpectation.

48

II.

II

This leads us to the relevance of this discussion for our presentpurpose. It is evident from the above that the level of employ-ment at any time depends, in a sense, not merely on the exist-ing state of expectation but on the states of expectation whichhave existed over a certain past period. Nevertheless past ex-pectations, which have not yet worked themselves out, areembodied in the to-day’s capital equipment with reference towhich the entrepreneur has to make to-day’s decisions, andonly influence his decisions in so far as they are so embod-ied. It follows, therefore, that, in spite of the above, to-day’semployment can be correctly described as being governed byto-day’s expectations taken in conjunction with to-day’s capi-tal equipment.

Express reference to current long-term expectations can sel-dom be avoided. But it will often be safe to omit expressreference to short-term expectation, in view of the fact that inpractice the process of revision of short-term expectation is agradual and continuous one, carried on largely in the light ofrealised results; so that expected and realised results run intoand overlap one another in their influence. For, although out-put and employment are determined by the producer’s short-term expectations and not by past results, the most recent re-sults usually play a predominant part in determining whatthese expectations are. It would be too complicated to workout the expectations de novo whenever a productive processwas being started; and it would, moreover, be a waste of timesince a large part of the circumstances usually continue sub-stantially unchanged from one day to the next. Accordingly itis sensible for producers to base their expectations on the as-sumption that the most recently realised results will continue,except in so far as there are definite reasons for expecting achange. Thus in practice there is a large overlap between theeffects on employment of the realised sale-proceeds of recentoutput and those of the sale-proceeds expected from currentinput; and producers’ forecasts are more often gradually mod-ified in the light of results than in anticipation of prospective

49


changes.4

Nevertheless, we must not forget that, in the case of durablegoods, the producer’s short-term expectations are based onthe current long-term expectations of the investor; and it isof the nature of long-term expectations that they cannot bechecked at short intervals in the light of realised results. More-over, as we shall see in chapter 12, where we shall considerlong-term expectations in more detail, they are liable to sud-den revision. Thus the factor of current long-term expecta-tions cannot be even approximately eliminated or replaced byrealised results.

4This emphasis on the expectation entertained when the decision to pro-duce is taken, meets, I think, Mr. Hawtrey’s point that input and accumulationof stocks before prices have fallen or disappointment in respect of output is re-flected in a realised loss relatively to expectation. For the accumulation of un-sold stocks (or decline of forward orders) is precisely the kind of event which ismost likely to cause input to differ from what mere statistics of the sale-proceedsof previous output would indicate if they were to be projected without criticisminto the next period.

50

CHAPTER 6

THE DEFINITION OF INCOME, SAVING ANDINVESTMENT

I Income

During any period of time an entrepreneur will have sold fin-ished output to consumers or to other entrepreneurs for a cer-tain sum which we will designate as A. He will also havespent a certain sum, designated byA1, on purchasing finishedoutput from other entrepreneurs. And he will end up with acapital equipment, which term includes both his stocks of un-finished goods or working capital and his stocks of finishedgoods, having a value G.

Some part, however, of A + G − A1 will be attributable,not to the activities of the period in question, but to the cap-ital equipment which he had at the beginning of the period.We must, therefore, in order to arrive at what we mean by theincome of the current period, deduct from A + G − A1 a cer-tain sum, to represent that part of its value which has been (insome sense) contributed by the equipment inherited from theprevious period. The problem of defining income is solved assoon as we have found a satisfactory method for calculatingthis deduction.

There are two possible principles for calculating it, each ofwhich has a certain significance;—one of them in connectionwith production, and the other in connection with consump-tion. Let us consider them in turn.

(i) The actual value G of the capital equipment at the endof the period is the net result of the entrepreneur, on the onehand, having maintained and improved it during the period,both by purchases from other entrepreneurs and by work done

51

6. THE DEFINITION OF INCOME, SAVING AND INVESTMENT

upon it by himself, and, on the other hand, having exhaustedor depreciated it through using it to produce output. If he haddecided not to use it to produce output, there is, neverthe-less, a certain optimum sum which it would have paid him tospend on maintaining and improving it. Let us suppose that,in this event, he would have spent B′ on its maintenance andimprovement, and that, having had this spent on it, it wouldhave been worth G′ at the end of the period. That is to say,G′ − B′ is the maximum net value which might have beenconserved from the previous period, if it had not been usedto produce A. The excess of this potential value of the equip-ment over G − A1 is the measure of what has been sacrificed(one way or another) to produce A. Let us call this quantity,namely

(G′ −B′)− (G−A1) ,

which measures the sacrifice of value involved in the produc-tion of A, the user cost of A. User cost will be written U .1 Theamount paid out by the entrepreneur to the other factors ofproduction in return for their services, which from their pointof view is their income, we will call the factor cost of A. Thesum of the factor cost F and the user cost U we shall call theprime cost of the output A.

We can then define the income2 of the entrepreneur as beingthe excess of the value of his finished output sold during theperiod over his prime cost. The entrepreneur’s income, thatis to say, is taken as being equal to the quantity, depending onhis scale of production, which he endeavours to maximise, i.e.to his gross profit in the ordinary sense of this term;—whichagrees with common sense. Hence, since the income of therest of the community is equal to the entrepreneur’s factorcost, aggregate income is equal to A− U .

Income, thus defined, is a completely unambiguous quan-tity. Moreover, since it is the entrepreneur’s expectation of the

1Some further observations on user cost are given in an appendix to thischapter.

2As distinguished from his net income which we shall define below.

52

I. Income

excess of this quantity over his outgoings to the other factorsof production which he endeavours to maximise when he de-cides how much employment to give to the other factors ofproduction, it is the quantity which is causally significant foremployment.

It is conceivable, of course, thatG−A1 may exceedG′−B′,so that user cost will be negative. For example, this may wellbe the case if we happen to choose our period in such a waythat input has been increasing during the period but with-out there having been time for the increased output to reachthe stage of being finished and sold. It will also be the case,whenever there is positive investment, if we imagine indus-try to be so much integrated that entrepreneurs make most oftheir equipment for themselves. Since, however, user cost isonly negative when the entrepreneur has been increasing hiscapital equipment by his own labour, we can, in an economywhere capital equipment is largely manufactured by differentfirms from those which use it, normally think of user cost asbeing positive. Moreover, it is difficult to conceive of a casewhere marginal user cost associated with an increase in A, i.e.∂U/∂A, will be other than positive.

It may be convenient to mention here, in anticipation of thelatter part of this chapter, that, for the community as a whole,the aggregate consumption (C) of the period is equal to Σ(A−A1), and the aggregate investment (I) is equal to Σ(A1 − U ).Moreover, U is the individual entrepreneur’s disinvestment(and −U his investment) in respect of his own equipment ex-clusive of what he buys from other entrepreneurs. Thus in acompletely integrated system (where A1 = 0) consumptionis equal to A and investment to −U , i.e. to G − (G′ − B′).The slight complication of the above, through the introduc-tion of A1, is simply due to the desirability of providing ina generalised way for the case of a non-integrated system ofproduction.

Furthermore, the effective demand is simply the aggregateincome (or proceeds) which the entrepreneurs expect to re-ceive, inclusive of the incomes which they will hand on tothe other factors of production, from the amount of current

53


employment which they decide to give. The aggregate de-mand function relates various hypothetical quantities of em-ployment to the proceeds which their outputs are expected toyield; and the effective demand is the point on the aggregatedemand function which becomes effective because, taken inconjunction with the conditions of supply, it corresponds tothe level of employment which maximises the entrepreneur’sexpectation of profit.

This set of definitions also has the advantage that we canequate the marginal proceeds (or income) to the marginal fac-tor cost; and thus arrive at the same sort of propositions relat-ing marginal proceeds thus defined to marginal factor costs ashave been stated by those economists who, by ignoring usercost or assuming it to be zero, have equated supply price3 tomarginal factor cost.4

(ii) We turn, next, to the second of the principles referredto above. We have dealt so far with that part of the changein the value of the capital equipment at the end of the pe-riod as compared with its value at the beginning which is dueto the voluntary decisions of the entrepreneur in seeking to

3Supply price is, I think, an incompletely defined term, if the problem ofdefining user cost has been ignored. The matter is further discussed in the ap-pendix to this chapter, where I argue that the exclusion of user cost from supplyprice, whilst sometimes appropriate in the case of aggregate supply price, isinappropriate to the problems of the supply price of a unit of output for anindividual firm.

4For example, let us take Zw = φ(N), or alternatively Z = W · φ(N) asthe aggregate supply function (where W is the wage-unit and W · Zw = Z).Then, since the proceeds of the marginal product is equal to the marginal factor-cost at every point on the aggregate supply curve, we have

∆N = ∆Aw −∆Uw = ∆Zw = ∆φ(N) ,

that is to say φ′(N) = 1; provided that factor cost bears a constant ratio towage-cost, and that the aggregate supply function for each firm (the numberof which is assumed to be constant) is independent of the number of men em-ployed in other industries, so that the terms of the above equation, which holdgood for each individual entrepreneur, can be summed for the entrepreneursas a whole. This means that, if wages are constant and other factor costs area constant proportion of the wages-bill, the aggregate supply function is linearwith a slope given by the reciprocal of the money-wage.

54

I. Income

maximise his profit. But there may, in addition, be an invol-untary loss (or gain) in the value of his capital equipment,occurring for reasons beyond his control and irrespective ofhis current decisions, on account of (e.g.) a change in mar-ket values, wastage by obsolescence or the mere passage oftime, or destruction by catastrophe such as war or earthquake.Now some part of these involuntary losses, whilst they are un-avoidable, are—broadly speaking—not unexpected; such aslosses through the lapse of time irrespective of use, and also‘normal’ obsolescence which, as Professor Pigou expresses it,‘is sufficiently regular to be foreseen, if not in detail, at least inthe large’, including, we may add, those losses to the commu-nity as a whole which are sufficiently regular to be commonlyregarded as ‘insurable risks’. Let us ignore for the moment thefact that the amount of the expected loss depends on when theexpectation is assumed to be framed, and let us call the depre-ciation of the equipment, which is involuntary but not unex-pected, i.e. the excess of the expected depreciation over theuser cost, the supplementary cost, which will be written V . Itis, perhaps, hardly necessary to point out that this definitionis not the same as Marshall’s definition of supplementary cost,though the underlying idea, namely, of dealing with that partof the expected depreciation which does not enter into primecost, is similar.

In reckoning, therefore, the net income and the net profit ofthe entrepreneur it is usual to deduct the estimated amountof the supplementary cost from his income and gross profit asdefined above. For the psychological effect on the entrepreneur,when he is considering what he is free to spend and to save,of the supplementary cost is virtually the same as though itcame off his gross profit. In his capacity as a producer decid-ing whether or not to use the equipment, prime cost and grossprofit, as defined above, are the significant concepts. But in hiscapacity as a consumer the amount of the supplementary costworks on his mind in the same way as if it were a part of theprime cost. Hence we shall not only come nearest to commonusage but will also arrive at a concept which is relevant to theamount of consumption, if, in defining aggregate net income,

55


we deduct the supplementary cost as well as the user cost, sothat aggregate net income is equal to A− U − V .

There remains the change in the value of the equipment,due to unforeseen changes in market values, exceptional ob-solescence or destruction by catastrophe, which is both invol-untary and—in a broad sense—unforeseen. The actual lossunder this head, which we disregard even in reckoning netincome and charge to capital account, may be called the wind-fall loss.

The causal significance of net income lies in the psycholog-ical influence of the magnitude of V on the amount of currentconsumption, since net income is what we suppose the ordi-nary man to reckon his available income to be when he is de-ciding how much to spend on current consumption. This isnot, of course, the only factor of which he takes account whenhe is deciding how much to spend. It makes a considerabledifference, for example, how much windfall gain or loss he ismaking on capital account. But there is a difference betweenthe supplementary cost and a windfall loss in that changesin the former are apt to affect him in just the same way aschanges in his gross profit. It is the excess of the proceeds ofthe current output over the sum of the prime cost and the sup-plementary cost which is relevant to the entrepreneur’s con-sumption; whereas, although the windfall loss (or gain) en-ters into his decisions, it does not enter into them on the samescale—a given windfall loss does not have the same effect asan equal supplementary cost.

We must now recur, however, to the point that the line be-tween supplementary costs and windfall losses, i.e. betweenthose unavoidable losses which we think it proper to debit toincome account and those which it is reasonable to reckon asa windfall loss (or gain) on capital account, is partly a con-ventional or psychological one, depending on what are thecommonly accepted criteria for estimating the former. For nounique principle can be established for the estimation of sup-plementary cost, and its amount will depend on our choiceof an accounting method. The expected value of the supple-mentary cost, when the equipment was originally produced,

56

I. Income

is a definite quantity. But if it is re-estimated subsequently, itsamount over the remainder of the life of the equipment mayhave changed as a result of a change in the meantime in ourexpectations; the windfall capital loss being the discountedvalue of the difference between the former and the revised ex-pectation of the prospective series of U + V . It is a widelyapproved principle of business accounting, endorsed by theInland Revenue authorities, to establish a figure for the sumof the supplementary cost and the user cost when the equip-ment is acquired and to maintain this unaltered during thelife of the equipment, irrespective of subsequent changes inexpectation. In this case the supplementary cost over any pe-riod must be taken as the excess of this predetermined figureover the actual user cost. This has the advantage of ensuringthat the windfall gain or loss shall be zero over the life of theequipment taken as a whole. But it is also reasonable in certaincircumstances to recalculate the allowance for supplementarycost on the basis of current values and expectations at an arbi-trary accounting interval, e.g. annually. Business men in factdiffer as to which course they adopt. It may be convenientto call the initial expectation of supplementary cost when theequipment is first acquired the basic supplementary cost, andthe same quantity recalculated up to date on the basis of cur-rent values and expectations the current supplementary cost.

Thus we cannot get closer to a quantitative definition ofsupplementary cost than that it comprises those deductionsfrom his income which a typical entrepreneur makes beforereckoning what he considers his net income for the purposeof declaring a dividend (in the case of a corporation) or of de-ciding the scale of his current consumption (in the case of anindividual). Since windfall charges on capital account are notgoing to be ruled out of the picture, it is clearly better, in caseof doubt, to assign an item to capital account, and to include insupplementary cost only what rather obviously belongs there.For any overloading of the former can be corrected by allow-ing it more influence on the rate of current consumption thanit would otherwise have had.

It will be seen that our definition of net income comes very

57


close to Marshall’s definition of income, when he decided totake refuge in the practices of the Income Tax Commission-ers and—broadly speaking to regard as income whatever they,with their experience, choose to treat as such. For the fabric oftheir decisions can be regarded as the result of the most care-ful and extensive investigation which is available, to interpretwhat, in practice, it is usual to treat as net income. It also cor-responds to the money value of Professor Pigou’s most recentdefinition of the National Dividend.5

It remains true, however, that net income, being based onan equivocal criterion which different authorities might inter-pret differently, is not perfectly clear-cut. Professor Hayek, forexample, has suggested that an individual owner of capitalgoods might aim at keeping the income he derives from hispossessions constant, so that he would not feel himself freeto spend his income on consumption until he had set asidesufficient to offset any tendency of his investment-income todecline for whatever reason.6 I doubt if such an individualexists; but, obviously, no theoretical objection can be raisedagainst this deduction as providing a possible psychologicalcriterion of net income. But when Professor Hayek infers thatthe concepts of saving and investment suffer from a corre-sponding vagueness, he is only right if he means net savingand net investment. The saving and the investment, whichare relevant to the theory of employment, are clear of this de-fect, and are capable of objective definition, as we have shownabove.

Thus it is a mistake to put all the emphasis on net income,which is only relevant to decisions concerning consumption,and is, moreover, only separated from various other factors af-fecting consumption by a narrow line; and to overlook (as hasbeen usual) the concept of income proper, which is the con-cept relevant to decisions concerning current production andis quite unambiguous.

5Economic Journal, June 1935, p. 235.6“The Maintenance of Capital”, Economica, August 1935, p. 241 et seq.

58

II. Saving and Investment

The above definitions of income and of net income are in-tended to conform as closely as possible to common usage. Itis necessary, therefore, that I should at once remind the readerthat in my Treatise on Money I defined income in a specialsense. The peculiarity in my former definition related to thatpart of aggregate income which accrues to the entrepreneurs,since I took neither the profit (whether gross or net) actuallyrealised from their current operations nor the profit whichthey expected when they decided to undertake their currentoperations, but in some sense (not, as I now think, sufficientlydefined if we allow for the possibility of changes in the scale ofoutput) a normal or equilibrium profit; with the result that onthis definition saving exceeded investment by the amount ofthe excess of normal profit over the actual profit. I am afraidthat this use of terms has caused considerable confusion, es-pecially in the case of the correlative use of saving; since con-clusions (relating, in particular, to the excess of saving over in-vestment), which were only valid if the terms employed wereinterpreted in my special sense, have been frequently adoptedin popular discussion as though the terms were being em-ployed in their more familiar sense. For this reason, and alsobecause I no longer require my former terms to express myideas accurately, I have decided to discard them—with muchregret for the confusion which they have caused.

II Saving and Investment

Amidst the welter of divergent usages of terms, it is agree-able to discover one fixed point. So far as I know, everyoneis agreed that saving means the excess of income over expen-diture on consumption. Thus any doubts about the meaningof saving must arise from doubts about the meaning either ofincome or of consumption. Income we have defined above. Ex-penditure on consumption during any period must mean thevalue of goods sold to consumers during that period, whichthrows us back to the question of what is meant by a consumer-purchaser. Any reasonable definition of the line between consumer-

59


purchasers and investor-purchasers will serve us equally well,provided that it is consistently applied. Such problem as thereis, e.g. whether it is right to regard the purchase of a motor-car as a consumer-purchase and the purchase of a house as aninvestor-purchase, has been frequently discussed and I havenothing material to add to the discussion.

The criterion must obviously correspond to where we drawthe line between the consumer and the entrepreneur. Thuswhen we have definedA1 as the value of what one entrepreneurhas purchased from another, we have implicitly settled thequestion. It follows that expenditure on consumption can beunambiguously defined as Σ(A − A1), where A is the totalsales made during the period and A1 is the total sales madeby one entrepreneur to another. In what follows it will beconvenient, as a rule, to omit and write A for the aggregatesales of all kinds, A1 for the aggregate sales from one en-trepreneur to another and U for the aggregate user costs ofthe entrepreneurs.

Having now defined both income and consumption, the defi-nition of saving, which is the excess of income over consump-tion, naturally follows. Since income is equal to A − U andconsumption is equal toA−A1, it follows that saving is equalto A1 − U . Similarly, we have net saving for the excess of netincome over consumption, equal to A1 − U − V .

Our definition of income also leads at once to the defini-tion of current investment. For we must mean by this the cur-rent addition to the value of the capital equipment which hasresulted from the productive activity of the period. This is,clearly, equal to what we have just defined as saving. For itis that part of the income of the period which has not passedinto consumption. We have seen above that as the result ofthe production of any period entrepreneurs end up with hav-ing sold finished output having a value A and with a capitalequipment which has suffered a deterioration measured by U(or an improvement measured by−U where U is negative) asa result of having produced and parted with A, after allow-ing for purchases A1 from other entrepreneurs. During thesame period finished output having a value A−A1 will have

60


passed into consumption. The excess of A − U over A − A1,namelyA1−U , is the addition to capital equipment as a resultof the productive activities of the period and is, therefore, theinvestment of the period. Similarly A1 − U − V ; which is thenet addition to capital equipment, after allowing for normalimpairment in the value of capital apart from its being usedand apart from windfall changes in the value of the equip-ment chargeable to capital account, is the net investment ofthe period.

Whilst, therefore, the amount of saving is an outcome ofthe collective behaviour of individual consumers and the amountof investment of the collective behaviour of individual en-trepreneurs, these two amounts are necessarily equal, sinceeach of them is equal to the excess of income over consump-tion. Moreover, this conclusion in no way depends on anysubtleties or peculiarities in the definition of income givenabove. Provided it is agreed that income is equal to the valueof current output, that current investment is equal to the valueof that part of current output which is not consumed, and thatsaving is equal to the excess of income over consumption—allof which is conformable both to common sense and to the tra-ditional usage of the great majority of economists—the equal-ity of saving and investment necessarily follows. In short-

Income = value of output = consumption + investment.Saving = income− consumption.

Therefore :

Saving = investment.

Thus any set of definitions which satisfy the above condi-tions leads to the same conclusion. It is only by denying thevalidity of one or other of them that the conclusion can beavoided.

The equivalence between the quantity of saving and thequantity of investment emerges from the bilateral character ofthe transactions between the producer on the one hand and,on the other hand, the consumer or the purchaser of capitalequipment.

61


Income is created by the value in excess of user cost whichthe producer obtains for the output he has sold; but the wholeof this output must obviously have been sold either to a con-sumer or to another entrepreneur; and each entrepreneur’scurrent investment is equal to the excess of the equipmentwhich he has purchased from other entrepreneurs over hisown user cost. Hence, in the aggregate the excess of incomeover consumption, which we call saving, cannot differ fromthe addition to capital equipment which we call investment.And similarly with net saving and net investment. Saving, infact, is a mere residual. The decisions to consume and the de-cisions to invest between them determine incomes. Assumingthat the decisions to invest become effective, they must in do-ing so either curtail consumption or expand income. Thus theact of investment in itself cannot help causing the residual ormargin, which we call saving, to increase by a correspondingamount.

It might be, of course, that individuals were so tête mon-tée in their decisions as to how much they themselves wouldsave and invest respectively, that there would be no point ofprice equilibrium at which transactions could take place. Inthis case our terms would cease to be applicable, since outputwould no longer have a definite market value, prices wouldfind no resting-place between zero and infinity. Experienceshows, however, that this, in fact, is not so; and that thereare habits of psychological response which allow of an equi-librium being reached at which the readiness to buy is equalto the readiness to sell. That there should be such a thing asa market value for output is, at the same time, a necessarycondition for money-income to possess a definite value anda sufficient condition for the aggregate amount which savingindividuals decide to save to be equal to the aggregate amountwhich investing individuals decide to invest.

Clearness of mind on this matter is best reached, perhaps,by thinking in terms of decisions to consume (or to refrainfrom consuming) rather than of decisions to save. A decisionto consume or not to consume truly lies within the power ofthe individual; so does a decision to invest or not to invest.

62


The amounts of aggregate income and of aggregate saving arethe results of the free choices of individuals whether or not toconsume and whether or not to invest; but they are neitherof them capable of assuming an independent value resultingfrom a separate set of decisions taken irrespective of the deci-sions concerning consumption and investment. In accordancewith this principle, the conception of the propensity to con-sume will, in what follows, take the place of the propensity ordisposition to save.

63


6.A Appendix on User Cost

I.

User cost has, I think, an importance for the classical theoryof value which has been overlooked. There is more to be saidabout it than would be relevant or appropriate in this place.But, as a digression, we will examine it somewhat further inthis appendix.

An entrepreneur’s user cost is by definition equal to

A1 + (G′ −B′)−G ,

whereA1 is the amount of our entrepreneur’s purchases fromother entrepreneurs, G the actual value of his capital equip-ment at the end of the period, and G′ the value it might havehad at the end of the period if he had refrained from using itand had spent the optimum sum B′ on its maintenance andimprovement. NowG−(G′−B′), namely the increment in thevalue of the entrepreneurs equipment beyond the net valuewhich he has inherited from the previous period, representsthe entrepreneur’s current investment in his equipment andcan be written I . Thus U , the user cost of his sales-turnoverA, is equal to A1 − I where A1 is what he has bought fromother entrepreneurs and I is what he has currently invested inhis own equipment. A little reflection will show that all thisis no more than common sense. Some part of his outgoingsto other entrepreneurs is balanced by the value of his currentinvestment in his own equipment, and the rest represents thesacrifice which the output he has sold must have cost him overand above the total sum which he has paid out to the factors ofproduction. If the reader tries to express the substance of thisotherwise, he will find that its advantage lies in its avoidanceof insoluble (and unnecessary) accounting problems. There is,I think, no other way of analysing the current proceeds of pro-duction unambiguously. If industry is completely integratedor if the entrepreneur has bought nothing from outside, sothat A1 = 0, the user cost is simply the equivalent of the cur-rent disinvestment involved in using the equipment; but we

64

6.A. Appendix on User Cost

are still left with the advantage that we do not require at anystage of the analysis to allocate the factor cost between thegoods which are sold and the equipment which is retained.Thus we can regard the employment given by a firm, whetherintegrated or individual, as depending on a single consoli-dated decision — a procedure which corresponds to the actualinterlocking character of the production of what is currentlysold with total production.

The concept of user cost enables us, moreover, to give aclearer definition than that usually adopted of the short-periodsupply price of a unit of a firm’s saleable output. For the short-period supply price is the sum of the marginal factor cost andthe marginal user cost.

Now in the modern theory of value it has been a usualpractice to equate the short-period supply price to the marginalfactor cost alone. It is obvious, however, that this is only legiti-mate if marginal user cost is zero or if supply-price is speciallydefined so as to be net of marginal user cost, just as I have de-fined (Chapter 3) “proceeds” and “aggregate supply price” asbeing net of aggregate user cost. But, whereas it may be occa-sionally convenient in dealing with output as a whole to deductuser cost, this procedure deprives our analysis of all reality ifit is habitually (and tacitly) applied to the output of a singleindustry or firm, since it divorces the “supply price” of an arti-cle from any ordinary sense of its “price”; and some confusionmay have resulted from the practice of doing so. It seems tohave been assumed that “supply price” has an obvious mean-ing as applied to a unit of the saleable output of an individualfirm, and the matter has not been deemed to require discus-sion. Yet the treatment both of what is purchased from otherfirms and of the wastage of the firm’s own equipment as aconsequence of producing the marginal output involves thewhole pack of perplexities which attend the definition of in-come. For, even if we assume that the marginal cost of pur-chases from other firms involved in selling an additional unitof output has to be deducted from the sale-proceeds per unitin order to give us what we mean by our firm’s supply price,we still have to allow for the marginal disinvestment in the

65


firm’s own equipment involved in producing the marginaloutput. Even if all production is carried on by a completely in-tegrated firm, it is still illegitimate to suppose that the marginaluser cost is zero, i.e. that the marginal disinvestment in equip-ment due to the production of the marginal output can gener-ally be neglected.

The concepts of user cost and of supplementary cost alsoenable us to establish a clearer relationship between long-periodsupply price and short-period supply price. Long-period costmust obviously include an amount to cover the basic supple-mentary cost as well as the expected prime cost appropriatelyaveraged over the life of the equipment. That is to say, thelong-period cost of the output is equal to the expected sumof the prime cost and the supplementary cost; and, further-more, in order to yield a normal profit, the long-period sup-ply price must exceed the long-period cost thus calculated byan amount determined by the current rate of interest on loansof comparable term and risk, reckoned as a percentage of thecost of the equipment. Or if we prefer to take a standard“pure” rate of interest, we must include in the long-periodcost a third term which we might call the risk-cost to cover theunknown possibilities of the actual yield differing from the ex-pected yield. Thus the long-period supply price is equal to thesum of the prime cost, the supplementary cost, the risk costand the interest cost, into which several components it can beanalysed. The short-period supply price, on the other hand,is equal to the marginal prime cost. The entrepreneur must,therefore, expect, when he buys or constructs his equipment,to cover his supplementary cost, his risk cost and his interestcost out of the excess marginal value of the prime cost over itsaverage value; so that in long-period equilibrium the excess ofthe marginal prime cost over the average prime cost is equalto the sum of the supplementary, risk and interest costs.7

7This way of putting it depends on the convenient assumption that themarginal prime cost curve is continuous throughout its length for changes inoutput. In fact, this assumption is often unrealistic, and there may be one ormore points of discontinuity, especially when we reach an output correspond-

66


The level of output, at which marginal prime cost is ex-actly equal to the sum of the average prime and supplemen-tary costs, has a special importance, because it is the point atwhich the entrepreneur’s trading account breaks even. It cor-responds, that is to say, to the point of zero net profit; whilstwith a smaller output than this he is trading at a net loss. Theextent to which the supplementary cost has to be provided forapart from the prime cost varies very much from one type ofequipment to another. Two extreme cases are the following:

(i) Some part of the maintenance of the equipment mustnecessarily take place pari passu with the act of using it (e.g.oiling the machine). The expense of this (apart from outsidepurchases) is included in the factor cost. If, for physical rea-sons, the exact amount of the whole of the current deprecia-tion has necessarily to be made good in this way, the amountof the user cost (apart from outside purchases) would be equaland opposite to that of the supplementary cost; and in long-period equilibrium the marginal factor cost would exceed theaverage factor cost by an amount equal to the risk and interestcost.

(ii) Some part of the deterioration in the value of the equip-ment only occurs if it is used. The cost of this is charged inuser cost, in so far as it is not made good pari passu with theact of using it. If loss in the value of the equipment could onlyoccur in this way, supplementary cost would be zero.

It may be worth pointing out that an entrepreneur doesnot use his oldest and worst equipment first, merely because

ing to the technical full capacity of the equipment. In this case the marginalanalysis partially breaks down; and the price may exceed the marginal primecost, where the latter is reckoned in respect of a small decrease of output. (Sim-ilarly there may often be a discontinuity in the downward direction. i.e. for areduction in output below a certain point). This is important when we are con-sidering the short-period supply price in long-period equilibrium, since in thatcase any discontinuities, which may exist corresponding to a point of technicalfull capacity, must be supposed to be in operation. Thus the short-period sup-ply price in long-period equilibrium may have to exceed the marginal primecost (reckoned in terms of a small decrease of output).

67


its user cost is low; since its low user cost may be outweighedby its relative inefficiency, i.e. by its high factor cost. Thus anentrepreneur uses by preference that part of his equipment forwhich the user cost plus factor cost is least per unit of output.8

It follows that for any given volume of output of the productin question there is a corresponding user cost,9 but that thistotal user cost does not bear a uniform relation to the marginaluser cost, i.e. to the increment of user cost due to an incrementin the rate of output.

II.

User cost constitutes one of the links between the present andthe future. For in deciding his scale of production an en-trepreneur has to exercise a choice between using up his equip-ment now and preserving it to be used later on. It is theexpected sacrifice of future benefit involved in present usewhich determines the amount of the user cost, and it is themarginal amount of this sacrifice which, together with themarginal factor cost and the expectation of the marginal pro-ceeds, determines his scale of production. How, then, is theuser cost of an act of production calculated by the entrepreneur?

We have defined the user cost as the reduction in the valueof the equipment due to using it as compared with not us-ing it, after allowing for the cost of the maintenance and im-provements which it would be worth while to undertake andfor purchases from other entrepreneurs. It must be arrivedat, therefore, by calculating the discounted value of the ad-ditional prospective yield which would be obtained at some

8Since user cost partly depends on expectations as to the future level ofwages, a reduction in the wage-unit which is expected to be short-lived willcause factor cost and user cost to move in different proportions and so affectwhat equipment is used, and, conceivably, the level of effective demand, sincefactor cost may enter into the determination of effective demand in a differentway from user cost.

9The user cost of the equipment which is first brought into use is not onlyindependent of the total volume of output (see below); i.e. the user cost may beaffected all along the line when the total volume of output is changed.

68


later date if it were not used now. Now this must be at leastequal to the present value of the opportunity to postpone re-placement which will result from laying up the equipment;and it may be more.10

If there is no surplus or redundant stock, so that more unitsof similar equipment are being newly produced every year ei-ther as an addition or in replacement, it is evident that marginaluser cost will be calculable by reference to the amount bywhich the life or efficiency of the equipment will be shortenedif it is used, and the current replacement cost. If, however,there is redundant equipment, then the user cost will also de-pend on the rate of interest and the current (i.e. re-estimated)supplementary cost over the period of time before the redun-dancy is expected to be absorbed through wastage, etc. In thisway interest cost and current supplementary cost enter indi-rectly into the calculation of user cost.

The calculation is exhibited in its simplest and most intel-ligible form when the factor cost is zero. e.g. in the case of aredundant stock of a raw material such as copper, on the lineswhich I have worked out in my Treatise on Money, vol. ii. chap.29. Let us take the prospective values of copper at various fu-ture dates, a series which will be governed by the rate at whichredundancy is being absorbed and gradually approaches theestimated normal cost. The present value or user cost of a tonof surplus copper will then be equal to the greatest of the val-ues obtainable by subtracting from the estimated future valueat any given date of a ton of copper the interest cost and thecurrent supplementary cost on a ton of copper between thatdate and the present.

In the same way the user cost of a ship or factory or ma-chine, when these equipments are in redundant supply, is itsestimated replacement cost discounted at the percentage rate

10It will be more when it is expected that a more than normal yield can beobtained at some later date, which, however, is not expected to last long enoughto justify (or give time for) the production of new equipment. To-day’s usercost is equal to the maximum of the discounted values of the potential expectedyields of all the tomorrows.

69


of its interest and current supplementary costs to the prospec-tive date of absorption of the redundancy.

We have assumed above that the equipment will be re-placed in due course by an identical article. If the equipmentin question will not be renewed identically when it is wornout, then its user cost has to be calculated by taking a pro-portion of the user cost of the new equipment, which will beerected to do its work when it is discarded, given by its com-parative efficiency.

III.

The reader should notice that, where the equipment is not ob-solescent but merely redundant for the time being, the differ-ence between the actual user cost and its normal value (i.e. thevalue when there is no redundant equipment) varies with theinterval of time which is expected to elapse before the redun-dancy is absorbed. Thus if the type of equipment in questionis of all ages and not “bunched” so that a fair proportion isreaching the end of its life annually, the marginal user costwill not decline greatly unless the redundancy is exception-ally excessive. In the case of a general slump, marginal usercost will depend on how long entrepreneurs expect the slumpto last. Thus the rise in the supply price when affairs begin tomend may be partly due to a sharp increase in marginal usercost due to a revision of their expectations.

It has sometimes been argued, contrary to the opinion ofbusiness men, that organised schemes for scrapping redun-dant plant cannot have the desired effect of raising prices un-less they apply to the whole of the redundant plant. But theconcept of user cost shows how the scrapping of (say) half theredundant plant may have the effect or raising prices immedi-ately. For absorption of the redundancy nearer, user cost andconsequently increasesthe current supply price. Thus busi-ness men would seem to have the notion of user cost implic-itly in mind, though they do not formulate it distinctly.

If the supplementary cost is heavy, it follows that the marginaluser cost will be low when there is surplus equipment. More-

70


over, where there is surplus equipment, the marginal factorand user costs are unlikely to be much in excess of their aver-age value. If both these conditions are fulfilled, the existenceof surplus equipment is likely to lead to the entrepreneur’sworking at a net loss, and perhaps at a heavy net loss. Therewill not be a sudden transition from this state of affairs to anormal profit, taking place at the moment when the redun-dancy is absorbed. As the redundancy becomes less, the usercost will gradually increase; and the excess of marginal overaverage factor and user cost may also gradually increase.

IV.

In Marshall’s Principles of Economics (6th ed. p. 360) a part ofuser cost is included in prime cost under the heading of “ex-tra wear-and-tear of plant”. But no guidance is given as tohow this item is to be calculated or as to its importance. Inhis Theory of Unemployment (p. 42) Professor Pigou expresslyassumes that the marginal disinvestment in equipment due tothe marginal output can, in general, be neglected: “The differ-ences in the quantity of wear-and-tear suffered by equipmentand in the costs of non-manual labour employed, that are as-sociated with differences in output, are ignored, as being, ingeneral, of secondary importance”.11 Indeed, the notion thatthe disinvestment in equipment is zero at the margin of pro-duction runs through a good deal of recent economic theory.But the whole problem is brought to an obvious head as soonas it is thought necessary to explain exactly what is meant bythe supply price of an individual firm.

It is true that the cost of maintenance of idle plant mayoften, for the reasons given above, reduce the magnitude ofmarginal user cost, especially in a slump which is expected tolast a long time. Nevertheless a very low user cost at the mar-gin is not a characteristic of the short period as such, but of

11Mr. Hawtrey (Economica, May 1934, p. 145) has called attention to Prof.Pigou’s identification of supply price with marginal labour cost, and has con-tended that Prof. Pigou’s argument is thereby seriously vitiated.

71


particular situations and types of equipment where the costof maintaining idle plant happens to be heavy, and of thosedisequilibria which are characterised by very rapid obsoles-cence or great redundancy, especially if it is coupled with alarge proportion of comparatively new plant.

In the case of raw materials the necessity of allowing foruser cost is obvious;— if a ton of copper is used up to-day itcannot be used to-morrow, and the value which the copperwould have for the purposes of to-morrow must clearly bereckoned as a part of the marginal cost. But the fact has beenoverlooked that copper is only an extreme case of what oc-curs whenever capital equipment is used to produce. The as-sumption that there is a sharp division between raw materialswhere we must allow for the disinvestment due to using themand fixed capital where we can safely neglect it does not cor-respond to the facts; — especially in normal conditions whereequipment is falling due for replacement every year and theuse of equipment brings nearer the date at which replacementis necessary.

It is an advantage of the concepts of user cost and sup-plementary cost that they are as applicable to working andliquid capital as to fixed capital. The essential difference be-tween raw materials and fixed capital lies not in their liabilityto user and supplementary costs, but in the fact that the returnto liquid capital consists of a single term; whereas in the caseof fixed capital, which is durable and used up gradually, thereturn consists of a series of user costs and profits earned insuccessive periods.

72

CHAPTER 7

THE MEANING OF SAVING AND INVESTMENTFURTHER CONSIDERED

I

In the previous chapter saving and investment have been sodefined that they are necessarily equal in amount, being, forthe community as a whole, merely different aspects of thesame thing. Several contemporary writers (including myselfin my Treatise on Money) have, however, given special defini-tions of these terms on which they are not necessarily equal.Others have written on the assumption that they may be un-equal without prefacing their discussion with any definitionsat all. It will be useful, therefore, with a view to relating theforegoing to other discussions of these terms, to classify someof the various uses of them which appear to be current.

So far as I know, everyone agrees in meaning by savingthe excess of income over what is spent on consumption. Itwould certainly be very inconvenient and misleading not tomean this. Nor is there any important difference of opinionas to what is meant by expenditure on consumption. Thusthe differences of usage arise either out of the definition ofinvestment or out of that of income.

II

Let us take investment first. In popular usage it is commonto mean by this the purchase of an asset, old or new, by anindividual or a corporation. Occasionally, the term might berestricted to the purchase of an asset on the Stock Exchange.But we speak just as readily of investing, for example, in a

73

7. THE MEANING OF SAVING AND INVESTMENT FURTHERCONSIDERED

house, or in a machine, or in a stock of finished or unfin-ished goods; and, broadly speaking, new investment, as dis-tinguished from reinvestment, means the purchase of a capi-tal asset of any kind out of income. If we reckon the sale of aninvestment as being negative investment, i.e. disinvestment,my own definition is in accordance with popular usage; sinceexchanges of old investments necessarily cancel out. We have,indeed, to adjust for the creation and discharge of debts (in-cluding changes in the quantity of credit or money); but sincefor the community as a whole the increase or decrease of theaggregate creditor position is always exactly equal to the in-crease or decrease of the aggregate debtor position, this com-plication also cancels out when we are dealing with aggregateinvestment. Thus, assuming that income in the popular sensecorresponds to my net income, aggregate investment in thepopular sense coincides with my definition of net investment,namely the net addition to all kinds of capital equipment, af-ter allowing for those changes in the value of the old capitalequipment which are taken into account in reckoning net in-come.

Investment, thus defined, includes, therefore, the incrementof capital equipment, whether it consists of fixed capital, work-ing capital or liquid capital; and the significant differences ofdefinition (apart from the distinction between investment andnet investment) are due to the exclusion from investment ofone or more of these categories.

Mr Hawtrey, for example, who attaches great importanceto changes in liquid capital, i.e. to undesigned increments(or decrements) in the stock of unsold goods, has suggesteda possible definition of investment from which such changesare excluded. In this case an excess of saving over investmentwould be the same thing as an undesigned increment in thestock of unsold goods, i.e. as an increase of liquid capital.Mr Hawtrey has not convinced me that this is the factor tostress; for it lays all the emphasis on the correction of changeswhich were in the first instance unforeseen, as compared withthose which are, rightly or wrongly, anticipated. Mr Hawtreyregards the daily decisions of entrepreneurs concerning their

74

III.

scale of output as being varied from the scale of the previousday by reference to the changes in their stock of unsold goods.Certainly, in the case of consumption goods, this plays an im-portant part in their decisions. But I see no object in excludingthe play of other factors on their decisions; and I prefer, there-fore, to emphasise the total change of effective demand andnot merely that part of the change in effective demand whichreflects the increase or decrease of unsold stocks in the previ-ous period. Moreover, in the case of fixed capital, the increaseor decrease of unused capacity corresponds to the increase ordecrease in unsold stocks in its effect on decisions to produce;and I do not see how Mr Hawtrey’s method can handle this atleast equally important factor.

It seems probable that capital formation and capital con-sumption, as used by the Austrian school of economists, arenot identical either with investment and disinvestment as de-fined above or with net investment and disinvestment. In par-ticular, capital consumption is said to occur in circumstanceswhere there is quite clearly no net decrease in capital equip-ment as defined above. I have, however, been unable to dis-cover a reference to any passage where the meaning of theseterms is clearly explained. The statement, for example, thatcapital formation occurs when there is a lengthening of theperiod of production does not much advance matters.

III

We come next to the divergences between saving and invest-ment which are due to a special definition of income and henceof the excess of income over consumption. My own use ofterms in my Treatise on Money is an example of this. For, asI have explained on p. 60 above, the definition of income,which I there employed, differed from my present definitionby reckoning as the income of entrepreneurs not their actuallyrealised profits but (in some sense) their ‘normal profit’. Thusby an excess of saving over investment I meant that the scaleof output was such that entrepreneurs were earning a less

75


than normal profit from their ownership of the capital equip-ment; and by an increased excess of saving over investment Imeant that a decline was taking place in the actual profits, sothat they would be under a motive to contract output.

As I now think, the volume of employment (and conse-quently of output and real income) is fixed by the entrepreneurunder the motive of seeking to maximise his present and prospec-tive profits (the allowance for user cost being determined byhis view as to the use of equipment which will maximise hisreturn from it over its whole life); whilst the volume of em-ployment which will maximise his profit depends on the ag-gregate demand function given by his expectations of the sumof the proceeds resulting from consumption and investmentrespectively on various hypotheses. In my Treatise on Moneythe concept of changes in the excess of investment over sav-ing, as there defined, was a way of handling changes in profit,though I did not in that book distinguish clearly between ex-pected and realised results.1 I there argued that change in theexcess of investment over saving was the motive force govern-ing changes in the volume of output. Thus the new argument,though (as I now think) much more accurate and instructive,is essentially a development of the old. Expressed in the lan-guage of my Treatise on Money, it would run: the expectationof an increased excess of investment over saving, given theformer volume of employment and output, will induce en-trepreneurs to increase the volume of employment and out-put. The significance of both my present and my former ar-guments lies in their attempt to show that the volume of em-ployment is determined by the estimates of effective demandmade by the entrepreneurs, an expected increase of invest-ment relatively to saving as defined in my Treatise on Moneybeing a criterion of an increase in effective demand. But theexposition in my Treatise on Money is, of course, very confus-ing and incomplete in the light of the further developmentshere set forth.

1My method there was to regard the current realised profit as determiningthe current expectation of profit.

76

IV.

Mr D. H. Robertson has defined to-day’s income as beingequal to yesterday’s consumption plus investment, so that to-day’s saving, in his sense, is equal to yesterday’s investmentplus the excess of yesterday’s consumption over to-day’s con-sumption. On this definition saving can exceed investment,namely, by the excess of yesterday’s income (in my sense) overto-day’s income. Thus when Mr Robertson says that there isan excess of saving over investment, he means literally thesame thing as I mean when I say that income is falling, andthe excess of saving in his sense is exactly equal to the declineof income in my sense. If it were true that current expecta-tions were always determined by yesterday’s realised results,to-day’s effective demand would be equal to yesterday’s in-come. Thus Mr Robertson’s method might be regarded as analternative. attempt to mine (being, perhaps, a first approxi-mation to it) to make the same distinction, so vital for causalanalysis, that I have tried to make by the contrast between ef-fective demand and income.2

IV

We come next to the much vaguer ideas associated with thephrase ’forced saving’. Is any clear significance discoverablein these? In my Treatise on Money (vol.1, p. 171, footnote [JMK,vol. V, p.154]) I gave some references to earlier uses of thisphrase and suggested that they bore some affinity to the dif-ference between investment and ‘saving’ in the sense in whichI there used the latter term. I am no longer confident that therewas in fact so much affinity as I then supposed. In any case, Ifeel sure that ‘forced saving’ and analogous phrases employedmore recently (e.g. by Professor Hayek or Professor Robbins)have no definite relation to the difference between investmentand ‘saving’ in the sense intended in my Treatise on Money.

2Vide Mr. Robertson’s article “Saving and Hoarding” (Economic Journal,September 1933, p. 399) and the discussion between Mr. Robertson, Mr.Hawtrey and myself (Economic Journal, December 1933, p. 658).

77


For whilst these authors have not explained exactly what theymean by this term, it is clear that ‘forced saving’, in their sense,is a phenomenon which results directly from, and is measuredby, changes in the quantity of money or bank-credit.

It is evident that a change in the volume of output and em-ployment will, indeed, cause a change in income measured inwage-units; that a change in the wage-unit will cause both aredistribution of income between borrowers and lenders anda change in aggregate income measured in money; and thatin either event there will (or may) be a change in the amountsaved. Since, therefore, changes in the quantity of money mayresult, through their effect on the rate of interest, in a changein the volume and distribution of income (as we shall showlater), such changes may involve, indirectly, a change in theamount saved. But such changes in the amounts saved are nomore ’forced savings’ than any other changes in the amountssaved due to a change in circumstances; and there is no meansof distinguishing between one case and another, unless wespecify the amount saved in certain given conditions as ournorm or standard. Moreover, as we shall see, the amount ofthe change in aggregate saving which results from a givenchange in the quantity of money is highly variable and de-pends on many other factors.

Thus ‘forced saving’ has no meaning until we have spec-ified some standard rate of saving. If we select (as might bereasonable) the rate of saving which corresponds to an estab-lished state of full employment, the above definition wouldbecome: ‘Forced saving is the excess of actual saving overwhat would be saved if there were full employment in a posi-tion of long-period equilibrium’. This definition would makegood sense, but a sense in which a forced excess of savingwould be a very rare and a very unstable phenomenon, and aforced deficiency of saving the usual state of affairs.

Professor Hayek’s interesting ‘Note on the Development ofthe Doctrine of Forced Saving’3 shows that this was in fact theoriginal meaning of the term. ‘Forced saving’ or ‘forced fru-

3Quarterly Journal of Economics, Nov. 1932, p. 123.

78

V.

gality’ was, in the first instance, a conception of Bentham’s;and Bentham expressly stated that he had in mind the conse-quences of an increase in the quantity of money (relatively tothe quantity of things vendible for money) in circumstancesof ‘all hands being employed and employed in the most ad-vantageous manner’4. In such circumstances, Bentham pointsout, real income cannot be increased, and, consequently, ad-ditional investment, taking place as a result of the transition,involves forced frugality ‘at the expense of national comfortand national justice’. All the nineteenth-century writers whodealt with this matter had virtually the same idea in mind. Butan attempt to extend this perfectly clear notion to conditionsof less than full employment involves difficulties. It is true, ofcourse (owing to the fact of diminishing returns to an increasein the employment applied to a given capital equipment), thatany increase in employment involves some sacrifice of real in-come to those who were already employed, but an attemptto relate this loss to the increase in investment which may ac-company the increase in employment is not likely to be fruit-ful. At any rate I am not aware of any attempt having beenmade by the modern writers who are interested in ‘forced sav-ing’ to extend the idea to conditions where employment is in-creasing; and they seem, as a rule, to overlook the fact thatthe extension of the Benthamite concept of forced frugality toconditions of less than full employment requires some expla-nation or qualification.

V

The prevalence of the idea that saving and investment, takenin their straightforward sense, can differ from one another, isto be explained, I think, by an optical illusion due to regardingan individual depositor’s relation to his bank as being a one-sided transaction, instead of seeing it as the two-sided trans-action which it actually is. It is supposed that a depositor and

4Loc. cit. p. 125.

79


his bank can somehow contrive between them to perform anoperation by which savings can disappear into the bankingsystem so that they are lost to investment, or, contrariwise,that the banking system can make it possible for investmentto occur, to which no saving corresponds. But no one can savewithout acquiring an asset, whether it be cash or a debt orcapital-goods; and no one can acquire an asset which he didnot previously possess, unless either an asset of equal value isnewly produced or someone else parts with an asset of thatvalue which he previously had. In the first alternative thereis a corresponding new investment: in the second alternativesomeone else must be dis-saving an equal sum. For his loss ofwealth must be due to his consumption exceeding his income,and not to a loss on capital account through a change in thevalue of a capital-asset, since it is not a case of his suffering aloss of value which his asset formerly had; he is duly receivingthe current value of his asset and yet is not retaining this valuein wealth of any form, i.e. he must be spending it on currentconsumption in excess of current income. Moreover, if it is thebanking system which parts with an asset, someone must beparting with cash. It follows that the aggregate saving of thefirst individual and of others taken together must necessarilybe equal to the amount of current new investment.

The notion that the creation of credit by the banking sys-tem allows investment to take place to which ‘no genuine sav-ing’ corresponds can only be the result of isolating one of theconsequences of the increased bank-credit to the exclusion ofthe others. If the grant of a bank credit to an entrepreneuradditional to the credits already existing allows him to makean addition to current investment which would not have oc-curred otherwise, incomes will necessarily be increased andat a rate which will normally exceed the rate of increased in-vestment. Moreover, except in conditions of full employment,there will be an increase of real income as well as of money-income. The public will exercise ‘a free choice’ as to the pro-portion in which they divide their increase of income betweensaving and spending; and it is impossible that the intention ofthe entrepreneur who has borrowed in order to increase in-

80

V.

vestment can become effective (except in substitution for in-vestment by other entrepreneurs which would have occurredotherwise) at a faster rate than the public decide to increasetheir savings. Moreover, the savings which result from thisdecision are just as genuine as any other savings. No onecan be compelled to own the additional money correspond-ing to the new bank-credit, unless he deliberately prefers tohold more money rather than some other form of wealth. Yetemployment, incomes and prices cannot help moving in sucha way that in the new situation someone does choose to holdthe additional money. It is true that an unexpected increase ofinvestment in a particular direction may cause an irregularityin the rate of aggregate saving and investment which wouldnot have occurred if it had been sufficiently foreseen. It is alsotrue that the grant of the bank-credit will set up three tenden-cies—(1) for output to increase, (2) for the marginal productto rise in value in terms of the wage-unit (which in conditionsof decreasing return must necessarily accompany an increaseof output), and (3) for the wage-unit to rise in terms of money(since this is a frequent concomitant of better employment);and these tendencies may affect the distribution of real incomebetween different groups. But these tendencies are character-istic of a state of increasing output as such, and will occur justas much if the increase in output has been initiated otherwisethan by an increase in bank-credit. They can only be avoidedby avoiding any course of action capable of improving em-ployment. Much of the above, however, is anticipating theresult of discussions which have not yet been reached.

Thus the old-fashioned view that saving always involvesinvestment, though incomplete and misleading, is formallysounder than the new-fangled view that there can be savingwithout investment or investment without ‘genuine’ saving.The error lies in proceeding to the plausible inference that,when an individual saves, he will increase aggregate invest-ment by an equal amount. It is true, that, when an individualsaves he increases his own wealth. But the conclusion thathe also increases aggregate wealth fails to allow for the possi-bility that an act of individual saving may react on someone

81


else’s savings and hence on someone else’s wealth.The reconciliation of the identity between saving and in-

vestment with the apparent ‘free-will’ of the individual to savewhat he chooses irrespective of what he or others may be in-vesting, essentially depends on saving being, like spending,a two-sided affair. For although the amount of his own sav-ing is unlikely to have any significant influence on his ownincome, the reactions of the amount of his consumption onthe incomes of others makes it impossible for all individualssimultaneously to save any given sums. Every such attemptto save more by reducing consumption will so affect incomesthat the attempt necessarily defeats itself. It is, of course, justas impossible for the community as a whole to save less thanthe amount of current investment, since the attempt to do sowill necessarily raise incomes to a level at which the sumswhich individuals choose to save add up to a figure exactlyequal to the amount of investment.

The above is closely analogous with the proposition whichharmonises the liberty, which every individual possesses, tochange, whenever he chooses, the amount of money he holds,with the necessity for the total amount of money, which indi-vidual balances add up to, to be exactly equal to the amountof cash which the banking system has created. In this lattercase the equality is brought about by the fact that the amountof money which people choose to hold is not independent oftheir incomes or of the prices of the things (primarily securi-ties), the purchase of which is the natural alternative to hold-ing money. Thus incomes and such prices necessarily changeuntil the aggregate of the amounts of money which individ-uals choose to hold at the new level of incomes and pricesthus brought about has come to equality with the amount ofmoney created by the banking system. This, indeed, is thefundamental proposition of monetary theory.

Both these propositions follow merely from the fact thatthere cannot be a buyer without a seller or a seller without abuyer. Though an individual whose transactions are small inrelation to the market can safely neglect the fact that demandis not a one-sided transaction, it makes nonsense to neglect it

82

V.

when we come to aggregate demand. This is the vital differ-ence between the theory of the economic behaviour of the ag-gregate and the theory of the behaviour of the individual unit,in which we assume that changes in the individual’s own de-mand do not affect his income.

83

Book III

The Propensity to Consume

CHAPTER 8

THE PROPENSITY TO CONSUME: I. THEOBJECTIVE FACTORS

I

We are now in a position to return to our main theme, fromwhich we broke off at the end of Book I in order to deal withcertain general problems of method and definition. The ul-timate object of our analysis is to discover what determinesthe volume of employment. So far we have established thepreliminary conclusion that the volume of employment is de-termined by the point of intersection of the aggregate sup-ply function with the aggregate demand function. The aggre-gate supply function, however, which depends in the main onthe physical conditions of supply, involves few considerationswhich are not already familiar. The form may be unfamiliarbut the underlying factors are not new. We shall return to theaggregate supply function in chapter 20, where we discuss itsinverse under the name of the employment function. But, in themain, it is the part played by the aggregate demand functionwhich has been overlooked; and it is to the aggregate demandfunction that we shall devote Books III and IV.

The aggregate demand function relates any given level ofemployment to the ‘proceeds’ which that level of employmentis expected to realise. The ‘proceeds’ are made up of the sumof two quantities—the sum which will be spent on consump-tion when employment is at the given level, and the sum whichwill be devoted to investment. The factors which govern thesetwo quantities are largely distinct. In this book we shall con-sider the former, namely what factors determine the sum whichwill be spent on consumption when employment is at a givenlevel; and in Book IV we shall proceed to the factors which

87

8. THE PROPENSITY TO CONSUME: I. THE OBJECTIVE FACTORS

determine the sum which will be devoted to investment.Since we are here concerned in determining what sum will

be spent on consumption when employment is at a given level,we should, strictly speaking, consider the function which re-lates the former quantity (C) to the latter (N ). It is more conve-nient, however, to work in terms of a slightly different func-tion, namely, the function which relates the consumption interms of wage-units (Cw) to the income in terms of wage-units(Yw) corresponding to a level of employment N . This suffersfrom the objection that Yw is not a unique function ofN , whichis the same in all circumstances. For the relationship betweenYw and N may depend (though probably in a very minor de-gree) on the precise nature of the employment. That is to say,two different distributions of a given aggregate employmentN between different employments might (owing to the differ-ent shapes of the individual employment functions—a matterto be discussed in Chapter 20 below) lead to different valuesof Yw. In conceivable circumstances a special allowance mighthave to be made for this factor. But in general it is a good ap-proximation to regard Yw as uniquely determined by N . Wewill therefore define what we shall call the propensity to con-sume as the functional relationship between Yw a given levelof income in terms of wage-units, and Cw the expenditure onconsumption out of that level of income, so that

Cw = (Yw) or C = W (Yw) .

The amount that the community spends on consumptionobviously depends (i) partly on the amount of its income, (ii)partly on the other objective attendant circumstances, and (iii)partly on the subjective needs and the psychological propen-sities and habits of the individuals composing it and the prin-ciples on which the income is divided between them (whichmay suffer modification as output is increased). The motivesto spending interact and the attempt to classify them runs thedanger of false division. Nevertheless it will clear our mindsto consider them separately under two broad heads which weshall call the subjective factors and the objective factors. The

88

II.

subjective factors, which we shall consider in more detail inthe next chapter, include those psychological characteristicsof human nature and those social practices and institutionswhich, though not unalterable, are unlikely to undergo a ma-terial change over a short period of time except in abnormalor revolutionary circumstances. In an historical enquiry or incomparing one social system with another of a different type,it is necessary to take account of the manner in which changesin the subjective factors may affect the propensity to consume.But, in general, we shall in what follows take the subjectivefactors as given; and we shall assume that the propensity toconsume depends only on changes in the objective factors.

II

The principal objective factors which influence the propensityto consume appear to be the following:

(1) A change in the wage-unit. Consumption (C) is obviouslymuch more a function of (in some sense) real income than ofmoney-income. In a given state of technique and tastes andof social conditions determining the distribution of income, aman’s real income will rise and fall with the amount of hiscommand over labour-units, i.e. with the amount of his in-come measured in wage-units; though when the aggregatevolume of output changes, his real income will (owing to theoperation of decreasing returns) rise less than in proportionto his income measured in wage-units. As a first approxima-tion, therefore, we can reasonably assume that, if the wage-unit changes, the expenditure on consumption correspondingto a given level of employment will, like prices, change in thesame proportion; though in some circumstances we may haveto make an allowance for the possible reactions on aggregateconsumption of the change in the distribution of a given realincome between entrepreneurs and rentiers resulting from achange in the wage-unit. Apart from this, we have already al-lowed for changes in the wage-unit by defining the propensity

89


to consume in terms of income measured in terms of wage-units.

(2) A change in the difference between income and net income.We have shown above that the amount of consumption de-pends on net income rather than on income, since it is, bydefinition, his net income that a man has primarily in mindwhen he is deciding his scale of consumption. In a given sit-uation there may be a somewhat stable relationship betweenthe two, in the sense that there will be a function uniquely re-lating different levels of income to the corresponding levelsof net income. If, however, this should not be the case, suchpart of any change in income as is not reflected in net incomemust be neglected since it will have no effect on consumption;and, similarly, a change in net income, not reflected in income,must be allowed for. Save in exceptional circumstances, how-ever, I doubt the practical importance of this factor. We willreturn to a fuller discussion of the effect on consumption ofthe difference between income and net income in the fourthsection of this chapter.

(3) Windfall changes in capital-values not allowed for in calcu-lating net income. These are of much more importance in mod-ifying the propensity to consume, since they will bear no sta-ble or regular relationship to the amount of income. The con-sumption of the wealth-owning class may be extremely sus-ceptible to unforeseen changes in the money-value of its wealth.This should be classified amongst the major factors capable ofcausing short-period changes in the propensity to consume.

(4) Changes in the rate of time-discounting, i.e. in the ratio ofexchange between present goods and future goods. This is not quitethe same thing as the rate of interest, since it allows for futurechanges in the purchasing power of money in so far as theseare foreseen. Account has also to be taken of all kinds of risks,such as the prospect of not living to enjoy the future goods orof confiscatory taxation. As an approximation, however, wecan identify this with the rate of interest.

The influence of this factor on the rate of spending out ofa given income is open to a good deal of doubt. For the clas-

90

II.

sical theory of the rate of interest1, which was based on theidea that the rate of interest was the factor which brought thesupply and demand for savings into equilibrium, it was con-venient to suppose that expenditure on consumption is cet.par. negatively sensitive to changes in the rate of interest, sothat any rise in the rate of interest would appreciably diminishconsumption. It has long been recognised, however, that thetotal effect of changes in the rate ofinterest on the readinessto spend on present consumption is complex and uncertain,being dependent on conflicting tendencies, since some of thesubjective motives towards saving will be more easily satis-fied if the rate of interest rises, whilst others will be weakened.Over a long period substantial changes in the rate of interestprobably tend to modify social habits considerably, thus af-fecting the subjective propensity to spend—though in whichdirection it would be hard to say, except in the light of actualexperience. The usual type of short-period fluctuation in therate of interest is not likely, however, to have much direct in-fluence on spending either way.

There are not many people who will alter their way of liv-ing because the rate of interest has fallen from 5 to 4 per cent, iftheir aggregate income is the same as before. Indirectly theremay be more effects, though not all in the same direction. Per-haps the most important influence, operating through changesin the rate of interest, on the readiness to spend out of a givenincome, depends on the effect of these changes on the appre-ciation or depreciation in the price of securities and other as-sets. For if a man is enjoying a windfall increment in the valueof his capital, it is natural that his motives towards currentspending should be strengthened, even though in terms of in-come his capital is worth no more than before; and weakenedif he is suffering capital losses. But this indirect influence wehave allowed for already under (3) above. Apart from this,the main conclusion suggested by experience is, I think, thatthe short-period influence of the rate of interest on individ-ual spending out of a given income is secondary and rela-

1Cf. Chapter 14 below.

91


tively unimportant, except, perhaps, where unusually largechanges are in question. When the rate of interest falls verylow indeed, the increase in the ratio between an annuity pur-chasable for a given sum and the annual interest on that summay, however, provide an important source of negative sav-ing by encouraging the practice of providing for old age bythe purchase of an annuity.

The abnormal situation, where the propensity to consumemay be sharply affected by the development of extreme un-certainty concerning the future and what it may bring forth,should also, perhaps, be classified under this heading.

(5) Changes in fiscal policy. In so far as the inducement to theindividual to save depends on the future return which he ex-pects, it clearly depends not only on the rate of interest but onthe fiscal policy of the government. Income taxes, especiallywhen they discriminate against ’unearned’ income, taxes oncapital-profits, death-duties and the like are as relevant as therate of interest; whilst the range of possible changes in fis-cal policy may be greater, in expectation at least, than for therate of interest itself. If fiscal policy is used as a deliberate in-strument for the more equal distribution of incomes, its effectin increasing the propensity to consume is, of course, all thegreater.2

We must also take account of the effect on the aggregatepropensity to consume of government sinking funds for thedischarge of debt paid for out of ordinary taxation. For theserepresent a species of corporate saving, so that a policy ofsubstantial sinking funds must be regarded in given circum-stances as reducing the propensity to consume. It is for thisreason that a change-over from a policy of government bor-rowing to the opposite policy of providing sinking funds (orvice versa) is capable of causing a severe contraction (or markedexpansion) of effective demand.

2It may be mentioned, in passing, that the effect of fiscal policy on thegrowth of wealth has been the subject of an important misunderstanding which,however, we cannot discuss adequately without the assistance of the theory ofthe rate of interest to be given in Book IV.

92

III.

(6) Changes in expectations of the relation between the presentand the future level of income. We must catalogue this factor forthe sake of formal completeness. But, whilst it may affect con-siderably a particular individual’s propensity to consume, itis likely to average out for the community as a whole. More-over, it is a matter about which there is, as a rule, too muchuncertainty for it to exert much influence.

We are left therefore, with the conclusion that in a given sit-uation the propensity to consume may be considered a fairlystable function, provided that we have eliminated changes inthe wage-unit in terms of money. Windfall changes in capital-values will be capable of changing the propensity to consume,and substantial changes in the rate of interest and in fiscal pol-icy may make some difference; but the other objective factorswhich might affect it, whilst they must not be overlooked, arenot likely to be important in ordinary circumstances.

The fact that, given the general economic situation, the ex-penditure on consumption in terms of the wage-unit dependsin the main, on the volume of output and employment is thejustification for summing up the other factors in the portman-teau function ‘propensity to consume’. For whilst the otherfactors are capable of varying (and this must not be forgotten),the aggregate income measured in terms of the wage-unit is,as a rule, the principal variable upon which the consumption-constituent of the aggregate demand function will depend.

III

Granted, then, that the propensity to consume is a fairly stablefunction so that, as a rule, the amount of aggregate consump-tion mainly depends on the amount of aggregate income (bothmeasured in terms of wage-units), changes in the propensityitself being treated as a secondary influence, what is the nor-mal shape of this function?

The fundamental psychological law, upon which we areentitled to depend with great confidence both a priori fromour knowledge of human nature and from the detailed facts of

93


experience, is that men are disposed, as a rule and on the aver-age, to increase their consumption as their income increases,but not by as much as the increase in their income. That isto say, if Cw is the amount of consumption and Yw is income(both measured in wage-units) ∆Cw has the same sign as ∆Y wbut is smaller in amount, i.e. ∂Cw/∂Yw is positive and lessthan unity.

This is especially the case where we have short periods inview, as in the case of the so-called cyclical fluctuations of em-ployment during which habits, as distinct from more perma-nent psychological propensities, are not given time enough toadapt themselves to changed objective circumstances. For aman’s habitual standard of life usually has the first claim onhis income, and he is apt to save the difference which dis-covers itself between his actual income and the expense ofhis habitual standard; or, if he does adjust his expenditure tochanges in his income, he will over short periods do so im-perfectly. Thus a rising income will often be accompanied byincreased saving, and a falling income by decreased saving,on a greater scale at first than subsequently.

But, apart from short-period changes in the level of in-come, it is also obvious that a higher absolute level of incomewill tend, as a rule, to widen the gap between income andconsumption. For the satisfaction of the immediate primaryneeds of a man and his family is usually a stronger motivethan the motives towards accumulation, which only acquireeffective sway when a margin of comfort has been attained.These reasons will lead, as a rule, to a greater proportion of in-come being saved as real income increases. But whether ornot a greater proportion is saved, we take it as a fundamen-tal psychological rule of any modern community that, whenits real income is increased, it will not increase its consump-tion by an equal absolute amount, so that a greater absoluteamount must be saved, unless a large and unusual change isoccurring at the same time in other factors. As we shall showsubsequently,3 the stability of the economic system essentially

3Cf. p. 251 below.

94

IV.

depends on this rule prevailing in practice. This means that, ifemployment and hence aggregate income increase, not all theadditional employment will be required to satisfy the needsof additional consumption.

On the other hand, a decline in income due to a declinein the level of employment, if it goes far, may even causeconsumption to exceed income not only by some individu-als and institutions using up the financial reserves which theyhave accumulated in better times, but also by the government,which will be liable, willingly or unwillingly, to run into abudgetary deficit or will provide unemployment relief; forexample, out of borrowed money. Thus, when employmentfalls to a low level, aggregate consumption will decline by asmaller amount than that by which real income has declined,by reason both of the habitual behaviour of individuals andalso of the probable policy of governments; which is the ex-planation why a new position of equilibrium can usually bereached within a modest range of fluctuation. Otherwise afall in employment and income, once started, might proceedto extreme lengths.

This simple principle leads, it will be seen, to the same con-clusion as before, namely, that employment can only increasepari passu with an increase in investment; unless, indeed, thereis a change in the propensity to consume. For since consumerswill spend less than the increase in aggregate supply pricewhen employment is increased, the increased employmentwill prove unprofitable unless there is an increase in invest-ment to fill the gap.

IV

We must not underestimate the importance of the fact alreadymentioned above that, whereas employment is a function ofthe expected consumption and the expected investment, con-sumption is, cet. par., a function of net income, i.e. of net in-vestment (net income being equal to consumption plus net in-vestment). In other words, the larger the financial provision

95


which it is thought necessary to make before reckoning netincome, the less favourable to consumption, and therefore toemployment, will a given level of investment prove to be.

When the whole of this financial provision (or supplemen-tary cost) is in fact currently expended in the upkeep of thealready existing capital equipment, this point is not likely tobe overlooked. But when the financial provision exceeds theactual expenditure on current upkeep, the practical results ofthis in its effect on employment are not always appreciated.For the amount of this excess neither directly gives rise tocurrent investment nor is available to pay for consumption.It has, therefore, to be balanced by new investment, the de-mand for which has arisen quite independently of the currentwastage of old equipment against which the financial provi-sion is being made; with the result that the new investmentavailable to provide current income is correspondingly dimin-ished and a more intense demand for new investment is nec-essary to make possible a given level of employment. More-over, much the same considerations apply to the allowancefor wastage included in user cost, in so far as the wastage isnot actually made good.

Take a house which continues to be habitable until it isdemolished or abandoned. If a certain sum is written off itsvalue out of the annual rent paid by the tenants, which thelandlord neither spends on upkeep nor regards as net incomeavailable for consumption, this provision, whether it is a partof U or of V ; constitutes a drag on employment all throughthe life of the house, suddenly made good in a lump when thehouse has to be rebuilt.

In a stationary economy all this might not be worth men-tioning, since in each year the depreciation allowances in re-spect of old houses would be exactly offset by the new housesbuilt in replacement of those reaching the end of their lives inthat year. But such factors may be serious in a non-static econ-omy, especially during a period which immediately succeedsa lively burst of investment in long-lived capital. For in suchcircumstances a very large proportion of the new items of in-vestment may be absorbed by the larger financial provisions

96

IV.

made by entrepreneurs in respect of existing capital equip-ment, upon the repairs and renewal of which, though it iswearing out with time, the date has not yet arrived for spend-ing anything approaching the full financial provision which isbeing set aside; with the result that incomes cannot rise abovea level which is low enough to correspond with a low aggre-gate of net investment. Thus sinking funds, etc., are apt towithdraw spending power from the consumer long before thedemand for expenditure on replacements (which such provi-sions are anticipating) comes into play; i.e. they diminish thecurrent effective demand and only increase it in the year inwhich the replacement is actually made. If the effect of this isaggravated by ‘financial prudence’, i.e. by its being thoughtadvisable to ‘write off’ the initial cost more rapidly than theequipment actually wears out, the cumulative result may bevery serious indeed.

In the United States, for example, by 1929 the rapid capitalexpansion of the previous five years had led cumulatively tothe setting up of sinking funds and depreciation allowances,in respect of plant which did not need replacement, on sohuge a scale that an enormous volume of entirely new in-vestment was required merely to absorb these financial pro-visions; and it became almost hopeless to find still more newinvestment on a sufficient scale to provide for such new sav-ing as a wealthy community in full employment would be dis-posed to set aside. This factor alone was probably sufficient tocause a slump. And, furthermore, since ‘financial prudence’of this kind continued to be exercised through the slump bythose great corporations which were still in a position to af-ford it, it offered a serious obstacle to early recovery.

Or again, in Great Britain at the present time (1935) the sub-stantial amount of house-building and of other new invest-ments since the war has led to an amount of sinking fundsbeing set up much in excess of any present requirements forexpenditure on repairs and renewals, a tendency which hasbeen accentuated, where the investment has been made by lo-cal authorities and public boards, by the principles of ’sound’finance which often require sinking funds sufficient to write

97


off the initial cost some time before replacement will actuallyfall due; with the result that even if private individuals wereready to spend the whole of their net incomes it would be asevere task to restore full employment in the face of this heavyvolume of statutory provision by public and semi-public au-thorities, entirely dissociated from any corresponding new in-vestment. The sinking funds of local authorities now stand, Ithink,4 at an annual figure of more than half the amount whichthese authorities are spending on the whole of their new de-velopments.5 Yet it is not certain that the Ministry of Healthare aware, when they insist on stiff sinking funds by local au-thorities, how much they may be aggravating the problem ofunemployment. In the case of advances by building societiesto help an individual to build his own house, the desire to beclear of debt more rapidly than the house actually deterioratesmay stimulate the house-owner to save more than he other-wise would;—though this factor should be classified, perhaps,as diminishing the propensity to consume directly rather thanthrough its effect on net income. In actual figures, repay-ments of mortgages advanced by building societies, whichamounted to £24,000,000 in 1925, had risen to £68,000,000 by1933, as compared with new advances of £103,000,000; andto-day the repayments are probably still higher.

That it is investment, rather than net investment, whichemerges from the statistics of output, is brought out forciblyand naturally in Mr Colin Clark’s National Income, 1924−1931.He also shows what a large proportion depreciation, etc., nor-mally bears to the value of investment. For example, he esti-mates that in Great Britain, over the years 1928−1931,6 the in-vestment and the net investment were as follows, though his

4The actual figures are deemed of so little interest that they are only pub-lished two years or more in arrears.

5In the year ending March 31, 1930, local authorities spent £87,000,000 oncapital account, of which £37,000,000 was provided by sinking funds, etc., inrespect of previous capital expenditure; in the year ending March 31, 1933, thefigures were £81,000,000 and £46,000,000.

6Op. cit. pp. 117 and 138.

98

IV.

gross investment is probably somewhat greater than my in-vestment, inasmuch as it may include a part of user cost, andit is not clear how closely his ’net investment’ corresponds tomy definition of this term:

(£ million)1928 1929 1930 1931

Gross Investment-Output 791 731 620 482‘Value of physical wastingof old capital’

433 435 437 439

Net Investment 358 296 183 43

Mr Kuznets has arrived at much the same conclusion incompiling the statistics of the Gross Capital Formation (as hecalls what I call investment) in the United States, 1919−1933.The physical fact, to which the statistics of output correspond,is inevitably the gross, and not the net, investment. Mr Kuznetshas also discovered the difficulties in passing from gross in-vestment to net investment. ‘The difficulty’, he writes, ‘ofpassing from gross to net capital formation, that is, the dif-ficulty of correcting for the consumption of existing durablecommodities, is not only in the lack of data. The very con-cept of annual consumption of commodities that last over anumber of years suffers from ambiguity’. He falls back, there-fore, ‘on the assumption that the allowance for depreciationand depletion on the books of business firms describes cor-rectly the volume of consumption of already existing, finisheddurable goods used by business firms’ On the other hand, heattempts no deduction at all in respect of houses and otherdurable commodities in the hands of individuals. His veryinteresting results for the United States can be summarised asfollows:

Several facts emerge with prominence from this table. Netcapital formation was very steady over the quinquennium 1925–1929, with only a 10 percent increase in the latter part of theupward movement. The deduction for entrepreneurs’ repairs,maintenance, depreciation and depletion remained at a highfigure even at the bottom of the slump. But Mr Kuznets’ method

99


(Millions

ofdollars)

19251926

19271928

19291930

19311932

1933G

rosscapital

formation

(afterallow

ingfor

netchange

inbusiness

inventories)

30,70633,571

31,15733,934

34,49127,538

18,7217,780

14,879

Entrepreneurs’servicing,repairs,

main-

tenance,de-

preciationand

depletion

7,6858,288

8,2238,481

9,0108,502

7,6236,543

8,204

Net

capitalfor-

mation

(onM

rK

uznets’defini-tion)

23,02125,283

22,93425,453

25,48119,036

11,0981,237

6,675

100

IV.

must surely lead to too low an estimate of the annual increasein depreciation, etc.; for he puts the latter at less than 1 1

2per

cent per annum of the new net capital formation. Above all,net capital formation suffered an appalling collapse after 1929,falling in 1932 to a figure no less than 95 per cent below the av-erage of the quinquennium 1925−1929.

The above is, to some extent, a digression. But it is impor-tant to emphasise the magnitude of the deduction which hasto be made from the income of a society, which already pos-sesses a large stock of capital, before we arrive at the net in-come which is ordinarily available for consumption. For if weoverlook this, we may underestimate the heavy drag on thepropensity to consume which exists even in conditions wherethe public is ready to consume a very large proportion of itsnet income.

Consumption—to repeat the obvious—is the sole end andobject of all economic activity. Opportunities for employmentare necessarily limited by the extent of aggregate demand.Aggregate demand can be derived only from present consump-tion or from present provision for future consumption. Theconsumption for which we can profitably provide in advancecannot be pushed indefinitely into the future. We cannot, asa community, provide for future consumption by financial ex-pedients but only by current physical output. In so far as oursocial and business organisation separates financial provisionfor the future from physical provision for the future so thatefforts to secure the former do not necessarily carry the latterwith them, financial prudence will be liable to diminish aggre-gate demand and thus impair well-being, as there are manyexamples to testify. The grcater, moreover, the consumptionfor which we have provided in advance, the more difficult itis to find something further to provide for in advance, and thegreater our dependence on present consumption as a sourceof demand. Yet the larger our incomes, the greater, unfortu-nately, is the margin between our incomes and our consump-tion. So, failing some novel expedient, there is, as we shallsee, no answer to the riddle, except that there must be suffi-

101


cient unemployment to keep us so poor that our consumptionfalls short of our income by no more than the equivalent ofthe physical provision for future consumption which it paysto produce to-day.

Or look at the matter thus. Consumption is satisfied partlyby objects produced currently and partly by objects producedpreviously, i.e. by disinvestment. To the extent that consump-tion is satisfied by the latter, there is a contraction of currentdemand, since to that extent a part of current expenditurefails to find its way back as a part of net income. Contrari-wise whenever an object is produced within the period with aview to satisfying consumption subsequently, an expansion ofcurrent demand is set up. Now all capital-investment is des-tined to result, sooner or later, in capital-disinvestment. Thusthe problem of providing that new capital-investment shall al-ways outrun capital-disinvestment sufficiently to fill the gapbetween net income and consumption, presents a problemwhich is increasingly difficult as capital increases. New capital-investment can only take place in excess of current capital-disinvestment if future expenditure on consumption is expectedto increase. Each time we secure to-day’s equilibrium by in-creased investment we are aggravating the difficulty of secur-ing equilibrium to-morrow. A diminished propensity to con-sume to-day can only be accommodated to the public advan-tage if an increased propensity to consume is expected to ex-ist some day. We are reminded of ‘The Fable of the Bees’—thegay of tomorrow are absolutely indispensable to provide a rai-son d’être for the grave of to-day. It is a curious thing, worthyof mention, that the popular mind seems only to be aware ofthis ultimate perplexity where public investment is concerned,as in the case of road-building and house-building and thelike. It is commonly urged as an objection to schemes for rais-ing employment by investment under the auspices of publicauthority that it is laying up trouble for the future. ‘What willyou do,’ it is asked, ‘when you have built all the houses androads and town halls and electric grids and water suppliesand so forth which the stationary population of the future canbe expected to require?’ But it is not so easily understood that

102

IV.

the same difficulty applies to private investment and to in-dustrial expansion; particularly to the latter, since it is mucheasier to see an early satiation of the demand for new factoriesand plant which absorb individually but little money, than ofthe demand for dwelling-houses.

The obstacle to a clear understanding is, in these examples,much the same as in many academic discussions of capital,namely, an inadequate appreciation of the fact that capital isnot a self-subsistent entity existing apart from consumption.On the contrary, every weakening in the propensity to con-sume regarded as a permanent habit must weaken the de-mand for capital as well as the demand for consumption.

103

CHAPTER 9

THE PROPENSITY TO CONSUME: II. THESUBJECTIVE FACTORS

I

There remains the second category of factors which affect theamount of consumption out of a given income—namely, thosesubjective and social incentives which determine how much isspent, given the aggregate of income in terms of wage-unitsand given the relevant objective factors which we have al-ready discussed. Since, however, the analysis of these factorsraises no point of novelty, it may be sufficient if we give a cat-alogue of the more important, without enlarging on them atany length.

There are, in general, eight main motives or objects of asubjective character which lead individuals to refrain fromspending out of their incomes:

(i) To build up a reserve against unforeseen contingencies;

(ii) To provide for an anticipated future relation betweenthe income and the needs of the individual or his family dif-ferent from that which exists in the present, as, for example,in relation to old age, family education, or the maintenance ofdependents;

(iii) To enjoy interest and appreciation, i.e. because a largerreal consumption at a later date is preferred to a smaller im-mediate consumption;

(iv) To enjoy a gradually increasing expenditure, since it grat-ifies a common instinct to look forward to a gradually improv-ing standard of life rather than the contrary, even though thecapacity for enjoyment may be diminishing;

105

9. THE PROPENSITY TO CONSUME: II. THE SUBJECTIVE FACTORS

(v) To enjoy a sense of independence and the power to dothings, though without a clear idea or definite intention ofspecific action;

(vi) To secure a masse de manoeuvre to carry out speculativeor business projects;(vii) To bequeath a fortune;(viii) To satisfy pure miserliness, i.e. unreasonable but insis-tent inhibitions against acts of expenditure as such.

These eight motives might be called the motives of Pre-caution, Foresight, Calculation, Improvement, Independence,Enterprise, Pride and Avarice; and we could also draw up acorresponding list of motives to consumption such as Enjoy-ment, Shortsightedness, Generosity, Miscalculation, Ostenta-tion and Extravagance.

Apart from the savings accumulated by individuals, thereis also the large amount of income, varying perhaps from one-third to two-thirds of the total accumulation in a modern in-dustrial community such as Great Britain or the United States,which is withheld by central and local government, by institu-tions and by business corporations—for motives largely anal-ogous to, but not identical with, those actuating individuals,and mainly the four following:

(i) The motive of enterprise—to secure resources to carryout further capital investment without incurring debt or rais-ing further capital on the market;

(ii) The motive of liquidity—to secure liquid resources tomeet emergencies, difficulties and depressions;

(iii) The motive of improvement—to secure a gradually in-creasing income, which, incidentally, will protect the manage-ment from criticism, since increasing income due to accumu-lation is seldom distinguished from increasing income due toefficiency;

(iv) The motive of financial prudence and the anxiety to be’on the right side’ by making a financial provision in excess

106

II.

of user and supplementary cost, so as to discharge debt andwrite off the cost of assets ahead of; rather than behind, theactual rate of wastage and obsolescence, the strength of thismotive mainly depending on the quantity and character of thecapital equipment and the rate of technical change.

Corresponding to these motives which favour the with-holding of a part of income from consumption, there are alsooperative at times motives which lead to an excess of con-sumption over income. Several of the motives towards pos-itive saving catalogued above as affecting individuals havetheir intended counterpart in negative saving at a later date,as, for example, with saving to provide for family needs orold age. Unemployment relief financed by borrowing is bestregarded as negative saving.

Now the strength of all these motives will vary enormouslyaccording to the institutions and organisation of the economicsociety which we presume, according to habits formed by race,education, convention, religion and current morals, accordingto present hopes and past experience, according to the scaleand technique of capital equipment, and according to the pre-vailing distribution of wealth and the established standardsof life. In the argument of this book, however, we shall notconcern ourselves, except in occasional digressions, with theresults of far-reaching social changes or with the slow effectsof secular progress. We shall, that is to say, take as given themain background of subjective motives to saving and to con-sumption respectively. In so far as the distribution of wealth isdetermined by the more or less permanent social structure ofthe community, this also can be reckoned a factor, subject onlyto slow change and over a long period, which we can take asgiven in our present context.

II

Since, therefore, the main background of subjective and socialincentives changes slowly, whilst the short-period influence of

107

9. THE PROPENSITY TO CONSUME: II. THE SUBJECTIVE FACTORS

changes in the rate of interest and the other objective factors isoften of secondary importance, we are left with the conclusionthat short-period changes in consumption largely depend onchanges in the rate at which income (measured in wage-units)is being earned and not on changes in the propensity to con-sume out of a given income.

We must, however, guard against a misunderstanding. Theabove means that the influence of moderate changes in therate of interest on the propensity to consume is usually small. Itdoes not mean that changes in the rate of interest have only asmall influence on the amounts actually saved and consumed.Quite the contrary. The influence of changes in the rate of in-terest on the amount actually saved is of paramount impor-tance, but is in the opposite direction to that usually supposed.For even if the attraction of the larger future income to beearned from a higher rate of interest has the effect of diminish-ing the propensity to consume, nevertheless we can be certainthat a rise in the rate of interest will have the effect of reducingthe amount actually saved. For aggregate saving is governedby aggregate investment; a rise in the rate of interest (unlessit is offset by a corresponding change in the demand-schedulefor investment) will diminish investment; hence a rise in therate of interest must have the effect of reducing incomes to alevel at which saving is decreased in the same measure as in-vestment. Since incomes will decrease by a greater absoluteamount than investment, it is, indeed, true that, when the rateof interest rises, the rate of consumption will decrease. Butthis does not mean that there will be a wider margin for sav-ing. On the contrary, saving and spending will both decrease.

Thus, even if it is the case that a rise in the rate of interestwould cause the community to save more out of a given income,we can be quite sure that a rise in the rate of interest (assumingno favourable change in the demand-schedule for investment)will decrease the actual aggregate of savings. The same lineof argument can even tell us by how much a rise in the rateof interest will, cet. par., decrease incomes. For incomes willhave to fall (or be redistributed) by just that amount which isrequired, with the existing propensity to consume to decrease

108

II.

savings by the same amount by which the rise in the rate ofinterest will, with the existing marginal efficiency of capital,decrease investment. A detailed examination of this aspectwill occupy our next chapter.

The rise in the rate of interest might induce us to save more,if our incomes were unchanged. But if the higher rate of in-terest retards investment, our incomes will not, and cannot, beunchanged. They must necessarily fall, until the declining ca-pacity to save has sufficiently offset the stimulus to save givenby the higher rate of interest. The more virtuous we are, themore determinedly thrifty, the more obstinately orthodox inour national and personal finance, the more our incomes willhave to fall when interest rises relatively to the marginal effi-ciency of capital. Obstinacy can bring only a penalty and noreward. For the result is inevitable.

Thus, after all, the actual rates of aggregate saving andspending do not depend on Precaution, Foresight, Calcula-tion, Improvement, Independence, Enterprise, Pride or Avarice.Virtue and vice play no part. It all depends on how far the rateof interest is favourable to investment, after taking account ofthe marginal efficiency of capital. No, this is an overstatement.If the rate of interest were so governed as to maintain contin-uous full employment, virtue would resume her sway;—therate of capital accumulation would depend on the weaknessof the propensity to consume. Thus, once again, the tributethat classical economists pay to her is due to their concealedassumption that the rate of interest always is so governed.

109

CHAPTER 10

THE MARGINAL PROPENSITY TO CONSUME ANDTHE MULTIPLIER

We established in chapter 8 that employment can only increasepari passu with investment unless there is a change in the propen-sity to consume. We can now carry this line of thought a stagefurther. For in given circumstances a definite ratio, to be calledthe multiplier, can be established between income and invest-ment and, subject to certain simplifications, between the totalemployment and the employment directly employed on in-vestment (which we shall call the primary employment). Thisfurther step is an integral part of our theory of employment,since it establishes a precise relationship, given the propen-sity to consume, between aggregate employment and incomeand the rate of investment. The conception of the multiplierwas first introduced into economic theory by Mr R. F. Kahnin his article on ‘The Relation of Home Investment to Un-employment’ (Economic Journal, June 1931). His argument inthis article depended on the fundamental notion that, if thepropensity to consume in various hypothetical circumstancesis (together with certain other conditions) taken as given andwe conceive the monetary or other public authority to takesteps to stimulate or to retard investment, the change in theamount of employment will be a function of the net changein the amount of investment; and it aimed at laying downgeneral principles by which to estimate the actual quantita-tive relationship between an increment of net investment andthe increment of aggregate employment which will be asso-ciated with it. Before coming to the multiplier, however, itwill be convenient to introduce the conception of the marginalpropensity to consume.

111

10. THE MARGINAL PROPENSITY TO CONSUME AND THEMULTIPLIER

I

The fluctuations in real income under consideration in thisbook are those which result from applying different quanti-ties of employment (i.e. of labour-units) to a given capitalequipment, so that real income increases and decreases withthe number of labour-units employed. If, as we assume ingeneral, there is a decreasing return at the margin as the num-ber of labour-units employed on the given capital equipmentis increased, income measured in terms of wage-units will in-crease more than in proportion to the amount of employment,which, in turn, will increase more than in proportion to theamount of real income measured (if that is possible) in termsof product. Real income measured in terms of product and in-come measured in terms of wage-units will, however, increaseand decrease together (in the short period when capital equip-ment is virtually unchanged). Since, therefore, real income, interms of product, may be incapable of precise numerical mea-surement, it is often convenient to regard income in terms ofwage-units (Yw) as an adequate working index of changes inreal income. In certain contexts we must not overlook the factthat, in general, Yw increases and decreases in a greater pro-portion than real income; but in other contexts the fact thatthey always increase and decrease together renders them vir-tually interchangeable.

Our normal psychological law that, when the real incomeof the community increases or decreases, its consumption willincrease or decrease but not so fast, can, therefore, be trans-lated—not, indeed, with absolute accuracy but subject to qual-ifications which are obvious and can easily be stated in a for-mally complete fashion into the propositions that ∆Cw and∆Yw have the same sign, but ∆Yw > ∆Cw, where Cw is theconsumption in terms of wage-units. This is merely a repeti-tion of the proposition already established in Chapter 3 above.Let us define, then, ∂Cw/∂Yw as the marginal propensity to con-sume.

This quantity is of considerable importance, because it tellsus how the next increment of output will have to be divided

112

II.

between consumption and investment. For ∆Yw = ∆Cw +∆Iw, where Cw and Iw are the increments of consumptionand investment; so that we can write ∆Yw = k∆Iw, where1− 1/k is equal to the marginal propensity to consume.

Let us call k the investment multiplier. It tells us that, whenthere is an increment of aggregate investment, income will in-crease by an amount which is k times the increment of invest-ment.

II

Mr Kahn’s multiplier is a little different from this, being whatwe may call the employment multiplier designated by k′, sinceit measures the ratio of the increment of total employmentwhich is associated with a given increment of primary em-ployment in the investment industries. That is to say, if the in-crement of investment ∆Iw leads to an increment of primaryemployment ∆N2 in the investment industries, the incrementof total employment ∆N = k′∆N2.

There is no reason in general to suppose that k = k′. Forthere is no necessary presumption that the shapes of the rel-evant portions of the aggregate supply functions for differ-ent types of industry are such that the ratio of the incrementof employment in the one set of industries to the incrementof demand which has stimulated it will be the same as inthe other set of industries.1 It is easy, indeed, to conceive

1More precisely, if ee and e′e are the elasticities of employment in industryas a whole and in the investment industries respectively; and if N and N2 arethe numbers of men employed in industry as a whole and in the investmentindustries, we have

∆Yw =Yw

(ee ·N) ·∆Nand

∆Iw =Iw

(e′e ·N2) ·∆N2,

so that

∆N =ee

e′e·Iw

N2·N

Yw

· k ·∆N2 ,

113


of cases, as, for example, where the marginal propensity toconsume is widely different from the average propensity, inwhich there would be a presumption in favour of some in-equality between ∆Yw/∆N and ∆Iw/∆N2, since there wouldbe very divergent proportionate changes in the demands forconsumption-goods and investment-goods respectively. If wewish to take account of such possible differences in the shapesof the relevant portions of the aggregate supply functions forthe two groups of industries respectively, there is no difficultyin rewriting the following argument in the more generalisedform. But to elucidate the ideas involved, it will be convenientto deal with the simplified case where k = k′.

It follows, therefore, that, if the consumption psychologyof the community is such that they will choose to consume,e.g. nine-tenths of an increment of income,2 then the multi-plier k is 10; and the total employment caused by (e.g.) in-creased public works will be ten times the primary employ-ment provided by the public works themselves, assuming noreduction of investment in other directions. Only in the eventof the community maintaining their consumption unchangedin spite of the increase in employment and hence in real in-come, will the increase of employment be restricted to theprimary employment provided by the public works. If, onthe other hand, they seek to consume the whole of any incre-ment of income, there will be no point of stability and priceswill rise without limit. With normal psychological suppo-sitions, an increase in employment will only be associatedwith a decline in consumption if there is at the same time achange in the propensity to consume—as the result, for in-

i.e.,

k′

=Iw

e′eN2·eeN

Yw

· k .

If however, there is no reason to expect any material relevant difference inthe shapes of the aggregate supply functions for industry as a whole and for theinvestment industries respectively, so that Iw

e′e·N2= Yw

ee·N , then it follows that∆Yw∆N = ∆Iw

∆N2and, therefore, that k = k′.

2Our quantities are measured throughout in terms of wage-units.

114

II.

stance, of propaganda in time of war in favour of restrictingindividual consumption; and it is only in this event that theincreased employment in investment will be associated withan unfavourable repercussion on employment in the indus-tries producing for consumption.

This only sums up in a formula what should by now beobvious to the reader on general grounds. An increment ofinvestment in terms of wage-units cannot occur unless thepublic are prepared to increase their savings in terms of wage-units. Ordinarily speaking, the public will not do this unlesstheir aggregate income in terms of wage-units is increasing.Thus their effort to consume a part of their increased incomeswill stimulate output until the new level (and distribution) ofincomes provides a margin of saving sufficient to correspondto the increased investment. The multiplier tells us by howmuch their employment has to be increased to yield an in-crease in real income sufficient to induce them to do the nec-essary extra saving, and is a function of their psychologicalpropensities.3 If saving is the pill and consumption is the jam,the extra jam has to be proportioned to the size of the addi-tional pill. Unless the psychological propensities of the publicare different from what we are supposing, we have here es-tablished the law that increased employment for investmentmust necessarily stimulate the industries producing for con-sumption and thus lead to a total increase of employmentwhich is a multiple of the primary employment required bythe investment itself.

It follows from the above that, if the marginal propensityto consume is not far short of unity, small fluctuations in in-vestment will lead to wide fluctuations in employment; but,at the same time, a comparatively small increment of invest-ment will lead to full employment. If, on the other hand,the marginal propensity to consume is not much above zero,small fluctuations in investment will lead to correspondingly

3Though in the more generalised case it is also a function of the physicalconditions of production in the investment and consumption industries respec-tively.

115


small fluctuations in employment; but, at the same time, itmay require a large increment of investment to produce fullemployment. In the former case involuntary unemploymentwould be an easily remedied malady, though liable to be trou-blesome if it is allowed to develop. In the latter case, em-ployment may be less variable but liable to settle down at alow level and to prove recalcitrant to any but the most dras-tic remedies. In actual fact the marginal propensity to con-sume seems to lie somewhere between these two extremes,though much nearer to unity than to zero; with the result thatwe have, in a sense, the worst of both worlds, fluctuationsin employment being considerable and, at the same time, theincrement in investment required to produce full employmentbeing too great to be easily handled. Unfortunately the fluctu-ations have been sufficient to prevent the nature of the maladyfrom being obvious, whilst its severity is such that it cannot beremedied unless its nature is understood.

When full employment is reached, any attempt to increaseinvestment still further will set up a tendency in money-pricesto rise without limit, irrespective of the marginal propensity toconsume; i.e. we shall have reached a state of true inflation.4

Up to this point, however, rising prices will be associated withan increasing aggregate real income.

III

We have been dealing so far with a net increment of invest-ment. If, therefore, we wish to apply the above without quali-fication to the effect of (e.g.) increased public works, we haveto assume that there is no offset through decreased investmentin other directions,—and also, of course, no associated changein the propensity of the community to consume. Mr Kahn wasmainly concerned in the article referred to above in consider-ing what offsets we ought to take into account as likely to beimportant, and in suggesting quantitative estimates. For in an

4Cf. Chapter 21, p. 303, below.

116

III.

actual case there are several factors besides some specific in-crease of investment of a given kind which enter into the finalresult. If, for example, a government employs 100,000 addi-tional men on public works, and if the multiplier (as definedabove) is 4, it is not safe to assume that aggregate employmentwill increase by 400,000. For the new policy may have adversereactions on investment in other directions.

It would seem (following Mr Kahn) that the following arelikely in a modern community to be the factors which it ismost important not to overlook (though the first two will notbe fully intelligible until after Book IV has been reached):

(i) The method of financing the policy and the increasedworking cash, required by the increased employment and theassociated rise of prices, may have the effect of increasing therate of interest and so retarding investment in other direc-tions, unless the monetary authority takes steps to the con-trary; whilst, at the same time, the increased cost of capitalgoods will reduce their marginal efficiency to the private in-vestor, and this will require an actual fall in the rate of interestto offset it.

(ii) With the confused psychology which often prevails, thegovernment programme may, through its effect on ’confidence’,increase liquidity-preference or diminish the marginal efficiencyof capital, which, again, may retard other investment unlessmeasures are taken to offset it.

(iii) In an open system with foreign-trade relations, somepart of the multiplier of the increased investment will accrueto the benefit of employment in foreign countries, since a pro-portion of the increased consumption will diminish our owncountry’s favourable foreign balance; so that, if we consideronly the effect on domestic employment as distinct from worldemployment, we must diminish the full figure of the multi-plier. On the other hand our own country may recover a por-tion of this leakage through favourable repercussions due tothe action of the multiplier in the foreign country in increasingits economic activity.

117


Furthermore, if we are considering changes of a substan-tial amount, we have to allow for a progressive change in themarginal propensity to consume, as the position of the marginis gradually shifted; and hence in the multiplier. The marginalpropensity to consume is not constant for all levels of employ-ment, and it is probable that there will be, as a rule, a tendencyfor it to diminish as employment increases; when real incomeincreases, that is to say, the community will wish to consumea gradually diminishing proportion of it.

There are also other factors, over and above the operationof the general rule Just mentioned, which may operate to mod-ify the marginal propensity to consume, and hence the multi-plier; and these other factors seem likely, as a rule, to accentu-ate the tendency of the general rule rather than to offset jt. For,in the first place, the increase of employment will tend, owingto the effect of diminishing returns in the short period, to in-crease the proportion of aggregate income which accrues tothe entrepreneurs, whose individual marginal propensity toconsume is probably less than the average for the communityas a whole. In the second place, unemployment is likely tobe associated with negative saving in certain quarters, privateor public, because the unemployed may be living either onthe savings of themselves and their friends or on public reliefwhich is partly financed out of loans; with the result that re-employment will gradually diminish these particular acts ofnegative saving and reduce, therefore, the marginal propen-sity to consume more rapidly than would have occurred froman equal increase in the community’s real income accruing indifferent circumstances.

In any case, the multiplier is likely to be greater for a smallnet increment of investment than for a large increment; sothat, where substantial changes are in view, we must be guidedby the average value of the multiplier based on the averagemarginal propensity to consume over the range in question.

Mr Kahn has examined the probable quantitative result ofsuch factors as these in certain hypothetical special cases. But,clearly, it is not possible to carry any generalisation very far.One can only say, for example, that a typical modern commu-

118

IV.

nity would probably tend to consume not much less than 80per cent of any increment of real income, if it were a closedsystem with the consumption of the unemployed paid for bytransfers from the consumption of other consumers, so thatthe multiplier after allowing for offsets would not be muchless than 5. In a country, however, where foreign trade ac-counts for, say, 20 per cent of consumption and where the un-employed receive out of loans or their equivalent up to, say,50 per cent of their normal consumption when in work, themultiplier may fall as low as 2 or 3 times the employment pro-vided by a specific new investment. Thus a given fluctuationof investment will be associated with a much less violent fluc-tuation of employment in a country in which foreign tradeplays a large part and unemployment relief is financed on alarger scale out of borrowing (as was the case, e.g. in GreatBritain in 1931), than in a country in which these factors areless important (as in the United States in 1932).5

It is, however, to the general principle of the multiplier towhich we have to look for an explanation of how fluctuationsin the amount of investment, which are a comparatively smallproportion of the national income, are capable of generatingfluctuations in aggregate employment and income so muchgreater in amplitude than themselves.

IV

The discussion has been carried on, so far, on the basis ofa change in aggregate investment which has been foreseensufficiently in advance for the consumption industries to ad-vance pari passu with the capital-goods industries without moredisturbance to the price of consumption-goods than is conse-quential, in conditions of decreasing returns, on an increase inthe quantity which is produced.

In general, however, we have to take account of the casewhere the initiative comes from an increase in the output of

5Cf., however, below, p. 128, for an American estimate.

119


the capital-goods industries which was not fully foreseen. It isobvious that an initiative of this description only produces itsfull effect on employment over a period of time. I have found,however, in discussion that this obvious fact often gives rise tosome confusion between the logical theory of the multiplier,which holds good continuously, without time-lag, at all mo-ments of time, and the consequences of an expansion in thecapital-goods industries which take gradual effect, subject totime-lag and only after an interval.

The relationship between these two things can be clearedup by pointing out, firstly that an unforeseen, or imperfectlyforeseen, expansion in the capital-goods industries does nothave an instantaneous effect of equal amount on the aggre-gate of investment but causes a gradual increase of the latter;and, secondly, that it may cause a temporary departure of themarginal propensity to consume away from its normal value,followed, however, by a gradual return to it.

Thus an expansion in the capital-goods industries causesa series of increments in aggregate investment occurring insuccessive periods over an interval of time, and a series ofvalues of the marginal propensity to consume in these succes-sive periods which differ both from what the values wouldhave been if the expansion had been foreseen and from whatthey will be when the community has settled down to a newsteady level of aggregate investment. But in every interval oftime the theory of the multiplier holds good in the sense thatthe increment of aggregate demand is equal to the product ofthe increment of aggregate investment and the multiplier asdetermined by the marginal propensity to consume.

The explanation of these two sets of facts can be seen mostclearly by taking the extreme case where the expansion of em-ployment in the capital-goods industries is so entirely unfore-seen that in the first instance there is no increase whatever inthe output of consumption-goods. In this event the efforts ofthose newly employed in the capital-goods industries to con-sume a proportion of their increased incomes will raise theprices of consumption-goods until a temporary equilibriumbetween demand and supply has been brought about partly

120

IV.

by the high prices causing a postponement of consumption,partly by a redistribution of income in favour of the savingclasses as an effect of the increased profits resulting from thehigher prices, and partly by the higher prices causing a deple-tion of stocks. So far as the balance is restored by a postpone-ment of consumption there is a temporary reduction of themarginal propensity to consume, i.e. of the multiplier itself,and in so far as there is a depletion of stocks, aggregate invest-ment increases for the time being by less than the increment ofinvestment in the capital-goods industries,—i.e. the thing tobe multiplied does not increase by the full increment of invest-ment in the capital-goods industries. As time goes on, how-ever, the consumption-goods industries adjust themselves tothe new demand, so that when the deferred consumption isenjoyed, the marginal propensity to consume rises temporar-ily above its normal level, to compensate for the extent towhich it previously fell below it, and eventually returns toits normal level; whilst the restoration of stocks to their pre-vious figure causes the increment of aggregate investment tobe temporarily greater than the increment of investment inthe capital-goods industries (the increment of working capitalcorresponding to the greater output also having temporarilythe same effect).

The fact that an unforeseen change only exercises its fulleffect on employment over a period of time is important incertain contexts;—in particular it plays a part in the analysisof the trade cycle (on lines such as I followed in my Treatise onMoney). But it does not in any way affect the significance ofthe theory of the multiplier as set forth in this chapter; norrender it inapplicable as an indicator of the total benefit toemployment to be expected from an expansion in the capital.goods industries. Moreover, except in conditions where theconsumption industries are already working almost at capac-ity so that an expansion of output requires an expansion ofplant and not merely the more intensive employment of theexisting plant, there is no reason to suppose that more thana brief interval of time nced elapse before employment in theconsumption industries is advancing pari passu with employ-

121


ment in the capital-goods industries with the multiplier oper-ating near its normal figure.

V

We have seen above that the greater the marginal propensityto consume, the greater the multiplier, and hence the greaterthe disturbance to employment corresponding to a given changein investment. This might seem to lead to the paradoxical con-clusion that a poor community in which saving is a very smallproportion of income will be more subject to violent fluctua-tions than a wealthy community where saving is a larger pro-portion of income and the multiplier consequently smaller.

This conclusion, however, would overlook the distinctionbetween the effects of the marginal propensity to consumeand those of the average propensity to consume. For whilst ahigh marginal propensity to consume involves a larger propor-tionate effect from a given percentage change in investment,the absolute effect will, nevertheless, be small if the averagepropensity to consume is also high. This may be illustrated asfollows by a numerical example.

Let us suppose that a community’s propensity to consumeis such that, so long as its real income does not exceed theoutput from employing 5,000,000 men on its existing capitalequipment, it consumes the whole of its income; that of theoutput of the next 100,000 additional men employed it con-sumes 99 per cent, of the next 100,000 after that 98 per cent,of the third 100,000 97 per cent and so on; and that 10,000,000men employed represents full employment. It follows fromthis that, when 5,000,000 + n 100,000 men are employed, themultiplier at the margin is 100/n, and n(n+i)

2(50+n)per cent of the

national income is invested.Thus when 5,200,000 men are employed the multiplier is

very large, namely 50, but investment is only a trifling pro-portion of current income, namely, 0.06 per cent; with theresult that if investment falls off by a large proportion, sayabout two-thirds, employment will only decline to 5,100,000,

122

V.

i.e. by about 2 per cent. On the other hand, when 9,000,000men are employed, the marginal multiplier is comparativelysmall, namely 2 1

2, but investment is now a substantial propor-

tion of current income, namely, 9 per cent; with the result thatif investment falls by two-thirds, employment will decline to6,900,000, namely, by 19 per cent. In the limit where invest-ment falls off to zero, employment will decline by about 4 percent in the former case, whereas in the latter case it will de-cline by 44 per cent.6

In the above example, the poorer of the two communitiesunder comparison is poorer by reason of under-employment.But the same reasoning applies by easy adaptation if the povertyis due to inferior skill, technique or equipment. Thus whilstthe multiplier is larger in a poor community, the effect on em-ployment of fluctuations in investment will be much greaterin a wealthy community, assuming that in the latter current in-vestment represents a much larger proportion of current out-put.7

It is also obvious from the above that the employment ofa given number of men on public works will (on the assump-tions made) have a much larger effect on aggregate employ-ment at a time when there is severe unemployment, than itwill have later on when full employment is approached. In theabove example, if, at a time when employment has fallen to5,200,000, an additional 100,000 men are employed on public

6Quantity of investment is measured, above, by the number of men em-ployed in producing it. Thus if there are diminishing returns per unit of em-ployment as employment increases, what is double the quantity of investmenton the above scale will be less than double on a physical (if such a scale is avail-able).

7More generally, the ratio of the proportional change in total demand to theproportional change in investment

=∆Y

Y

∆I

I=

∆Y

Y·

Y − C∆Y −∆C

=1− C

Y

1− ∂C∂Y

As wealth increases ∂Y∂Y diminishes, but C

Y also diminishes. Thus the frac-tion increases or diminishes according as consumption increases or diminishesin a smaller or greater proportion than income.

123


works, total employment will rise to 6,400,000. But if employ-ment is already 9,000,000 when the additional 100,000 men aretaken on for public works, total employment will only rise to9,200,000. Thus public works even of doubtful utility may payfor themselves over and over again at a time of severe unem-ployment, if only from the diminished cost of relief expendi-ture, provided that we can assume that a smaller proportion ofincome is saved when unemployment is greater; but they maybecome a more doubtful proposition as a state of full employ-ment is approached. Furthermore, if our assumption is correctthat the marginal propensity to consume falls off steadily aswe approach full employment, it follows that it will becomemore and more troublesome to secure a further given increaseof employment by further increasing investment. It shouldnot be difficult to compile a chart of the marginal propensityto consume at each stage of a trade cycle from the statistics(if they were available) of aggregate incorne and aggregate in-vestment at successive dates. At present, however, our statis-tics are not accurate enough (or compiled sufficiently with thisspecific object in view) to allow us to infer more than highlyapproximate estimates. The best for the purpose, of which Iam aware, are Mr Kuznets’ figures for the United States (al-ready referred to, p.103 above), though they are, nevertheless,very precarious. Taken in conjunction with estimates of na-tional income these suggest, for what they are worth, botha lower figure and a more stable figure for the investmentmultiplier than I should have expected. If single years aretaken in isolation, the results look rather wild. But if theyare grouped in pairs, the multiplier seems to have been lessthan 3 and probably fairly stable in the neighbourhood of 2.5.This suggests a marginal propensity to consume not exceed-ing 6o to 70 per cent—a figure quite plausible for the boom,but surprisingly, and, in my judgment, improbably low forthe slump. It is possible, however, that the extreme financialconservatism of corporate finance in the United States, evenduring the slump, may account for it. In other words, if, wheninvestment is falling heavily through a failure to undertake re-pairs and replacements, financial provision is made, neverthe-

124

VI.

less, in respect of such wastage, the effect is to prevent the risein the marginal propensity to consume which would have oc-curred otherwise. I suspect that this factor may have played asignificant part in aggravating the degree of the recent slumpin the United States. On the other hand, it is possible that thestatistics somewhat overstate the decline in investment, whichis alleged to have fallen off by more than 75 per cent in 1932compared with 1929, whilst net ’capital formation’ declinedby more than 95 per cent;—a moderate change in these esti-mates being capable of making a substantial difference to themultiplier.

VI

When involuntary unemployment exists, the marginal disutil-ity of labour is necessarily less than the utility of the marginalproduct. Indeed it may be much less. For a man who hasbeen long unemployed some measure of labour, instead ofinvolving disutility, may have a positive utility. If this is ac-cepted, the above reasoning shows how ‘wasteful’ loan ex-penditure8 may nevertheless enrich the community on bal-ance. Pyramid-building, earthquakes, even wars may serveto increase wealth, if the education of our statesmen on theprinciples of the classical economics stands in the way of any-thing better.

It is curious how common sense, wriggling for an escapefrom absurd conclusions, has been apt to reach a preferencefor wholly ‘wasteful’ forms of loan expenditure rather thanfor partly wasteful forms, which, because they are not wholly

8It is often convenient to use the term “loan expenditure” to include bothpublic investment financed by borrowing from individuals and also any othercurrent public expenditure which is so financed. Strictly speaking, the lattershould be reckoned as negative saving, but official action of this kind is notinfluenced by the same sort of psychological motives as those which governprivate saying. Thus “loan expenditure” is a convenient expression for the netborrowings of public authorities on all accounts, whether on capital accountor to meet a budgetary deficit. The one form of loan expenditure operates byincreasing investment and the other by increasing the propensity to consume.

125


wasteful, tend to be judged on strict ‘business’ principles. Forexample, unemployment relief financed by loans is more read-ily accepted than the financing of improvements at a chargebelow the current rate of interest; whilst the form of diggingholes in the ground known as gold-mining, which not onlyadds nothing whatever to the real wealth of the world butinvolves the disutility of labour, is the most acceptable of allsolutions.

If the Treasury were to fill old bottles with banknotes, burythem at suitable depths in disused coalmines which are thenfilled up to the surface with town rubbish, and leave it toprivate enterprise on well-tried principles of laissez-faire todig the notes up again (the right to do so being obtained,of course, by tendering for leases of the note-bearing terri-tory), there need be no more unemployment and, with thehelp of the repercussions, the real income of the community,and its capital wealth also, would probably become a gooddeal greater than it actually is. It would, indeed, be more sen-sible to build houses and the like; but if there are political andpractical difficulties in the way of this, the above would bebetter than nothing.

The analogy between this expedient and the goldmines ofthe real world is complete. At periods when gold is availableat suitable depths experience shows that the real wealth of theworld increases rapidly; and when but little of it is so availableour wealth suffers stagnation or decline. Thus gold-mines areof the greatest value and importance to civilisation. Just aswars have been the only form of large-scale loan expenditurewhich statesmen have thought justifiable, so gold-mining isthe only pretext for digging holes in the ground which hasrecommended itself to bankers as sound finance; and each ofthese activities has played its part in progress—failing some-thing better. To mention a detail, the tendency in slumps forthe price of gold to rise in terms of labour and materials aidseventual recovery, because it increases the depth at which gold-digging pays and lowers the minimum grade of ore which ispayable.

In addition to the probable effect of increased supplies of

126

VI.

gold on the rate of interest, gold-mining is for two reasons ahighly practical form of investment, if we are precluded fromincreasing employment by means which at the same time in-crease our stock of useful wealth. In the first place, owing tothe gambling attractions which it offers it is carried on withouttoo close a regard to the ruling rate of interest. In the secondplace the result, namely, the increased stock of gold, does not,as in other cases, have the effect of diminishing its marginalutility. Since the value of a house depends on its utility, everyhouse which is built serves to diminish the prospective rentsobtainable from further house-building and therefore lessensthe attraction of further similar investment unless the rate ofinterest is falling pari passu. But the fruits of gold-mining donot suffer from this disadvantage, and a check can only comethrough a rise of the wage-unit in terms of gold, which is notlikely to occur unless and until employment is substantiallybetter. Moreover, there is no subsequent reverse effect on ac-count of provision for user and supplementary costs, as in thecase of less durable forms of wealth.

Ancient Egypt was doubly fortunate, and doubtless owedto this its fabled wealth, in that it possessed two activities,namely, pyramid-building as well as the search for the pre-cious metals, the fruits of which, since they could not servethe needs of man by being consumed, did not stale with abun-dance. The Middle Ages built cathedrals and sang dirges.Two pyramids, two masses for the dead, are twice as goodas one; but not so two railways from London to York. Thuswe are so sensible, have schooled ourselves to so close a sem-blance of prudent financiers, taking careful thought before weadd to the ’financial’ burdens of posterity by building themhouses to live in, that we have no such easy escape from thesufferings of unemployment. We have to accept them as aninevitable result of applying to the conduct of the State themaxims which are best calculated to ’enrich’ an individual byenabling him to pile up claims to enjoyment which he doesnot intend to exercise at any definite time.

127

Book IV

The Inducement to Invest

CHAPTER 11

THE MARGINAL EFFICIENCY OF CAPITAL

I

When a man buys an investment or capital-asset, he purchasesthe right to the series of prospective returns, which he expectsto obtain from selling its output, after deducting the runningexpenses of obtaining that output, during the life of the asset.This series of annuities Q1, Q2, . . . , Qn it is convenient to callthe prospective yield of the investment.

Over against the prospective yield of the investment wehave the supply price of the capital-asset, meaning by this, notthe market-price at which an asset of the type in questioncan actually be purchased in the market, but the price whichwould just induce a manufacturer newly to produce an ad-ditional unit of such assets, i.e. what is sometimes called itsreplacement cost. The relation between the prospective yield ofa capital-asset and its supply price or replacement cost, i.e. therelation between the prospective yield of one more unit of thattype of capital and the cost of producing that unit, furnishesus with the marginal efficiency of capital of that type. More pre-cisely, I define the marginal efficiency of capital as being equalto that rate of discount which would make the present valueof the series of annuities given by the returns expected fromthe capital-asset during its life just equal to its supply price.This gives us the marginal efficiencies of particular types ofcapital-assets. The greatest of these marginal efficiencies canthen be regarded as the marginal efficiency of capital in gen-eral.

The reader should note that the marginal efficiency of cap-ital is here defined in terms of the expectation of yield and ofthe current supply price of the capital-asset. It depends on the

131

11. THE MARGINAL EFFICIENCY OF CAPITAL

rate of return expected to be obtainable on money if it wereinvested in a newly produced asset; not on the historical resultof what an investment has yielded on its original cost if welook back on its record after its life is over.

If there is an increased investment in any given type ofcapital during any period of time, the marginal efficiency ofthat type of capital will diminish as the investment in it is in-creased, partly because the prospective yield will fall as thesupply of that type of capital is increased, and partly because,as a rule, pressure on the facilities for producing that type ofcapital will cause its supply price to increase; the second ofthese factors being usually the more important in producingequilibrium in the short run, but the longer the period in viewthe more does the first factor take its place. Thus for each typeof capital we can build up a schedule, showing by how muchinvestment in it will have to increase within the period, in or-der that its marginal efficiency should fall to any given fig-ure. We can then aggregate these schedules for all the differenttypes of capital, so as to provide a schedule relating the rateof aggregate investment to the corresponding marginal effi-ciency of capital in general which that rate of investment willestablish. We shall call this the investment demand-schedule;or, alternatively, the schedule of the marginal efficiency of cap-ital.

Now it is obvious that the actual rate of current investmentwill be pushed to the point where there is no longer any classof capital-asset of which the marginal efficiency exceeds thecurrent rate of interest. In other words, the rate of invest-ment will be pushed to the point on the investment demand-schedule where the marginal efficiency of capital in general isequal to the market rate of interest.1

The same thing can also be expressed as follows. If Qr is

1For the sake of simplicity of statement I have slurred the point that weare dealing with complexes of rates of interest and discount corresponding tothe different lengths of time which will elapse before the various prospectivereturns from the asset are realised. But it is not difficult to re-state the argumentso as to cover this point.

132

II.

the prospective yield from an asset at time r, and dr is thepresent value of £1 deferred r years at the current rate of in-terest, ΣQrdr is the demand price of the investment; and in-vestment will be carried to the point where ΣQrdr becomesequal to the supply price of the investment as defined above.If, on the other hand, ΣQrdr falls short of the supply price,there will be no current investment in the asset in question.

It follows that the inducement to invest depends partly onthe investment demand-schedule and partly on the rate of in-terest. Only at the conclusion of Book IV will it be possible totake a comprehensive view of the factors determining the rateof investment in their actual complexity. I would, however,ask the reader to note at once that neither the knowledge ofan asset’s prospective yield nor the knowledge of the marginalefficiency of the asset enables us to deduce either the rate ofinterest or the present value of the asset. We must ascertainthe rate of interest from some other source, and only then canwe value the asset by ‘capitalising’ its prospective yield.

II

How is the above definition of the marginal efficiency of cap-ital related to common usage? The Marginal Productivity orYield or Efficiency or Utility of Capital are familiar terms whichwe have all frequently used. But it is not easy by searchingthe literature of economics to find a clear statement of whateconomists have usually intended by these terms.

There are at least three ambiguities to clear up. There is,to begin with, the ambiguity whether we are concerned withthe increment of physical product per unit of time due to theemployment of one more physical unit of capital, or with theincrement of value due to the employment of one more valueunit of capital. The former involves difficulties as to the defi-nition of the physical unit of capital, which I believe to be bothinsoluble and unnecessary. It is, of course, possible to say thatten labourers will raise more wheat from a given area whenthey are in a position to make use of certain additional ma-

133


chines; but I know no means of reducing this to an intelligiblearithmetical ratio which does not bring in values. Neverthe-less many discussions of this subject seem to be mainly con-cerned with the physical productivity of capital in some sense,though the writers fail to make themselves clear.

Secondly, there is the question whether the marginal effi-ciency of capital is some absolute quantity or a ratio. The con-texts in which it is used and the practice of treating it as beingof the same dimension as the rate of interest seem to requirethat it should be a ratio. Yet it is not usually made clear whatthe two terms of the ratio are supposed to be.

Finally, there is the distinction, the neglect of which hasbeen the main cause of confusion and misunderstanding, be-tween the increment of value obtainable by using an addi-tional quantity of capital in the existing situation, and the se-ries of increments which it is expected to obtain over the wholelife of the additional capital asset; — i.e. the distinction be-tween Q1 and the complete series Q1, Q2, . . . , Qr , . . . . Thisinvolves the whole question of the place of expectation in eco-nomic theory. Most discussions of the marginal efficiency ofcapital seem to pay no attention to any member of the seriesexcept Q1 . Yet this cannot be legitimate except in a Static the-ory, for which all the Q’s are equal. The ordinary theory ofdistribution, where it is assumed that capital is getting now itsmarginal productivity (in some sense or other), is only valid ina stationary state. The aggregate current return to capital hasno direct relationship to its marginal efficiency; whilst its cur-rent return at the margin of production (i.e. the return to capi-tal which enters into the supply price of output) is its marginaluser cost, which also has no close connection with its marginalefficiency.

There is, as I have said above, a remarkable lack of anyclear account of the matter. At the same time I believe thatthe definition which I have given above is fairly close to whatMarshall intended to mean by the term. The phrase whichMarshall himself uses is ‘marginal net efficiency’ of a factorof production; or, alternatively, the ‘marginal utility of capi-tal’. The following is a summary of the most relevant passage

134

II.

which I can find in his Principles (6th ed. pp. 519-520). I haverun together some non-consecutive sentences to convey thegist of what he says:

In a certain factory an extra £100 worth of machinery can beapplied so as not to involve any other extra expense, and so asto add annually £3 worth to the net output of the factory afterallowing for its own wear and tear. If the investors of capitalpush it into every occupation in which it seems likely to gain ahigh reward; and if, after this has been done and equilibriumhas been found, it still pays and only just pays to employ thismachinery, we can infer from this fact that the yearly rate ofinterest is 3 per cent. But illustrations of this kind merely indi-cate part of the action of the great causes which govern value.They cannot be made into a theory of interest, any more thaninto a theory of wages, without reasoning in a circle . . . . Sup-pose that the rate of interest is 3 per cent. per annum on per-fectly good security; and that the hat-making trade absorbsa capital of one million pounds. This implies that the hat-making trade can turn the whole million pounds’ worth ofcapital to so good account that they would pay 3 per cent. perannum net for the use of it rather than go without any of it.There may be machinery which the trade would have refusedto dispense with if the rate of interest had been 20 per cent.per annum. If the rate had been 10 per cent., more wouldhave been used; if it had been 6 per cent., still more; if 4 percent. still more; and finally, the rate being 3 per cent., they usemore still. When they have this amount, the marginal utilityof the machinery, i.e. the utility of that machinery which it isonly just worth their while to employ, is measured by 3 percent.

It is evident from the above that Marshall was well awarethat we are involved in a circular argument if we try to de-termine along these lines what the rate of interest actually is.2

In this passage he appears to accept the view set forth above,that the rate of interest determines the point to which new in-

2But was he not wrong in supposing that the marginal productivity theoryof wages is equally circular?

135


vestment will be pushed, given the schedule of the marginalefficiency of capital. If the rate of interest is 3 per cent, thismeans that no one will pay £100 for a machine unless he hopesthereby to add £3 to his annual net output after allowing forcosts and depreciation. But we shall see in chapter 14 thatin other passages Marshall was less cautious — though stilldrawing back when his argument was leading him on to du-bious ground.

Although he does not call it the ‘marginal efficiency of cap-ital’, Professor Irving Fisher has given in his Theory of Interest(1930) a definition of what he calls ‘the rate of return over cost’which is identical with my definition. ‘The rate of return overcost’, he writes,3 ‘is that rate which, employed in computingthe present worth of all the costs and the present worth of allthe returns, will make these two equal.’ Professor Fisher ex-plains that the extent of investment in any direction will de-pend on a comparison between the rate of return over costand the rate of interest. To induce new investment ‘the rate ofreturn over cost must exceed the rate of interest’.4 ‘This newmagnitude (or factor) in our study plays the central rôle on theinvestment opportunity side of interest theory.’5 Thus Profes-sor Fisher uses his ‘rate of return over cost’ in the same senseand for precisely the same purpose as I employ ‘the marginalefficiency of capital’.

III

The most important confusion concerning the meaning andsignificance of the marginal efficiency of capital has ensuedon the failure to see that it depends on the prospective yieldof capital, and not merely on its current yield. This can be bestillustrated by pointing out the effect on the marginal efficiencyof capital of an expectation of changes in the prospective cost

3Op. cit. p. 1684Op. cit. p. 159.5Op. cit. p. 155.

136

III.

of production, whether these changes are expected to comefrom changes in labour cost, i.e. in the wage-unit, or frominventions and new technique. The output from equipmentproduced to-day will have to compete, in the course of its life,with the output from equipment produced subsequently, per-haps at a lower labour cost, perhaps by an improved tech-nique, which is content with a lower price for its output andwill be increased in quantity until the price of its output hasfallen to the lower figure with which it is content. Moreover,the entrepreneur’s profit (in terms of money) from equipment,old or new, will be reduced, if all output comes to be producedmore cheaply. In so far as such developments are foreseen asprobable, or even as possible, the marginal efficiency of capi-tal produced to-day is appropriately diminished.

This is the factor through which the expectation of changesin the value of money influences the volume of current output.The expectation of a fall in the value of money stimulates in-vestment, and hence employment generally, because it raisesthe schedule of the marginal efficiency of capital, i.e. the in-vestment demand-schedule; and the expectation of a rise inthe value of money is depressing, because it lowers the sched-ule of the marginal efficiency of capital.

This is the truth which lies behind Professor Irving Fisher’stheory of what he originally called ‘Appreciation and Inter-est’ — the distinction between the money rate of interest andthe real rate of interest where the latter is equal to the for-mer after correction for changes in the value of money. Itis difficult to make sense of this theory as stated, because itis not clear whether the change in the value of money is oris not assumed to be foreseen. There is no escape from thedilemma that, if it is not foreseen, there will be no effect oncurrent affairs; whilst, if it is foreseen, the prices of existinggoods will be forthwith so adjusted that the advantages ofholding money and of holding goods are again equalised, andit will be too late for holders of money to gain or to suffer achange in the rate of interest which will offset the prospectivechange during the period of the loan in the value of the moneylent. For the dilemma is not successfully escaped by Professor

137


Pigou’s expedient of supposing that the prospective changein the value of money is foreseen by one set of people but notforeseen by another.

The mistake lies in supposing that it is the rate of intereston which prospective changes in the value of money will di-rectly react, instead of the marginal efficiency of a given stockof capital. The prices of existing assets will always adjustthemselves to changes in expectation concerning the prospec-tive value of money. The significance of such changes in ex-pectation lies in their effect on the readiness to produce newassets through their reaction on the marginal efficiency of cap-ital. The stimulating effect of the expectation of higher pricesis due, not to its raising the rate of interest (that would be aparadoxical way of stimulating output — in so far as the rateof interest rises, the stimulating effect is to that extent offset),but to its raising the marginal efficiency of a given stock ofcapital. If the rate of interest were to rise pari passu with themarginal efficiency of capital, there would be no stimulatingeffect from the expectation of rising prices. For the stimulus tooutput depends on the marginal efficiency of a given stock ofcapital rising relatively to the rate of interest. Indeed ProfessorFisher’s theory could be best re-written in terms of a ‘real rateof interest’ defined as being the rate of interest which wouldhave to rule, consequently on a change in the state of expecta-tion as to the future value of money, in order that this changeshould have no effect on current output.6 It is worth notingthat an expectation of a future fall in the rate of interest willhave the effect of lowering the schedule of the marginal effi-ciency of capital; since it means that the output from equip-ment produced to-day will have to compete during part ofits life with the output from equipment which is content witha lower return. This expectation will have no great depress-ing effect, since the expectations, which are held concerningthe complex of rates of interest for various terms which willrule in the future, will be partially reflected in the complex

6Cf. Mr. Robertson’s article on “Industrial Fluctuations and the NaturalRate of Interest”, Economic Journal, December 1934.

138

IV.

of rates of interest which rule to-day. Nevertheless there maybe some depressing effect, since the output from equipmentproduced to-day, which will emerge towards the end of thelife of this equipment, may have to compete with the outputof much younger equipment which is content with a lowerreturn because of the lower rate of interest which rules for pe-riods subsequent to the end of the life of equipment producedto-day.

It is important to understand the dependence of the marginalefficiency of a given stock of capital on changes in expecta-tion, because it is chiefly this dependence which renders themarginal efficiency of capital subject to the somewhat violentfluctuations which are the explanation of the trade cycle. Inchapter 22 below we shall show that the succession of boomand slump can be described and analysed in terms of the fluc-tuations of the marginal efficiency of capital relatively to therate of interest.

IV

Two types of risk affect the volume of investment which havenot commonly been distinguished, but which it is importantto distinguish. The first is the entrepreneur’s or borrower’srisk and arises out of doubts in his own mind as to the prob-ability of his actually earning the prospective yield for whichhe hopes. If a man is venturing his own money, this is the onlyrisk which is relevant.

But where a system of borrowing and lending exists, bywhich I mean the ranting of loans with a margin of real or per-sonal security, a second type of risk is relevant which we maycall the lender’s risk. This may be due either to moral haz-ard, i.e. voluntary default or other means of escape, possiblylawful, from the fulfilment of the obligation, or to the possibleinsufficiency of the margin of security, i.e. involuntary defaultdue to the disappointment of expectation. A third source ofrisk might be added, namely, a possible adverse change in thevalue of the monetary standard which renders a money-loan

139


to this extent less secure than a real asset; though all or mostof this should be already reflected, and therefore absorbed, inthe price of durable real assets.

Now the first type of risk is, in a sense, a real social cost,though susceptible to diminution by averaging as well as byan increased accuracy of foresight. The second, however, isa pure addition to the cost of investment which would notexist if the borrower and lender were the same person. More-over, it involves in part a duplication of a proportion of theentrepreneur’s risk, which is added twice to the pure rate ofinterest to give the minimum prospective yield which will in-duce the investment. For if a venture is a risky one, the bor-rower will require a wider margin between his expectation ofyield and the rate of interest at which he will think it worthhis while to borrow; whilst the very same reason will lead thelender to require a wider margin between what he chargesand the pure rate of interest in order to induce him to lend(except where the borrower is so strong and wealthy that he isin a position to offer an exceptional margin of security). Thehope of a very favourable outcome, which may balance therisk in the mind of the borrower, is not available to solace thelender.

This duplication of allowance for a portion of the risk hasnot hitherto been emphasised, so far as I am aware; but it maybe important in certain circumstances. During a boom thepopular estimation of the magnitude of both these risks, bothborrower’s risk and lender’s risk, is apt to become unusuallyand imprudently low.

V

The schedule of the marginal efficiency of capital is of fun-damental importance because it is mainly through this factor(much more than through the rate of interest) that the expec-tation of the future influences the present. The mistake of re-garding the marginal efficiency of capital primarily in termsof the current yield of capital equipment, which would be cor-

140

V.

rect only in the static state where there is no changing futureto influence the present, has had the result of breaking the the-oretical link between to-day and to-morrow. Even the rate ofinterest is, virtually,7 a current phenomenon; and if we reducethe marginal efficiency of capital to the same status, we cutourselves off from taking any direct account of the influenceof the future in our analysis of the existing equilibrium.

The fact that the assumptions of the static state often un-derlie present-day economic theory, imports into it a large el-ement of unreality. But the introduction of the concepts ofuser cost and of the marginal efficiency of capital, as definedabove, will have the effect, I think, of bringing it back to re-ality, whilst reducing to a minimum the necessary degree ofadaptation.

It is by reason of the existence of durable equipment thatthe economic future is linked to the present. It is, therefore,consonant with, and agreeable to, our broad principles of thought,that the expectation of the future should affect the presentthrough the demand price for durable equipment.

7Not completely; for its value partly reflects the uncertainty of the future.Moreover, the relation between rates of interest for different terms depends onexpectations.

141

CHAPTER 12

THE STATE OF LONG-TERM EXPECTATION

I

We have seen in the previous chapter that the scale of in-vestment depends on the relation between the rate of interestand the schedule of the marginal efficiency of capital corre-sponding to different scales of current investment, whilst themarginal efficiency of capital depends on the relation betweenthe supply price of a capital-asset and its prospective yield. Inthis chapter we shall consider in more detail some of the fac-tors which determine the prospective yield of an asset.

The considerations upon which expectations of prospec-tive yields are based are partly existing facts which we canassume to be known more or less for certain, and partly futureevents which can only be forecasted with more or less confi-dence. Amongst the first may be mentioned the existing stockof various types of capital-assets and of capital-assets in gen-eral and the strength of the existing consumers’ demand forgoods which require for their efficient production a relativelylarger assistance from capital. Amongst the latter are futurechanges in the type and quantity of the stock of capital-assetsand in the tastes of the consumer, the strength of effective de-mand from time to time during the life of the investment un-der consideration, and the changes in the wage-unit in termsof money which may occur during its life. We may sum up thestate of psychological expectation which covers the latter asbeing the state of long-term expectation;—as distinguished fromthe short-term expectation upon the basis of which a producerestimates what he will get for a product when it is finishedif he decides to begin producing it to-day with the existingplant, which we examined in chapter 5.

143

12. THE STATE OF LONG-TERM EXPECTATION

II

It would be foolish, in forming our expectations, to attachgreat weight to matters which are very uncertain.1 It is reason-able, therefore, to be guided to a considerable degree by thefacts about which we feel somewhat confident, even thoughthey may be less decisively relevant to the issue than otherfacts about which our knowledge is vague and scanty. Forthis reason the facts of the existing situation enter, in a sensedisproportionately, into the formation of our long-term expec-tations; our usual practice being to take the existing situationand to project it into the future, modified only to the extentthat we have more or less definite reasons for expecting achange.

The state of long-term expectation, upon which our de-cisions are based, does not solely depend, therefore, on themost probable forecast we can make. It also depends on theconfidence with which we make this forecast—on how highlywe rate the likelihood of our best forecast turning out quitewrong. If we expect large changes but are very uncertain asto what precise form these changes will take, then our confi-dence will be weak.

The state of confidence, as they term it, is a matter to whichpractical men always pay the closest and most anxious at-tention. But economists have not analysed it carefully andhave been content, as a rule, to discuss it in general terms.In particular it has not been made clear that its relevance toeconomic problems comes in through its important influenceon the schedule of the marginal efficiency of capital. Thereare not two separate factors affecting the rate of investment,namely, the schedule of the marginal efficiency of capital andthe state of confidence. The state of confidence is relevant be-cause it is one of the major factors determining the former,which is the same thing as the investment demand-schedule.

There is, however, not much to be said about the state of

1By “very uncertain” I do not mean the same thing as “improbable”. Cf.my Treatise on Probability, chap. 6, on “The Weight of Arguments”.

144

III.

confidence a priori. Our conclusions must mainly depend uponthe actual observation of markets and business psychology.This is the reason why the ensuing digression is on a differentlevel of abstraction from most of this book.

For convenience of exposition we shall assume in the fol-lowing discussion of the state of confidence that there are nochanges in the rate of interest; and we shall write, throughoutthe following sections, as if changes in the values of invest-ments were solely due to changes in the expectation of theirprospective yields and not at all to changes in the rate of in-terest at which these prospective yields are capitalised. Theeffect of changes in the rate of interest is, however, easily su-perimposed on the effect of changes in the state of confidence.

III

The outstanding fact is the extreme precariousness of the ba-sis of knowledge on which our estimates of prospective yieldhave to be made. Our knowledge of the factors which willgovern the yield of an investment some years hence is usuallyvery slight and often negligible. If we speak frankly, we haveto admit that our basis of knowledge for estimating the yieldten years hence of a railway, a copper mine, a textile factory,the goodwill of a patent medicine, an Atlantic liner, a build-ing in the City of London amounts to little and sometimes tonothing; or even five years hence. In fact, those who seriouslyattempt to make any such estimate are often so much in theminority that their behaviour does not govern the market.

In former times, when enterprises were mainly owned bythose who undertook them or by their friends and associates,investment depended on a sufficient supply of individuals ofsanguine temperament and constructive impulses who em-barked on business as a way of life, not really relying on aprecise calculation of prospective profit. The affair was partlya lottery, though with the ultimate result largely governedby whether the abilities and character of the managers wereabove or below the average. Some would fail and some would

145


succeed. But even after the event no one would know whetherthe average results in terms of the sums invested had exceeded,equalled or fallen short of the prevailing rate of interest; though,if we exclude the exploitation of natural resources and mo-nopolies, it is probable that the actual average results of in-vestments, even during periods of progress and prosperity,have disappointed the hopes which prompted them. Businessmen play a mixed game of skill and chance, the average re-sults of which to the players are not known by those who takea hand. If human nature felt no temptation to take a chance,no satisfaction (profit apart) in constructing a factory, a rail-way, a mine or a farm, there might not be much investmentmerely as a result of cold calculation.

Decisions to invest in private business of the old-fashionedtype were, however, decisions largely irrevocable, not only forthe community as a whole, but also for the individual. Withthe separation between ownership and management whichprevails to-day and with the development of organised in-vestment markets, a new factor of great importance has en-tered in, which sometimes facilitates investment but some-times adds greatly to the instability of the system. In the ab-sence of security markets, there is no object in frequently at-tempting to revalue an investment to which we are commit-ted. But the Stock Exchange revalues many investments everyday and the revaluations give a frequent opportunity to theindividual (though not to the community as a whole) to re-vise his commitments. It is as though a farmer, having tappedhis barometer after breakfast, could decide to remove his cap-ital from the farming business between 10 and II in the morn-ing and reconsider whether he should return to it later in theweek. But the daily revaluations of the Stock Exchange, thoughthey are primarily made to facilitate transfers of old invest-ments between one individual and another, inevitably exerta decisive influence on the rate of current investment. Forthere is no sense in building up a new enterprise at a costgreater than that at which a similar existing enterprise can bepurchased; whilst there is an inducement to spend on a newproject what may seem an extravagant sum, if it can be floated

146

IV.

off on the Stock Exchange at an immediate profit.2 Thus cer-tain classes of investment are governed by the average expec-tation of those who deal on the Stock Exchange as revealedin the price of shares, rather than by the genuine expectationsof the professional entrepreneur.3 How then are these highlysignificant daily, even hourly, revaluations of existing invest-ments carried out in practice?

IV

In practice we have tacitly agreed, as a rule, to fall back onwhat is, in truth, a convention. The essence of this conven-tion—though it does not, of course, work out quite so sim-ply—lies in assuming that the existing state of affairs will con-tinue indefinitely, except in so far as we have specific reasonsto expect a change. This does not mean that we really be-lieve that the existing state of affairs will continue indefinitely.We know from extensive experience that this is most unlikely.The actual results of an investment over a long term of yearsvery seldom agree with the initial expectation. Nor can werationalise our behaviour by arguing that to a man in a stateof ignorance errors in either direction are equally probable, sothat there remains a mean actuarial expectation based on equi-probabilities. For it can easily be shown that the assumptionof arithmetically equal probabilities based on a state of igno-

2In my Treatise on Money (vol. ii. p. 195) I pointed out that when a com-pany’s shares are quoted very high so that it can raise more capital by issuingmore shares on favourable terms, this has the same effect as if it could borrowat a low rate of interest. I should now describe this by saying that a high quo-tation for existing equities involves an increase in the marginal efficiency of thecorresponding type of capital and therefore has the same effect (since invest-ment depends on a comparison between the marginal efficiency of capital andthe rate of interest) as a fall in the rate of interest.

3This does not apply, of course, to classes of enterprise which are not read-ily marketable or to which no negotiable instrument closely corresponds. Thecategories failing within this exception were formerly extensive. But measuredas a proportion of the total value of new investment they are rapidly decliningin importance.

147


rance leads to absurdities. We are assuming, in effect, thatthe existing market valuation, however arrived at, is uniquelycorrect in relation to our existing knowledge of the facts whichwill influence the yield of the investment, and that it will onlychange in proportion to changes in this knowledge; though,philosophically speaking, it cannot be uniquely correct, sinceour existing knowledge does not provide a sufficient basis fora calculated mathematical expectation. In point of fact, allsorts of considerations enter into the market valuation whichare in no way relevant to the prospective yield.

Nevertheless the above conventional method of calcula-tion will be compatible with a considerable measure of con-tinuity and stability in our affairs, so long as we can rely on themaintenance of the convention.

For if there exist organised investment markets and if wecan rely on the maintenance of the convention, an investor canlegitimately encourage himself with the idea that the only riskhe runs is that of a genuine change in the news over the nearfuture, as to the likelihood of which he can attempt to formhis own judgment, and which is unlikely to be very large.For, assuming that the convention holds good, it is only thesechanges which can affect the value of his investment, and heneed not lose his sleep merely because he has not any no-tion what his investment will be worth ten years hence. Thusinvestment becomes reasonably ‘safe’ for the individual in-vestor over short periods, and hence over a succession of shortperiods however many, if he can fairly rely on there beingno breakdown in the convention and on his therefore havingan opportunity to revise his judgment and change his invest-ment, before there has been time for much to happen. Invest-ments which are ‘fixed’ for the community are thus made ‘liq-uid’ for the individual.

It has been, I am sure, on the basis of some such procedureas this that our leading investment markets have been devel-oped. But it is not surprising that a convention, in an absoluteview of things so arbitrary, should have its weak points. It isits precariousness which creates no small part of our contem-porary problem of securing sufficient investment.

148

V.

V

Some of the factors which accentuate this precariousness maybe briefly mentioned.

(1) As a result of the gradual increase in the proportion ofthe equity in the community’s aggregate capital investmentwhich is owned by persons who do not manage and haveno special knowledge of the circumstances, either actual orprospective, of the business in question, the element of realknowledge in the valuation of investments by whose who ownthem or contemplate purchasing them has seriously declined.

(2) Day-to-day fluctuations in the profits of existing invest-ments, which are obviously of an ephemeral and non-significantcharacter, tend to have an altogether excessive, and even anabsurd, influence on the market. It is said, for example, thatthe shares of American companies which manufacture ice tendto sell at a higher price in summer when their profits are sea-sonally high than in winter when no one wants ice. The recur-rence of a bank-holiday may raise the market valuation of theBritish railway system by several million pounds.

(3) A conventional valuation which is established as the out-come of the mass psychology of a large number of ignorant in-dividuals is liable to change violently as the result ofa suddenfluctuation of opinion due to factors which do not really makemuch difference to the prospective yield; since there will be nostrong roots of conviction to hold it steady. In abnormal timesin particular, when the hypothesis of an indefinite continu-ance of the existing state of affairs is less plausible than usualeven though there are no express grounds to anticipate a def-inite change, the market will be subject to waves of optimisticand pessimistic sentiment, which are unreasoning and yet ina sense legitimate where no solid basis exists for a reasonablecalculation.

(4) But there is one feature in particular which deserves ourattention. It might have been supposed that competition be-tween expert professionals, possessing judgment and knowl-

149


edge beyond that of the average private investor, would cor-rect the vagaries of the ignorant individual left to himself.It happens, however, that the energies and skill of the pro-fessional investor and speculator are mainly occupied oth-erwise. For most of these persons are, in fact, largely con-cerned, not with making superior long-term forecasts of theprobable yield of an investment over its whole life, but withforeseeing changes in the conventional basis of valuation ashort time ahead of the general public. They are concerned,not with what an investment is really worth to a man whobuys it ‘for keeps’, but with what the market will value it at,under the influence of mass psychology, three months or ayear hence. Moreover, this behaviour is not the outcome of awrong-headed propensity. It is an inevitable result of an in-vestment market organised along the lines described. For it isnot sensible to pay 25 for an investment of which you believethe prospective yield to justify a value of 30, if you also believethat the market will value it at 20 three months hence.

Thus the professional investor is forced to concern himselfwith the anticipation of impending changes, in the news orin the atmosphere, of the kind by which experience showsthat the mass psychology of the market is most influenced.This is the inevitable result of investment markets organisedwith a view to so-called ‘liquidity’. Of the maxims of ortho-dox finance none, surely, is more anti-social than the fetish ofliquidity, the doctrine that it is a positive virtue on the partof investment institutions to concentrate their resources uponthe holding of ‘liquid’ securities. It forgets that there is nosuch thing as liquidity of investment for the community as awhole. The social object of skilled investment should be todefeat the dark forces of time and ignorance which envelopour future. The actual, private object of the most skilled in-vestment to-day is ‘to beat the gun’, as the Americans so wellexpress it, to outwit the crowd, and to pass the bad, or depre-ciating, half-crown to the other fellow.

This battle of wits to anticipate the basis of conventionalvaluation a few months hence, rather than the prospectiveyield of an investment over a long term of years, does not even

150

V.

require gulls amongst the public to feed the maws of the pro-fessional;—it can be played by professionals amongst them-selves. Nor is it necessary that anyone should keep his simplefaith in the conventional basis of valuation having any gen-uine long-term validity. For it is, so to speak, a game of Snap,of Old Maid, of Musical Chairs—a pastime in which he is vic-tor who says Snap neither too soon nor too late, who passedthe Old Maid to his neighbour before the game is over, whosecures a chair for himself when the music stops. These gamescan be played with zest and enjoyment, though all the play-ers know that it is the Old Maid which is circulating, or thatwhen the music stops some of the players will find themselvesunseated.

Or, to change the metaphor slightly, professional invest-ment may be likened to those newspaper competitions in whichthe competitors have to pick out the six prettiest faces from ahundred photographs, the prize being awarded to the com-petitor whose choice most nearly corresponds to the averagepreferences of the competitors as a whole; so that each com-petitor has to pick, not those faces which he himself finds pret-tiest, but those which he thinks likeliest to catch the fancy ofthe other competitors, all of whom are looking at the problemfrom the same point of view. It is not a case of choosing thosewhich, to the best of one’s judgment, are really the prettiest,nor even those which average opinion genuinely thinks theprettiest. We have reached the third degree where we devoteour intelligences to anticipating what average opinion expectsthe average opinion to be. And there are some, I believe, whopractise the fourth, fifth and higher degrees.

If the reader interjects that there must surely be large prof-its to be gained from the other players in the long run by askilled individual who, unperturbed by the prevailing pas-time, continues to purchase investments on the best genuinelong-term expectations he can frame, he must be answered,first of all, that there are, indeed, such serious-minded indi-viduals and that it makes a vast difference to an investmentmarket whether or not they predominate in their influenceover the game-players. But we must also add that there are

151


several factors which jeopardise the predominance of such in-dividuals in modern investment markets. Investment basedon genuine long-term expectation is so difficult to-day as tobe scarcely practicable. He who attempts it must surely leadmuch more laborious days and run greater risks than he whotries to guess better than the crowd how thc crowd will be-have; and, given equal intelligence, he may make more dis-astrous mistakes. There is no clear evidence from experiencethat the investment policy which is socially advantageous co-incides with that which is most profitable. It needs more in-telligence to defeat the forces of time and our ignorance ofthe future than to beat the gun. Moreover, life is not longenough;—human nature desires quick results, there is a pecu-liar zest in making money quickly, and remoter gains are dis-counted by the average man at a very high rate. The game ofprofessional investment is intolerably boring and over-exactingto anyone who is entirely exempt from the gambling instinct;whilst he who has it must pay to this propensity the appro-priate toll. Furthermore, an investor who proposes to ignorenear-term market fluctuations needs greater resources for safetyand must not operate on so large a scale, if at all, with bor-rowed money—a further reason for the higher return from thepastime to a given stock of intelligence and resources. Finallyit is the long-term investor, he who most promotes the publicinterest, who will in practice come in for most criticism, wher-ever investment funds are managed by committees or boardsor banks.4 For it is in the essence of his behaviour that heshould be eccentric, unconventional and rash in the eyes ofaverage opinion. If he is successful, that will only confirmthe general belief in his rashness; and if in the short run heis unsuccessful, which is very likely, he will not receive muchmercy. Worldly wisdom teaches that it is better for reputationto fail conventionally than to succeed unconventionally.

4The practice, usually considered prudent, by which an investment trustor an insurance office frequently calculates not only the income from its invest-ment portfolio but also its capital valuation in the market, may also tend todirect too much attention to short-term fluctuations in the latter.

152

VI.

(5) So far we have had chiefly in mind the state of confi-dence of the speculator or speculative investor himself andmay have seemed to be tacitly assuming that, if he himselfis satisfied with the prospects, he has unlimited commandover money at the market rate of interest. This is, of course,not the case. Thus we must also take account of the otherfacet of the state of confidence, namely, the confidence of thelending institutions towards those who seek to borrow fromthem, sometimes described as the state of credit. A collapsein the price of equities, which has had disastrous reactions onthe marginal efficiency of capital, may have been due to theweakening either of speculative confidence or of the state ofcredit. But whereas the weakening of either is enough to causea collapse, recovery requires the revival of both. For whilstthe weakening of credit is sufficient to bring about a collapse,its strengthening, though a necessary condition of recovery, isnot a sufficient condition.

VI

These considerations should not lie beyond the purview ofthe economist. But they must be relegated to their right per-spective. If I may be allowed to appropriate the term specula-tion for the activity of forecasting the psychology of the mar-ket, and the term enterprise for the activity of forecasting theprospective yield of assets over their whole life, it is by nomeans always the case that speculation predominates over en-terprise. As the organisation of investment markets improves,the risk of the predominance of speculation does, however, in-crease. In one of the greatest investment markets in the world,namely, New York, the influence of speculation (in the abovesense) is enormous. Even outside the field of finance, Ameri-cans are apt to be unduly interested in discovering what aver-age opinion believes average opinion to be; and this nationalweakness finds its nemesis in the stock market. It is rare, oneis told, for an American to invest, as many Englishmen still do,‘for income’; and he will not readily purchase an investment

153


except in the hope of capital appreciation. This is only anotherway of saying that, when he purchases an investment, theAmerican is attaching his hopes, not so much to its prospec-tive yield, as to a favourable change in the conventional basisof valuation, i.e. that he is, in the above sense, a speculator.Speculators may do no harm as bubbles on a steady streamof enterprise. But the position is serious when enterprise be-comes the bubble on a whirlpool of speculation. When thecapital development of a country becomes a by-product ofthe activities of a casino, the job is likely to be ill-done. Themeasure of success attained by Wall Street, regarded as an in-stitution of which the proper social purpose is to direct newinvestment into the most profitable channels in terms of futureyield, cannot be claimed as one of the outstanding triumphsof laissez-faire capitalism—which is not surprising, if I am rightin thinking that the best brains of Wall Street have been in factdirected towards a different object.

These tendencies are a scarcely avoidable outcome of ourhaving successfully organised ‘liquid’ investment markets. Itis usually agreed that casinos should, in the public interest, beinaccessible and expensive. And perhaps the same is true ofstock exchanges. That the sins of the London Stock Exchangeare less than those of Wall Street may be due, not so muchto differences in national character, as to the fact that to theaverage Englishman Throgmorton Street is, compared withWall Street to the average American, inaccessible and very ex-pensive. The jobber’s ‘turn’, the high brokerage charges andthe heavy transfer tax payable to the Exchequer, which attenddealings on the London Stock Exchange, sufficiently dimin-ish the liquidity of the market (although the practice of fort-nightly accounts operates the other way) to rule out a largeproportion of the transaction characteristic of Wall Street.5 Theintroduction of a substantial government transfer tax on all

5It is said that, when Wall Street is active, at least a half of the purchasesor sales of investments are entered upon with an intention on the part of thespeculator to reverse them the same day. This is often true of the commodityexchanges also.

154

VI.

transactions might prove the most serviceable reform avail-able, with a view to mitigating the predominance of specula-tion over enterprise in the United States.

The spectacle of modern investment markets has some-times moved me towards the conclusion that to make the pur-chase of an investment permanent and indissoluble, like mar-riage, except by reason of death or other grave cause, might bea useful remedy for our contemporary evils. For this wouldforce the investor to direct his mind to the long-term prospectsand to those only. But a little consideration of this expedientbrings us up against a dilemma, and shows us how the liq-uidity of investment markets often facilitates, though it some-times impedes, the course of new investment. For the fact thateach individual investor flatters himself that his commitmentis ‘liquid’ (though this cannot be true for all investors collec-tively) calms his nerves and makes him much more willing torun a risk. If individual purchases of investments were ren-dered illiquid, this might seriously impede new investment,so long as alternative ways in which to hold his savings areavailable to the individual. This is the dilemma. So long asit is open to the individual to employ his wealth in hoardingor lending money, the alternative of purchasing actual capitalassets cannot be rendered sufficiently attractive (especially tothe man who does not manage the capital assets and knowsvery little about them), except by organising markets whereinthese assets can be easily realised for money.

The only radical cure for the crises of confidence which af-flict the economic life of the modern world would be to allowthe individual no choice between consuming his income andordering the production of the specific capital-asset which,even though it be on precarious evidence, impresses him asthe most promising investment available to him. It mightbe that, at times when he was more than usually assailed bydoubts concerning the future, he would turn in his perplex-ity towards more consumption and less new investment. Butthat would avoid the disastrous, cumulative and far-reachingrepercussions of its being open to him, when thus assailed bydoubts, to spend his income neither on the one nor on the

155


other.Those who have emphasised the social dangers of the hoard-

ing of money have, of course, had something similar to theabove in mind. But they have overlooked the possibility thatthe phenomenon can occur without any change, or at least anycommensurate change, in the hoarding of money.

VII

Even apart from the instability due to speculation, there isthe instability due to the characteristic of human nature thata large proportion of our positive activities depend on spon-taneous optimism rather than on a mathematical expectation,whether moral or hedonistic or economic. Most, probably, ofour decisions to do something positive, the full consequencesof which will be drawn out over many days to come, can onlybe taken as a result of animal spirits—of a spontaneous urgeto action rather than inaction, and not as the outcome of aweighted average of quantitative benefits multiplied by quan-titative probabilities. Enterprise only pretends to itself to bemainly actuated by the statements in its own prospectus, how-ever candid and sincere. Only a little more than an expeditionto the South Pole, is it based on an exact calculation of ben-efits to come. Thus if the animal spirits are dimmed and thespontaneous optimism falters, leaving us to depend on noth-ing but a mathematical expectation, enterprise will fade anddie;—though fears of loss may have a basis no more reason-able than hopes of profit had before.

It is safe to say that enterprise which depends on hopesstretching into the future benefits the community as a whole.But individual initiative will only be adequate when reason-able calculation is supplemented and supported by animalspirits, so that the thought of ultimate loss which often over-takes pioneers, as experience undoubtedly tells us and them,is put aside as a healthy man puts aside the expectation ofdeath.

This means, unfortunately, not only that slumps and de-

156

VIII.

pressions are exaggerated in degree, but that economic pros-perity is excessively dependent on a political and social at-mosphere which is congenial to the average business man. Ifthe fear of a Labour Government or a New Deal depressesenterprise, this need not be the result either of a reasonablecalculation or of a plot with political intent;—it is the mereconsequence of upsetting the delicate balance of spontaneousoptimism. In estimating the prospects of investment, we musthave regard, therefore, to the nerves and hysteria and even thedigestions and reactions to the weather of those upon whosespontaneous activity it largely depends.

We should not conclude from this that everything dependson waves of irrational psychology. On the contrary, the stateof long-term expectation is often steady, and, even when it isnot, the other factors exert their compensating effects. We aremerely reminding ourselves that human decisions affectingthe future, whether personal or political or economic, cannotdepend on strict mathematical expectation, since the basis formaking such calculations does not exist; and that it is our in-nate urge to activity which makes the wheels go round, ourrational selves choosing between the alternatives as best weare able, calculating where we can, but often falling back forour motive on whim or sentiment or chance.

VIII

There are, moreover, certain important factors which some-what mitigate in practice the effects of our ignorance of thefuture. Owing to the operation of compound interest com-bined with the likelihood of obsolescence with the passageof time, there are many individual investments of which theprospective yield is legitimately dominated by the returns ofthe comparatively near future. In the case of the most im-portant class of very long-term investments, namely build-ings, the risk can be frequently transferred from the investorto the occupier, or at least shared between them, by means oflong-term contracts, the risk being outweighed in the mind of

157


the occupier by the advantages of continuity and security oftenure. In the case of another important class of long-term in-vestments, namely public utilities, a substantial proportion ofthe prospective yield is practically guaranteed by monopolyprivileges coupled with the right to charge such rates as willprovide a certain stipulated margin. Finally there is a grow-ing class of investments entered upon by, or at the risk of;public authorities, which are frankly influenced in making theinvestment by a general presumption of there being prospec-tive social advantages from the investment, whatever its com-mercial yield may prove to be within a wide range, and with-out seeking to be satisfied that the mathematical expectationof the yield is at least equal to the current rate of interest,—though the rate which the public authority has to pay maystill play a decisive part in determining the scale of invest-ment operations which it can afford.

Thus after giving full weight to the importance of the influ-ence of short-period changes in the state of long-term expec-tation as distinct from changes in the rate of interest, we arestill entitled to return to the latter as exercising, at any rate,in normal circumstances, a great, though not a decisive, in-fluence on the rate of investment. Only experience, however,can show how far management of the rate of interest is ca-pable of continuously stimulating the appropriate volume ofinvestment.

For my own part I am now somewhat sceptical of the suc-cess of a merely monetary policy directed towards influencingthe rate of interest. I expect to see the State, which is in a po-sition to calculate the marginal efficiency of capital-goods onlong views and on the basis of the general social advantage,taking an ever greater responsibility for directly organising in-vestment; since it seems likely that the fluctuations in the mar-ket estimation of the marginal efficiency of different types ofcapital, calculated on the principles I have described above,will be too great to be offset by any practicable changes in therate of interest.

158

CHAPTER 13

THE GENERAL THEORY OF THE RATE OFINTEREST

I

We have shown in chapter 11 that, whilst there are forces caus-ing the rate of investment to rise or fall so as to keep themarginal efficiency of capital equal to the rate of interest, yetthe marginal efficiency of capital is, in itself; a different thingfrom the ruling rate of interest. The schedule of the marginalefficiency of capital may be said to govern the terms on whichloanable funds are demanded for the purpose of new invest-ment; whilst the rate of interest governs the terms on whichfunds are being currently supplied. To complete our theory,therefore, we need to know what determines the rate of inter-est.

In chapter 14 and its Appendix we shall consider the an-swers to this question which have been given hitherto. Broadlyspeaking, we shall find that they make the rate of interest todepend on the interaction of the schedule of the marginal ef-ficiency of capital with the psychological propensity to save.But the notion that the rate of interest is the balancing factorwhich brings the demand for saving in the shape of new in-vestment forthcoming at a iven rate of interest into equalitywith the supply of saving which results at that rate of inter-est from the community’s psychological propensity to save,breaks down as soon as we perceive that it is impossible todeduce the rate of interest merely from a knowledge of thesetwo factors. What, then, is our own answer to this question?

159

13. THE GENERAL THEORY OF THE RATE OF INTEREST

II

The psychological time-preferences of an individual requiretwo distinct sets of decisions to carry them out completely.The first is concerned with that aspect of time-preference whichI have called the propensity to consume, which, operating underthe influence of the various motives set forth in Book III, de-termines for each individual how much of his income he willconsume and how much he will reserve in some form of com-mand over future consumption.

But this decision having been made, there is a further deci-sion which awaits him, namely, in what form he will hold thecommand over future consumption which he has reserved,whether out of his current income or from previous savings.Does he want to hold it in the form of immediate, liquid com-mand (i.e. in money or its equivalent)? Or is he preparedto part with immediate command for a specified or indefiniteperiod, leaving it to future market conditions to determine onwhat terms he can, if necessary, convert deferred commandover specific goods into immediate command over goods ingeneral? In other words, what is the degree of his liquidity-preference—where an individual’s liquidity-preference is givenby a schedule of the amounts of his resources, valued in termsof money or of wage-units, which he will wish to retain in theform of money in different sets of circumstances?

We shall find that the mistake in the accepted theories ofthe rate of interest lies in their attempting to derive the rate ofinterest from the first of these two constituents of psycholog-ical time-preference to the neglect of the second; and it is thisneglect which we must endeavour to repair.

It should be obvious that the rate of interest cannot be areturn to saving or waiting as such. For if a man hoards hissavings in cash, he earns no interest, though he saves just asmuch as before. On the contrary, the mere definition of therate of interest tells us in so many words that the rate of inter-est is the reward for parting with liquidity for a specified pe-riod. For the rate of interest is, in itself; nothing more than theinverse proportion between a sum of money and what can be

160

II.

obtained for parting with control over the money in exchangefor a debt1 for a stated period of time.2

Thus the rate of interest at any time, being the reward forparting with liquidity, is a measure of the unwillingness ofthose who possess money to part with their liquid controlover it. The rate of interest is not the ‘price’ which bringsinto equilibrium the demand for resources to invest with thereadiness to abstain from present consumption. It is the ’price’which equilibrates the desire to hold wealth in the form ofcash with the available quantity of cash;—which implies thatif the rate of interest were lower, i.e. if the reward for part-ing with cash were diminished, the aggregate amount of cashwhich the public would wish to hold would exceed the avail-able supply, and that if the rate of interest were raised, therewould be a surplus of cash which no one would be willingto hold. If this explanation is correct, the quantity of money isthe other factor, which, in conjunction with liquidity-preference,determines the actual rate of interest in given circumstances.Liquidity-preference is a potentiality or functional tendency,which fixes the quantity of money which the public will holdwhen the rate of interest is given; so that if r is the rate of inter-est, M the quantity of money and L the function of liquidity-preference, we have M = L(r). This is where, and how, thequantity of money enters into the economic scheme.

1Without disturbance to this definition, we can draw the line between“money” and “debts” at whatever point is most convenient for handling a par-ticular problem. For example, we can treat as money any command over generalpurchasing power which the owner has not parted with for a period in excessof three months, and as debt what cannot be recovered for a longer period thanthis; or we can substitute for “three months” one month or three days or threehours or any other period; or we can exclude from money whatever is not legaltender on the spot. It is often convenient in practice to include in money time-deposits with banks and, occasionally, even such instruments as (e.g.) treasurybills. As a rule, I shall, as in my Treatise on Money, assume that money is coex-tensive with bank deposits.

2In general discussion, as distinct from specific problems where the periodof the debt is expressly specified, it is convenient to mean by the rate of interestthe complex of the various rates of interest current for different periods of time,i.e. for debts of different maturities.

161


At this point, however, let us turn back and consider whysuch a thing as liquidity-preference exists. In this connectionwe can usefully employ the ancient distinction between theuse of money for the transaction of current business and itsuse as a store of wealth. As regards the first of these twouses, it is obvious that up to a point it is worth while to sac-rifice a certain amount of interest for the convenience of liq-uidity. But, given that the rate of interest is never negative,why should anyone prefer to hold his wealth in a form whichyields little or no interest to holding it in a form which yieldsinterest (assuming, of course, at this stage, that the risk of de-fault is the same in respect of a bank balance as of a bond)?A full explanation is complex and must wait for chapter 15.There is, however, a necessary condition failing which the ex-istence of a liquidity-preference for money as a means of hold-ing wealth could not exist.

This necessary condition is the existence of uncertainty asto the future of the rate of interest, i.e. as to the complex ofrates of interest for varying maturities which will rule at fu-ture dates. For if the rates of interest ruling at all future timescould be foreseen with certainty, all future rates of interestcould be inferred from the present rates of interest for debtsof different maturities, which would be adjusted to the knowl-edge of the future rates. For example, if 1dr is the value ln thepresent year 1 of £1 deferred r years and it is known that ndr

will be the value in the year n of £1 deferred r years from thatdate, we have

ndr =n dn+r/1dn

whence it follows that the rate at which any debt can be turnedinto cash n years hence is given by two out of the complexof current rates of interest. If the current rate of interest ispositive for debts of every maturity, it must always be moreadvantageous to purchase a debt than to hold cash as a storeof wealth.

If, on the contrary, the future rate of interest is uncertain wecannot safely infer that ndr will prove to be equal to 1dn+r/1dn

when the time comes. Thus if a need for liquid cash may con-

162

II.

ceivably arise before the expiry of n years, there is a risk of aloss being incurred in purchasing a long-term debt and subse-quently turning it into cash, as compared with holding cash.The actuarial profit or mathematical expectation of gain calcu-lated in accordance with the existing probabilities—if it can beso calculated, which is doubtful—must be sufficient to com-pensate for the risk of disappointment.

There is, moreover, a further ground for liquidity-preferencewhich results from the existence of uncertainty as to the fu-ture of the rate of interest, provided that there is an organlsedmarket for dealing in debts. For different people will esti-mate the prospects differently and anyone who differs fromthe predominant opinion as expressed in market quotationsmay have a good reason for keeping liquid resources in orderto profit, if he is right, from its turning out in due course thatthe 1dr’s were in a mistaken relationship to one another.3

This is closely analogous to what we have already discussedat some length in connection with the marginal efficiency ofcapital. Just as we found that the marginal efficiency of cap-ital is fixed, not by the ‘best’ opinion, but by the market val-uation as determined by mass psychology, so also expecta-tions as to the future of the rate of interest as fixed by masspsychology have their reactions on liquidity-preference;—butwith this addition that the individual, who believes that fu-ture rates of interest will be above the rates assumed by themarket, has a reason for keeping actual liquid cash4, whilstthe individual who differs from the market in the other direc-tion will have a motive for borrowing money for short periodsin order to purchase debts of longer term. The market price

3This is the same point as I discussed in my Treatise on Money under thedesignation of the two views and the “bull-bear” position.

4It might be thought that, in the same way, an individual, who believed thatthe prospective yield of investments will be below what the market is expecting,will have a sufficient reason for holding liquid cash. But this is not the case. Hehas a sufficient reason for holding cash or debts in preference to equities; but thepurchase of debts will be a preferable alternative to holding cash, unless he alsobelieves that the future rate of interest will prove to be higher than the marketis supposing.

163


will be fixed at the point at which the sales of the ‘bears’ andthe purchases of the ‘bulls’ are balanced.

The three divisions of liquidity-preference which we havedistinguished above may be defined as depending on (i) thetransactions-motive, i.e. the need of cash for the current trans-action of personal and business exchanges; (ii) the precautionary-motive, i.e. the desire for security as to the future cash equiv-alent of a certain proportion of total resources; and (iii) thespeculative-motive, i.e. the object of securing profit from know-ing better than the market what the future will bring forth. Aswhen we were discussing the marginal efficiency of capital,the question of the desirability of having a highly organisedmarket for dealing with debts presents us with a dilemma.For, in the absence of an organised market, liquidity-preferencedue to the precautionary-motive would be greatly increased;whereas the existence of an organised market gives an oppor-tunity for wide fluctuations in liquidity-preference due to thespeculative-motive.

It may illustrate the argument to point out that, if the liquidity-preferences due to the transactions-motive and the precautionary-motive are assumed to absorb a quantity of cash which is notvery sensitive to changes in the rate of interest as such andapart from its reactions on the level of income, so that the to-tal quantity of money, less this quantity, is available for satis-fying liquidity-preferences due to the speculative-motive, therate of interest and the price of bonds have to be fixed at thelevel at which the desire on the part of certain individuals tohold cash (because at that level they feel ‘bearish’ of the fu-ture of bonds) is exactly equal to the amount of cash availablefor the speculative-motive. Thus each increase in the quan-tity of money must raise the price of bonds sufficiently to ex-ceed the expectations of some ‘bull’ and so influence him tosell his bond for cash and join the ‘bear’ brigade. If, however,there is a negligible demand for cash from the speculative-motive except for a short transitional interval, an increase inthe quantity of money will have to lower the rate of inter-est almost forthwith, in whatever degree is necessary to raiseemployment and the wage-unit sufficiently to cause the addi-

164

II.

tional cash to be absorbed by the transactions-motive and theprecautionary-motive.

As a rule, we can suppose that the schedule of liquidity-preference relating the quantity of money to the rate of interestis given by a smooth curve which shows the rate of interestfalling as the quantity of money is increased. For there areseveral different causes all leading towards this result.

In the first place, as the rate of interest falls, it is likely, cet.par., that more money will be absorbed by liquidity-preferencesdue to the transactions-motive. For if the fall in the rate ofinterest increases the national income, the amount of moneywhich it is convenient to keep for transactions will be increasedmore or less proportionately to the increase in income; whilst,at the same time, the cost of the convenience of plenty of readycash in terms of loss of interest will be diminished. Unlesswe measure liquidity-preference in terms of wage-units ratherthan of money (which is convenient in some contexts), similarresults follow if the increased employment ensuing on a fallin the rate of interest leads to an increase of wages, i.e. to anincrease in the money value of the wage-unit. In the secondplace, every fall in the rate of interest may, as we have justseen, increase the quantity of cash which certain individualswill wish to hold because their views as to the future of therate of interest differ from the market views.

Nevertheless, circumstances can develop in which even alarge increase in the quantity of money may exert a compar-atively small influence on the rate of interest. For a large in-crease in the quantity of money may cause so much uncer-tainty about the future that liquidity-preferences due to theprecautionary-motive may be strengthened; whilst opinionabout the future of the rate of interest may be so unanimousthat a small change in present rates may cause a mass move-ment into cash. It is interesting that the stability of the sys-tem and its sensitiveness to changes in the quantity of moneyshould be so dependent on the existence of a variety of opin-ion about what is uncertain. Best of all that we should knowthe future. But if not, then, if we are to control the activityof the economic system by changing the quantity of money, it

165


is important that opinions should differ Thus this method ofcontrol is more precarious in the United States, where every-one tends to hold the same opinion at the same time, than inEngland where differences of opinion are more usual.

III

We have now introduced money into our causal nexus for thefirst time, and we are able to catch a first glimpse of the wayin which changes in the quantity of money work their wayinto the economic system. If, however, we are tempted toassert that money is the drink which stimulates the systemto activity, we must remind ourselves that there may be sev-eral slips between the cup and the lip. For whilst an increasein the quantity of money may be expected, cet. par., to re-duce the rate of interest, this will not happen if the liquidity-preferences of the public are increasing more than the quan-tity of money; and whilst a decline in the rate of interest maybe expected, cet. par., to increase the volume of investment,this will not happen if the schedule of the marginal efficiencyof capital is falling more rapidly than the rate of intere t; andwhilst an increase in the volume of investment may be ex-pected, cet. par., to increase employment, this may not happenif the propensity to consume is falling off. Finally, if employ-ment increases, prices will rise in a degree partly governedby the shapes of the physical supply functions, and partly bythe liability of the wage-unit to rise in terms of money. Andwhen output has increased and prices have risen, the effect ofthis on liquidity-preference will be to increase the quantity ofmoney necessary to maintain a given rate of interest.

IV

Whilst liquidity-preference due to the speculative-motive cor-responds to what in my Treatise on Money I called ‘the stateof bearishness’, it is by no means the same thing. For ‘bear-

166

V.

ishness’ is there defined as the functional relationship, not be-tween the rate of interest (or price of debts) and the quantityof money, but between the price of assets and debts, taken to-gether, and the quantity of money. This treatment, however,involved a confusion between results due to a change in therate of interest and those due to a change in the schedule ofthe marginal efficiency of capital, which I hope I have hereavoided.

V

The concept of hoarding may be regarded as a first approx-imation to the concept of liquidity-preference. Indeed if wewere to substitute ‘propensity to hoard’ for ‘hoarding’, it wouldcome to substantially the same thing. But if we mean by ‘hoard-ing’ an actual increase in cash-holding, it is an incompleteidea—and seriously misleading if it causes us to think of ‘hoard-ing’ and ‘not-hoarding’ as simple alternatives. For the deci-sion to hoard is not taken absolutely or without regard to theadvantages offered for parting with liquidity;—it results froma balancing of advantages, and we have, therefore, to knowwhat lies in the other scale. Moreover it is impossible for theactual amount of hoarding to change as a result of decisionson the part of the public, so long as we mean by ‘hoarding’ theactual holding of cash. For the amount of hoarding must beequal to the quantity of money (or—on some definitions—tothe quantity of money minus what is required to satisfy thetransactions-motive); and the quantity of money is not deter-mined by the public. All that the propensity of the public to-wards hoarding can achieve is to determine the rate of interestat which the aggregate desire to hoard becomes equal to theavailable cash. The habit of overlooking the relation of therate of interest to hoarding may be a part of the explanationwhy interest has been usually regarded as the reward of not-spending, whereas in fact it is the reward of not-hoarding.

167

CHAPTER 14

THE CLASSICAL THEORY OF THE RATE OFINTEREST

I

What is the classical theory of the rate of interest? It is some-thing upon which we have all been brought up and which wehave accepted without much reserve until recently. Yet I findit difficult to state it precisely or to discover an explicit accountof it in the leading treatises of the modern classical school.1

It is fairly clear, however, that this tradition has regardedthe rate of interest as the factor which brings the demand forinvestment and the willingness to save into equilibrium withone another. Investment represents the demand for investibleresources and saving represents the supply, whilst the rate ofinterest is the ’price’ of investible resources at which the twoare equated. Just as the price of a commodity is necessarilyfixed at that point where the demand for it is equal to the sup-ply, so the rate of interest necessarily comes to rest under theplay of market forces at the point where the amount of invest-ment at that rate of interest is equal to the amount of saving atthat rate.

The above is not to be found in Marshall’s Principles in somany words. Yet his theory seems to be this, and it is what Imyself was brought up on and what I taught for many yearsto others. Take, for example, the following passage from hisPrinciples: ‘Interest, being the price paid for the use of capitalin any market, tends towards an equilibrium level such thatthe aggregate demand for capital in that market, at that rate

1See the Appendix to this Chapter for an abstract of what I have been ableto find.

169

14. THE CLASSICAL THEORY OF THE RATE OF INTEREST

of interest, is equal to the aggregate stock forthcoming at thatrate’.2 Or again in Professor Cassel’s Nature and Necessity of In-terest it is explained that investment constitutes the ‘demandfor waiting’ and saving the ‘supply of waiting’, whilst inter-est is a ‘price’ which serves, it is implied, to equate the two,though here again I have not found actual words to quote.Chapter vi of Professor Carver’s Distribution of Wealth clearlyenvisages interest as the factor which brings into equilibriumthe marginal disutility of waiting with the marginal produc-tivity of capital.3 Sir Alfred Flux (Economic Principles, p. 95)writes: ‘If there is justice in the contentions of our generaldiscussion, it must be admitted that an automatic adjustmenttakes place between saving and the opportunities for employ-ing capital profitably. . . Saving will not have exceeded itspossibilities of usefulness. . . so long as the rate of net interestis in excess of zero.’ Professor Taussig (Principles, vol. ii. p.29) draws a supply curve of saving and a demand curve rep-resenting ‘the diminishing productiveness of the several in-stalments of capital’, having previously stated (p.20) that ‘therate of interest settles at a point where the marginal productiv-ity of capital suffices to bring out the marginal instalment ofsaving’.4 Walras, in Appendix I (III) of his Éléments d’économiepure, where he deals with ‘l’échange d’épargnes contre capi-taux neufs’, argues expressly that, corresponding to each pos-sible rate of interest, there is a sum which individuals will

2Cf. Appendix p. 186 below for a further discussion of this passage.3Prof. Carver’s discussion of Interest is difficult to follow (1) through his in-

consistency a to whether he means by “marginal productivity of capital” quan-tity of marginal product or value of marginal product, and (2) through his mak-ing no attempt to define quantity of capital.

4In a very recent discussion of these problems (“Capital, Time and the In-terest Rate”, by Prof. F. H. Knight, Economica, August 1932), a discussion whichcontains many interesting and profound observations on the nature of capital,and confirms the soundness of the Marshallian tradition as to the uselessness ofthe Böhm-Bawerkian analysis, the theory of interest is given precisely in the tra-ditional, classical mould. Equilibrium in the field of capital production means,according to Prof. Knight, “such a rate of interest that savings flow into themarket at precisely the same time-rate or speed as they flow into investmentproducing the same net rate of return as that which is paid savers for their use”.

170

I.

save and also a sum which they will invest in new capital as-sets, that these two aggregates tend to equality with one an-other, and that the rate of interest is the variable which bringsthem to equality; so that the rate of interest is fixed at the pointwhere saving, which represents the supply of new capital, isequal to the demand for it. Thus he is strictly in the classicaltradition.

Certainly the ordinary man—banker, civil servant or politi-cian—brought up on the traditional theory, and the trainedeconomist also, has carried away with him the idea that when-ever an individual performs an act of saving he has done some-thing which automatically brings down the rate of interest,that this automatically stimulates the output of capital, andthat the fall in the rate of interest is just so much as is necessaryto stimulate the output of capital to an extent which is equalto the increment of saving; and, further, that this is a self-regulatory process of adjustment which takes place withoutthe necessity for any special intervention or grandmotherlycare on the part of the monetary authority. Similarly—and thisis an even more general belief, even to-day—each additionalact of investment will necessarily raise the rate of interest, if itis not offset by a change in the readiness to save.

Now the analysis of the previous chapters will have madeit plain that this account of the matter must be erroneous. Intracing to its source the reason for the difference of opinion,let us, however, begin with the matters which are agreed.

Unlike the neo-classical school, who believe that savingand investment can be actually unequal, the classical schoolproper has accepted the view that they are equal. Marshall,for example, surely believed, although he did not expresslysay so, that aggregate saving and aggregate investment arenecessarily equal. Indeed, most members of the classical schoolcarried this belief much too far; since they held that every actof increased saving by an individual necessarily brings intoexistence a corresponding act of increased investment. Noris there any material difference, relevant in this context, be-tween my schedule of the marginal efficiency of capital or in-vestment demand-schedule and the demand curve for capital

171


contemplated by some of the classical writers who have beenquoted above. When we come to the propensity to consumeand its corollary the propensity to save, we are nearer to a dif-ference of opinion, owing to the emphasis which they haveplaced on the influence of the rate of interest on the propen-sity to save. But they would, presumably, not wish to denythat the level of income also has an important influence onthe amount saved; whilst I, for my part, would not deny thatthe rate of interest may perhaps have an influence (thoughperhaps not of the kind which they suppose) on the amountsaved out of a given income. All these points of agreement canbe summed up in a proposition which the classical schoolwould accept and I should not dispute; namely, that, if thelevel of income is assumed to be given, we can infer that thecurrent rate of interest must lie at the point where the demandcurve for capital corresponding to different rates of interestcuts the curve of the amounts saved out of the given incomecorresponding to different rates of interest.

But this is the point at which definite error creeps into theclassical theory. If the classical school merely inferred fromthe above proposition that, given the demand curve for capitaland the influence of changes in the rate of interest on the readi-ness to save out of given incomes, the level of income and therate of interest must be uniquely correlated, there would benothing to quarrel with. Moreover, this proposition wouldlead naturally to another proposition which embodies an im-portant truth; namely, that, if the rate of interest is given aswell as the demand curve for capital and the influence of therate of interest on the readiness to save out of given levels ofincome, the level of income must be the factor which bringsthe amount saved to equality with the amount invested. But,in fact, the classical theory not merely neglects the influenceof changes in the level of income, but involves formal error.

For the classical theory, as can be seen from the above quo-tations, assumes that it can then proceed to consider the effecton the rate of interest of (e.g.) a shift in the demand curve forcapital, without abating or modifying its assumption as to theamount of the given income out of which the savings are to

172

I.

be made. The independent variables of the classical theory ofthe rate of interest are the demand curve for capital and theinfluence of the rate of interest on the amount saved out ofa given income; and when (e.g.) the demand curve for capi-tal shifts, the new rate of interest, according to this theory, isgiven by the point of intersection between the new demandcurve for capital and the curve relating the rate of interest tothe amounts which will be saved out of the given income. Theclassical theory of the rate of interest seems to suppose that, ifthe demand curve for capital shifts or if the curve relating therate of interest to the amounts saved out of a given incomeshifts or if both these curves shift, the new rate of interest willbe given by the point of intersection of the new positions ofthe two curves. But this is a nonsense theory. For the as-sumption that income is constant is inconsistent with the as-sumption that these two curves can shift independently of oneanother. If either of them shift, then, in general, income willchange; with the result that the whole schematism based onthe assumption of a given income breaks down. The positioncould only be saved by some complicated assumption provid-ing for an automatic change in the wage-unit of an amountjust sufficient in its effect on liquidity-preference to establish arate of interest which would just offset the supposed shift, soas to leave output at the same level as before. In fact, there isno hint to be found in the above writers as to the necessity forany such assumption; at the best it would be plausible onlyin relation to long-period equilibrium and could not form thebasis of a short-period theory; and there is no ground for sup-posing it to hold even in the long-period. In truth, the clas-sical theory has not been alive to the relevance of changes inthe level of income or to the possibility of the level of incomebeing actually a function of the rate of the investment.

The above can be illustrated by a diagram5 as follows:

5This diagram was suggested to me by Mr. R. F. Harrod. Cf. also a partlysimilar schematism by Mr. D. H. Robertson, Economic Journal, December 1934,p. 652.

173


In this diagram the amount of investment (or saving) Iis measured vertically, and the rate of interest r horizontally.X1X

′1 is the first position of the investment demand-schedule,

and X2X′2 is a second position of this curve. The curve Y1 re-

lates the amounts saved out of an income Y1 to various levelsof the rate of interest, the curves Y2, Y3, etc., being the corre-sponding curves for levels of income Y2, Y3, etc. Let us sup-pose that the curve Y1 is the Y -curve consistent with an invest-ment demand-scheduleX1X

′1 and a rate of interest r1. Now if

the investment demand-schedule shifts from X1X′1 to X2X

′2,

income will, in general, shift also. But the above diagram doesnot contain enough data to tell us what its new value will be;and, therefore, not knowing which is the appropriate Y -curve,we do not know at what point the new investment demand-schedule will cut it. If, however, we introduce the state ofliquidity-preference and the quantity of money and these be-tween them tell us that the rate of interest is r2, then the wholeposition becomes determinate. For the Y -curve which inter-sects X2X

′2 at the point vertically above r2, namely, the curve

Y2, will be the appropriate curve. Thus the X-curve and theY -curves tell us nothing about the rate of interest. They only

174

I.

tell us what income will be, if from some other source we cansay what the rate of interest is. If nothing has happened tothe state of liquidity-preference and the quantity of money, sothat the rate of interest is unchanged, then the curve Y ′

2 whichintersects the new investment demand-schedule vertically be-low the point where the curve Y1 intersected the old invest-ment demand-schedule will be the appropriate Y -curve, andY ′

2 will be the new level of income.Thus the functions used by the classical theory, namely,

the response of investment and the response of the amountsaved out of a given income to change in the rate of interest,do not furnish material for a theory of the rate of interest; butthey could be used to tell us what the level of income will be,given (from some other source) the rate of interest; and, alter-natively, what the rate of interest will have to be, if the levelof income is to be maintained at a given figure (e.g. the levelcorresponding to full employment).

The mistake originates from regarding interest as the re-ward for waiting as such, instead of as the reward for not-hoarding; just as the rates of return on loans or investments in-volving different degrees of risk, are quite properly regardedas the reward, not of waiting as such, but of running the risk.There is, in truth, no sharp line between these and the so-called ‘pure’ rate of interest, all of them being the reward forrunning the risk of uncertainty of one kind or another. Onlyln the event of money being used solely for transactions andnever as a store of value, would a different theory become ap-propriate.6

There are, however, two familiar points which might, per-haps, have warned the classical school that something waswrong. In the first place, it has been agreed, at any rate sincethe publication of Professor Cassel’s Nature and Necessity of In-terest, that it is not certain that the sum saved out of a givenincome necessarily increases when the rate of interest is in-creased; whereas no one doubts that the investment demand-schedule falls with a rising rate of interest. But if the Y -curves


175


and the X-curves both fall as the rate of interest rises, thereis no guarantee that a given Y -curve will intersect a givenX-curve anywhere at all. This suggests that it cannot be theY -curve and the X-curve alone which determine the rate ofinterest.

In the second place, it has been usual to suppose that anincrease in the quantity of money has a tendency to reducethe rate of interest, at any rate in the first instance and in theshort period. Yet no reason has been given why a changein the quantity of money should affect either the investmentdemand-schedule or the readiness to save out of a given in-come. Thus the classical school have had quite a different the-ory of the rate of interest in volume I dealing with the the-ory of value from what they have had in volume II dealingwith the theory of money. They have seemed undisturbed bythe conflict and have made no attempt, so far as I know, tobuild a bridge between the two theories. The classical schoolproper, that is to say; since it is the attempt to build a bridge onthe part of the neo-classical school which has led to the worstmuddles of all. For the latter have inferred that there must betwo sources of supply to meet the investment demand-schedule;namely, savings proper, which are the savings dealt with bythe classical school, plus the sum made available by any in-crease in the quantity of money (this being balanced by somespecies of levy on the public, called ‘forced saving’ or the like).This leads on to the idea that there is a ‘natural’ or ‘neutral’7

or ‘equilibrium’ rate of interest, namely, that rate of interestwhich equates investment to classical savings proper with-out any addition from ‘forced savings’; and finally to what,assuming they are on the right track at the start, is the mostobvious solution of all, namely, that, if the quantity of moneycould only be kept constant in all circumstances, none of thesecomplications would arise, since the evils supposed to resultfrom the supposed excess of investment over savings proper

7The “neutral” rate of interest of contemporary economists is different bothfrom the “natural” rate of Böhm-Bawerk and from the “natural” rate of Wick-sell.

176

I.

would cease to be possible. But at this point we are in deepwater. ‘The wild duck has dived down to the bottom—as deepas she can get—and bitten fast hold of the weed and tangleand all the rubbish that is down there, and it would need anextraordinarily clever dog to dive after and fish her up again.’

Thus the traditional analysis is faulty because it has failedto isolate correctly the independent variables of the system.Saving and investment are the determinates of the system, notthe determinants. They are the twin results of the system’sdeterminants, namely, the propensity to consume, the sched-ule of the marginal efficiency of capital and the rate of inter-est. These determinants are, indeed, themselves complex andeach is capable of being affected by prospective changes in theothers. But they remain independent in the sense that theirvalues cannot be inferred from one another. The traditionalanalysis has been aware that saving depends on income but ithas overlooked the fact that income depends on investment,in such fashion that, when investment changes, income mustnecessarily change in just that degree which is necessary tomake the change in saving equal to the change in investment.

Nor are those theories more successful which attempt tomake the rate of interest depend on ‘the marginal efficiency ofcapital’. It is true that in equilibrium the rate of interest will beequal to the marginal efficiency of capital, since it will be prof-itable to increase (or decrease) the current scale of investmentuntil the point of equality has been reached. But to make thisinto a theory of the rate of interest or to derive the rate of inter-est from it involves a circular argument, as Marshall discov-ered after he had got half-way into giving an account of therate of interest along these lines.8 For the ‘marginal efficiencyof capital’ partly depends on the scale of current investment,and we must already know the rate of interest before we cancalculate what this scale will be. The significant conclusion isthat the output of new investment will be pushed to the pointat which the marginal efficiency of capital becomes equal tothe rate of interest; and what the schedule of the marginal ef-

8See the Appendix to this Chapter.

177


ficiency of capital tells us, is, not what the rate of interest is,but the point to which the output of new investment will bepushed, given the rate of interest.

The reader will readily appreciate that the problem hereunder discussion is a matter of the most fundamental theoret-ical significance and of overwhelming practical importance.For the economic principle, on which the practical advice ofeconomists has been almost invariably based, has assumed,in effect, that, cet. par., a decrease in spending will tend tolower the rate of interest and an increase in investment to raiseit. But if what these two quantities determine is, not the rateof interest, but the aggregate volume of employment, thenour outlook on the mechanism of the economic system willbe profoundly changed. A decreased readiness to spend willbe looked on in quite a different light If, instead of being re-garded as a factor which will, cet. par., increase investment, itis seen as a factor which will, cet. par., diminish employment.

178

14.A. Appendix on the Rate of Interest in Marshall’s Principles ofEconomics, Ricardo’s Principles of Political Economy, and Elsewhere

14.A Appendix on the Rate of Interest inMarshall’s Principles of Economics,

Ricardo’s Principles of Political Economy,and Elsewhere

I.

There is no consecutive discussion of the rate of interest in theworks of Marshall, Edgeworth or Professor Pigou,—nothingmore than a few obiter dicta. Apart from the passage alreadyquoted above (p. 139) the only important clues to Marshall’sposition on the rate of interest are to be found in his Principlesof Economics (6th edn.), Book VI. p. 534 and p. 593, the gist ofwhich is given by the following quotations:

‘Interest, being the price paid for the use ofcapital9 in any market, tends towards an equilib-rium level such that the aggregate demand for cap-ital in that market, at that rate of interest, is equalto the aggregate stock forthcoming there at thatrate. If the market, which we are considering, isa small one—say a single town, or a single tradein a progressive country—an increased demandfor capital in it will be promptly met by an in-creased supply drawn from surrounding districtsor trades. But if we are considering the wholeworld, or even the whole of a large country, asone market for capital, we cannot regard the ag-gregate supply of it as altered quickly and to a

9It is to be noticed that Marshall uses the word “capital” not “money” andthe word “stock” not “loans”; interest is a payment for borrowing money, and“demand for capital” in this context should mean “demand for loans of moneyfor the purpose of buying a stock of capital-goods”. But the equality betweenthe stock of capital-goods offered and the stock demanded will be broughtabout by the prices of capital-goods, not by the rate of interest. It is equality be-tween the demand and supply of loans of money, i.e. of debts, which is broughtabout by the rate of interest.

179


considerable extent by a change in the rate of in-terest. For the general fund of capital is the prod-uct of labour and waiting; and the extra work,10

and the extra waiting, to which a rise in the rateof interest would act as an incentive, would notquickly amount to much, as compared with thework and waiting, of which the total existing stockof capital is the result. An extensive increase inthe demand for capital in general will thereforebe met for a time not so much by an increase ofsupply, as by a rise in the rate of interest11; whichwill cause capital to withdraw itself partially fromthose uses in which its marginal utility is lowest.It is only slowly and gradually that the rise in the

10This assumes that income is not constant. But it is not obvious in what waya rise in the rate of interest will lead to “extra work”. Is the suggestion that a risein the rate of interest is to be regarded, by reason of its increasing the attractive-ness of working in order to save, as constituting a sort of increase in real wageswhich will induce the factors of production to work for a lower wage? This is, Ithink, in Mr. D. H. Robertson’s mind in a similar context. Certainly this “wouldnot quickly amount to much”; and an attempt to explain the actual fluctuationsin the amount of investment by means of this factor would be most unplausible,indeed absurd. My rewriting of the latter half of this sentence would be: “andif an extensive increase in the demand for capital in general, due to an increasein the schedule of the marginal efficiency of capital, is not offset by a rise in therate of interest, the extra employment and the higher level of income, whichwill ensue as a result of the increased production of capital-goods, will lead toan amount of extra waiting which in terms of money will be exactly equal tothe value of the current increment of capital-goods and will, therefore, preciselyprovide for it.”

11Why not by a rise in the supply price of capital-goods? Suppose, for ex-ample, that the “extensive increase in the demand for capital in general” is dueto a fall in the rate of interest. I would suggest that the sentence should berewritten: “In so far, therefore, as the extensive increase in the demand forcapital-goods cannot be immediately met by an increase in the total stock, itwill have to be held in check for the time being by a rise in the supply price ofcapital-goods sufficient to keep the marginal efficiency of capital in equilibriumwith the rate of interest without there being any material change in the scaleof investment; meanwhile (as always) the factors of production adapted for theoutput of capital-goods will be used in producing those capital-goods of whichthe marginal efficiency is greatest in the new conditions.”

180


rate of interest will increase the total stock of cap-ital’ (p.534).

‘It cannot be repeated too often that the phrase“the rate of interest” is applicable to old invest-ments of capital only in a very limited sense.12 Forinstance, we may perhaps estimate that a tradecapital of some seven thousand millions is investedin the different trades of this country at about 3per cent net interest. But such a method of speak-ing, though convenient and justifiable for manypurposes, is not accurate. What ought to be saidis that, taking the rate of net interest on the in-vestments of new capital in each of those trades[i.e. on marginal investments] to be about 3 percent; then the aggregate net income rendered bythe whole of the trade-capital invested in the var-ious trades is such that, if capitalised at 33 years’purchase (that is, on the basis of interest at 3 percent), it would amount to some seven thousandmillion pounds. For the value of the capital al-ready invested in improving land or erecting a build-ing, in making a railway or a machine, is the ag-gregate discounted value of its estimated futurenet incomes [or quasi-rents]; and if its prospec-tive income-yielding power should diminish, itsvalue would fall accordingly and would be thecapitalised value of that smaller income after al-lowing for depreciation’ (p.593).

In his Economics of Welfare (3rd edn.), p. 163, ProfessorPigou writes: ‘The nature of the service of “waiting” has beenmuch misunderstood. Sometimes it has been supposed toconsist in the provision of money, sometimes in the provi-sion of time, and, on both suppositions, it has been argued

12In fact we cannot speak of it at all. We can only properly speak of therate of interest on money borrowed for the purpose of purchasing investmentsof capital, new or old (or for any other purpose).

181


that no contribution whatever is made by it to the dividend.Neither supposition is correct. “Waiting” simply means post-poning consumption which a person has power to enjoy im-mediately, thus allowing resources, which might have beendestroyed, to assume the form of production instruments.13

The unit of “waiting” is, therefore, the use of a given quan-tity of resources14—for example, labour or machinery—for agiven time. . . . In more general terms we may say that theunit of waiting is a year-value-unit, or, in the simpler, if lessaccurate, language of Dr Cassel, a year-pound. . . A cautionmay be added against the common view that the amount ofcapital accumulated in any year is necessarily equal to theamount of “savings” made in it. This is not so, even whensavings are interpreted to mean net savings, thus eliminat-ing the savings of one man that are lent to increase the con-sumption of another, and when temporary accumulations ofunused claims upon services in the form of bank-money areignored; for many savings which are meant to become capitalin fact fail of their purpose through misdirection into wastefuluses.’15

Professor Pigou’s only significant reference to what deter-mines the rate of interest is, I think, to be found in his In-dustrial Fluctuations (1st edn.), pp. 251−3, where he contro-verts the view that the rate of interest, being determined by

13Here the wording is ambiguous as to whether we are to infer that thepostponement of consumption necessarily has this effect, or whether it merelyreleases resources which are then either unemployed or used for investmentaccording to circumstances.

14Not, be it noted, the amount of money which the recipient of incomemight, but does not, spend on consumption; so that the reward of waiting isnot interest but quasi-rent. This sentence seems to imply that the released re-sources are necessarily used. For what is the reward of waiting if the releasedsources are left unemployed?

15We are not told in this passage whether net savings would or would not beequal to the increment of capital, if we were to ignore misdirected investmentbut were to take account of “temporary accumulations of unused claims uponservices in the form of bank-money”. But in Industrial Fluctuations (p. 22) Prof.Pigou makes it clear that such accumulations have no effect on what he calls“real savings”.

182


the general conditions of demand and supply of real capital,lies outside the central or any other bank’s control. Againstthis view he argues that: ‘When bankers create more creditfor business men, they make, in their interest, subject to theexplanations given in chapter xiii. of part i.16, a forced levy ofreal things from the public, thus increasing the stream of realcapital available for them, and causing a fall in the real rate ofinterest on long and short loans alike. It is true, in short, thatthe bankers’ rate for money is bound by a mechanical tie tothe real rate of interest on long loans; but it is not true that thisreal rate is determined by conditions wholly outside bankers’control.’

My running comments on the above have been made in thefootnotes. The perplexity which I find in Marshall’s accountof the matter is fundamentally due, I think, to the incursion ofthe concept ’interest’, which belongs to a monetary economy,into a treatise which takes no account of money. ’Interest’ hasreally no business to turn up at all in Marshall’s Principles ofEconomics,—it belongs to another branch of the subject.

Professor Pigou, conformably with his other tacit assump-tions, leads us (in his Economics of Welfare) to infer that the unitof waiting is the same as the unit of current investment andthat the reward of waiting is quasi-rent, and practically nevermentions interest, which is as it should be. Nevertheless these

16This reference (op. cit. pp. 129-134) contains Prof. Pigou’s view as to theamount by which a new credit creation by the banks increases the stream of realcapital available for entrepreneurs. In effect he attempts to deduct “from thefloating credit handed over to business men through credit creations the float-ing capital which would have been contributed in other ways if the banks hadnot been there”. After these deductions have been made, the argument is oneof deep obscurity. To begin with, the rentiers have an income of 1500, of whichthey consume 500 and save 1000; the act of credit creation reduces their incometo 1300, of which they consume 500 - x and save 800 + x; and x, Prof. Pigou con-cludes, represents the net increase of capital made available by the act of creditcreation. Is the entrepreneurs’ income supposed to be swollen by the amountwhich they borrow from the banks (after making the above deductions)? Or isit swollen by the amount, i.e. 200, by which the rentiers’ income is reduced? Ineither case, are they supposed to save the whole of it? Is the increased invest-ment equal to the credit creations minus the deductions? Or is it equal to x?The argument seems to stop just where it should begin.

183


writers are not dealing with a non-monetary economy (if thereis such a thing). They quite clearly presume that money isused and that there is a banking system. Moreover, the rateof interest scarcely plays a larger part in Professor Pigou’s In-dustrial Fluctuations (which is mainly a study of fluctuations inthe marginal efficiency of capital) or in his Theory of Unemploy-ment (which is mainly a study of what determines changes inthe volume of employment, assuming that there is no invol-untary unemployment) than in his Economics of Welfare.

II.

The following from his Principles of Political Economy (p. 511)puts the substance of Ricardo’s theory of the rate of interest:

‘The interest of money is not regulated by therate at which the Bank will lend, whether it be 5,3 or 2 per cent., but by the rate of profit whichcan be made by the employment of capital, andwhich is totally independent of the quantity or ofthe value of money. Whether the Bank lent onemillion, ten millions, or a hundred millions, theywould not permanently alter the market rate ofinterest; they would alter only the value of themoney which they thus issued. In one case, tenor twenty times more money might be required tocarry on the same business than what might be re-quired in the other. The applications to the Bankfor money, then, depend on the comparison be-tween the rate of profits that may be made by theemployment of it, and the rate at which they arewilling to lend it. If they charge less than the mar-ket rate of interest, there is no amount of moneywhich they might not lend;—if they charge morethan that rate, none but spendthrifts and prodigalswould be found to borrow of them.’

184


This is so clear-cut that it affords a better starting-point fora discussion than the phrases of later writers who, without re-ally departing from the essence of the Ricardian doctrine, arenevertheless sufficiently uncomfortable about it to seek refugein haziness. The above is, of course, as always with Ricardo,to be interpreted as a long-period doctrine, with the empha-sis on the word ’permanently’ half-way through the passage;and it is interesting to consider the assumptions required tovalidate it.

Once again the assumption required is the usual classicalassumption, that there is always full employment; so that, as-suming no change in the supply curve of labour in terms ofproduct, there is only one possible level of employment inlong-period equilibrium. On this assumption with the usualceteris paribus, i.e. no change in psychological propensities andexpectations other than those arising out of a change in thequantity of money, the Ricardian theory is valid, in the sensethat on these suppositions there is only one rate of interestwhich will be compatible with full employment in the longperiod. Ricardo and his successors overlook the fact that evenin the long period the volume of employment is not neces-sarily full but is capable of varying, and that to every bank-ing policy there corresponds a different long-period level ofemployment; so that there are a number of positions of long-period equilibrium corresponding to different conceivable in-terest policies on the part of the monetary authority.

If Ricardo had been content to present his argument solelyas applying to any given quantity of money created by themonetary authority, it would still have been correct on the as-sumption of flexible money-wages. If, that is to say, Ricardohad argued that it would make no permanent alteration tothe rate of interest whether the quantity of money was fixedby the monetary authority at ten millions or at a hundred mil-lions, his conclusion would hold. But if by the policy of themonetary authority we mean the terms on which it will in-crease or decrease the quantity of money, i.e. the rate of in-terest at which it will, either by a change in the volume ofdiscounts or by open-market operations, increase or decrease

185


its assets—which is what Ricardo expressly does mean in theabove quotation—then it is not the case either that the policyof the monetary authority is nugatory or that only one policyis compatible with long-period equilibrium; though in the ex-treme case where money-wages are assumed to fall withoutlimit in face of involuntary unemployment through a futilecompetition for employment between the unemployed labour-ers, there will, it is true, be only two possible long-period po-sitions—full employment and the level of employment corre-sponding to the rate of interest at which liquidity-preferencebecomes absolute (in the event of this being less than full em-ployment). Assuming flexible money-wages, the quantity ofmoney as such is, indeed, nugatory in the long period; but theterms on which the monetary authority will change the quan-tity of money enters as a real determinant into the economicscheme.

It is worth adding that the concluding sentences of the quo-tation suggest that Ricardo was overlooking the possible changesin the marginal efficiency of capital according to the amountinvested. But this again can be interpreted as another exampleof his greater internal consistency compared with his succes-sors. For if the quantity of employment and the psychologicalpropensities of the community are taken as given, there is infact only one possible rate of accumulation of capital and, con-sequently, only one possible value for the marginal efficiencyof capital. Ricardo offers us the supreme intellectual achieve-ment, unattainable by weaker spirits, of adopting a hypotheti-cal world remote from experience as though it were the worldof experience and then living in it consistently. With most ofhis successors common sense cannot help breaking in—withinjury to their logical consistency.

III.

A peculiar theory of the rate of interest has been propoundedby Professor von Mises and adopted from him by ProfessorHayek and also, I think, by Professor Robbins; namely, thatchanges in the rate of interest can be identified with changes

186


in the relative price levels of consumption-goods and capital-goods.17 It is not clear how this conclusion is reached. Butthe argument seems to run as follows. By a somewhat dras-tic simplification the marginal efficiency of capital is taken asmeasured by the ratio of the supply price of new consumers’goods to the supply price of new producers’ goods.18 This isthen identified with the rate of interest. The fact is called tonotice that a fall in the rate of interest is favourable to invest-ment. Ergo, a fall in the ratio of the price of consumers’ goodsto the price of producer’s goods is favourable to investment.

By this means a link is established between tncreased sav-ing by an individual and increased aggregate investment. Forit is common gound that increased individual saving will causea fall in the price of consumers’ goods, and, quite possibly, agreater fall than in the price of producers’ goods; hence, ac-cording to the above reasoning, it means a reduction in therate of interest which will stimulate investment. But, of course,a lowering of the marginal efficiency of particular capital as-sets, and hence a lowering of the schedule of the marginal ef-ficiency of capital in general, has exactly the opposite effect towhat the above argument assumes. For investment is stimu-lated either by a raising of the schedule of the marginal effi-ciency or by a lowering of the rate of interest. As a result ofconfusing the marginal efficiency of capital with the rate ofinterest, Professor von Mises and his disciples have got theirconclusions exactly the wrong way round. A good exampleof a confusion along these lines is given by the following pas-sage by Professor Alvin Hansen:19 ‘It has been suggested bysome economists that the net effect of reduced spending will

17The Theory of Money and Credit, p. 339 et passim, particularly p. 363.18f we are in long-period equilibrium, special assumptions might be de-

vised on which this could be justified. But when the prices in question are theprices prevailing in slump conditions, the simplification of supposing that theentrepreneur will, in forming his expectations, assume these prices to be perma-nent, is certain to be misleading. Moreover, if he does, the prices of the existingstock of producers’ goods will fall in the same proportion as the prices of con-sumers’ goods.

19Economic Reconstruction, p. 233.

187


be a lower price level of consumers’ goods than would other-wise have been the case, and that, in consequence, the stim-ulus to investment in fixed capital would thereby tend to beminimised. This view is, however, incorrect and is based ona confusion of the effect on capital formation of (i) higher orlower prices of consumers’ goods, and (2) a change in the rateof interest. It is true that in consequence of the decreasedspending and increased saving, consumers’ prices would below relative to the prices of producers’ goods. But this, in ef-fect, means a lower rate of interest, and a lower rate of intereststimulates an expansion of capital investment in fields whichat higher rates would be unprofitable.’

188

CHAPTER 15

THE PSYCHOLOGICAL AND BUSINESSINCENTIVES TO LIQUIDITY

I

We must now develop in more detail the analysis of the mo-tives to liquidity-preference which were introduced in a pre-liminary way in Chapter 13. The subject is substantially thesame as that which has been sometimes discussed under theheading of the Demand for Money. It is also closely connectedwith what is called the income-velocity of money; — for theincome-velocity of money merely measures what proportionof their incomes the public chooses to hold in cash, so that anincreased income-velocity of money may be a symptom of adecreased liquidity-preference. It is not the same thing, how-ever, since it is in respect of his stock of accumulated savings,rather than of his income, that the individual can exercise hischoice between liquidity and illiquidity. And, anyhow, theterm “income-velocity of money” carries with it the mislead-ing suggestion of a presumption in favour of the demand formoney as a whole being proportional, or having some deter-minate relation, to income, whereas this presumption shouldapply, as we shall see, only to a portion of the public’s cashholdings; with the result that it overlooks the part played bythe rate of interest.

In my Treatise on Money I studied the total demand for moneyunder the headings of income-deposits, business-deposits, andsavings-deposits, and I need not repeat here the analysis whichI gave in Chapter 3 of that book. Money held for each of thethree purposes forms, nevertheless, a single pool, which theholder is under no necessity to segregate into three water-tightcompartments; for they need not be sharply divided even in

189

15. THE PSYCHOLOGICAL AND BUSINESS INCENTIVES TOLIQUIDITY

his own mind, and the same sum can be held primarily for onepurpose and secondarily for another. Thus we can — equallywell, and, perhaps, better — consider the individual’s aggre-gate demand for money in given circumstances as a single de-cision, though the composite result of a number of differentmotives.

In analysing the motives, however, it is still convenientto classify them under certain headings, the first of whichbroadly corresponds to the former classification of income-deposits and business-deposits, and the two latter to that ofsavings-deposits. These I have briefly introduced in Chapter13 under the headings of the transactions-motive, which canbe further classified as the income-motive and the business-motive, the precautionary-motive and the speculative-motive.

(i) The Income-motive. — One reason for holding cash is tobridge the interval between the receipt of income and its dis-bursement. The strength of this motive in inducing a decisionto hold a given aggregate of cash will chiefly depend on theamount of income and the normal length of the interval be-tween its receipt and its disbursement. It is in this connectionthat the concept of the income-velocity of money is strictly ap-propriate.

(ii) The Business-motive. — Similarly, cash is held to bridgethe interval between the time of incurring business costs andthat of the receipt of the sale-proceeds; cash held by dealersto bridge the interval between purchase and realisation beingincluded under this heading. The strength of this demand willchiefly depend on the value of current output (and hence oncurrent income), and on the number of hands through whichoutput passes.

(iii) The Precautionary-motive. — To provide for contingen-cies requiring sudden expenditure and for unforeseen oppor-tunities of advantageous purchases, and also to hold an assetof which the value is fixed in terms of money to meet a sub-sequent liability fixed in terms of money, are further motivesfor holding cash.

190

I.

The strength of all these three types of motive will partlydepend on the cheapness and the reliability of methods of ob-taining cash, when it is required, by some form of temporaryborrowing, in particular by overdraft or its equivalent. Forthere is no necessity to hold idle cash to bridge over intervalsif it can be obtained without difficulty at the moment when itis actually required. Their strength will also depend on whatwe may term the relative cost of holding cash. If the cash canonly be retained by forgoing the purchase of a profitable asset,this increases the cost and thus weakens the motive towardsholding a given amount of cash. If deposit interest is earnedor if bank charges are avoided by holding cash, this decreasesthe cost and strengthens the motive. It may be, however, thatthis is likely to be a minor factor except where large changesin the cost of holding cash are in question.

(iv) There remains the Speculative-motive. — This needs amore detailed examination than the others, both because itis less well understood and because it is particularly impor-tant in transmitting the effects of a change in the quantity ofmoney.

In normal circumstances the amount of money required tosatisfy the transactions-motive and the precautionary-motiveis mainly a resultant of the general activity of the economicsystem and of the level of money-income. But it is by playingon the speculative-motive that monetary management (or, inthe absence of management, chance changes in the quantityof money) is brought to bear on the economic system. Forthe demand for money to satisfy the former motives is gener-ally irresponsive to any influence except the actual occurrenceof a change in the general economic activity and the levelof incomes; whereas experience indicates that the aggregatedemand for money to satisfy the speculative-motive usuallyshows a continuous response to gradual changes in the rateof interest, i.e. there is a continuous curve relating changes inthe demand for money to satisfy the speculative motive and

191


changes in the rate of interest as given by changes in the pricesof bonds and debts of various maturities.

Indeed, if this were not so, “open market operations” wouldbe impracticable. I have said that experience indicates the con-tinuous relationship stated above, because in normal circum-stances the banking system is in fact always able to purchase(or sell) bonds in exchange for cash by bidding the price ofbonds up (or down) in the market by a modest amount; andthe larger the quantity of cash which they seek to create (orcancel) by purchasing (or selling) bonds and debts, the greatermust be the fall (or rise) in the rate of interest. Where, how-ever, (as in the United States, 1933-1934) open-market oper-ations have been limited to the purchase of very short-datedsecurities, the effect may, of course, be mainly confined to thevery short-term rate of interest and have but little reaction onthe much more important long-term rates of interest.

In dealing with the speculative-motive it is, however, im-portant to distinguish between the changes in the rate of inter-est which are due to changes in the supply of money availableto satisfy the speculative-motive, without there having beenany change in the liquidity function, and those which are pri-marily due to changes in expectation affecting the liquidityfunction itself. Open-market operations may, indeed, influ-ence the rate of interest through both channels; since they maynot only change the volume of money, but may also give riseto changed expectations concerning the future policy of theCentral Bank or of the Government. Changes in the liquidityfunction itself, due to a change in the news which causes re-vision of expectations, will often be discontinuous, and will,therefore, give rise to a corresponding discontinuity of changein the rate of interest. Only, indeed, in so far as the change inthe news is differently interpreted by different individuals oraffects individual interests differently will there be room forany increased activity of dealing in the bond market. If thechange in the news affects the judgment and the requirementsof everyone in precisely the same way, the rate of interest (asindicated by the prices of bonds and debts) will be adjustedforthwith to the new situation without any market transac-

192

I.

tions being necessary.Thus, in the simplest case, where everyone is similar and

similarly placed, a change in circumstances or expectationswill not be capable of causing any displacement of moneywhatever; — it will simply change the rate of interest in what-ever degree is necessary to offset the desire of each individ-ual, felt at the previous rate, to change his holding of cash inresponse to the new circumstances or expectations; and, sinceeveryone will change his ideas as to the rate which would in-duce him to alter his holdings of cash in the same degree, notransactions will result. To each set of circumstances and ex-pectations there will correspond an appropriate rate of inter-est, and there will never be any question of anyone changinghis usual holdings of cash.

In general, however, a change in circumstances or expec-tations will cause some realignment in individual holdingsof money; — since, in fact, a change will influence the ideasof different individuals differently by reason partly of differ-ences in environment and the reason for which money is heldand partly of differences in knowledge and interpretation ofthe new situation. Thus the new equilibrium rate of interestwill be associated with a redistribution of money-holdings.Nevertheless it is the change in the rate of interest, rather thanthe redistribution of cash, which deserves our main attention.The latter is incidental to individual differences, whereas theessential phenomenon is that which occurs in the simplestcase. Moreover, even in the general case, the shift in the rate ofinterest is usually the most prominent part of the reaction toa change in the news. The movement in bond-prices is, as thenewspapers are accustomed to say, “out of all proportion tothe activity of dealing”; — which is as it should be, in view ofindividuals being much more similar than they are dissimilarin their reaction to news.

193


II

Whilst the amount of cash which an individual decides tohold to satisfy the transactions-motive and the precautionary-motive is not entirely independent of what he is holding tosatisfy the speculative-motive, it is a safe first approximationto regard the amounts of these two sets of cash-holdings as be-ing largely independent of one another. Let us, therefore, forthe purposes of our further analysis, break up our problem inthis way.

Let the amount of cash held to satisfy the transactions- andprecautionary-motives be M1, and the amount held to satisfythe speculative-motive be M2. Corresponding to these twocompartments of cash, we then have two liquidity functionsL1 and L2. L1 mainly depends on the level of income, whilstL2 mainly depends on the relation between the current rate ofinterest and the state of expectation. Thus

M = M1 +M2 = L1(Y ) + L2(r) ,

where L1 is the liquidity function corresponding to an incomeY , which determines M1, and L2 is the liquidity function ofthe rate of interest r, which determines M2. It follows thatthere are three matters to investigate: (i) the relation of changesin M to Y and r, (ii) what determines the shape of L1, (iii)what determines the shape of L2.

(i) The relation of changes in M to Y and r depends, in thefirst instance, on the way in which changes in M come about.Suppose that M consists of gold coins and that changes inM can only result from increased returns to the activities ofgold-miners who belong to the economic system under exam-ination. In this case changes in M are, in the first instance,directly associated with changes in Y , since the new gold ac-crues as someone’s income. Exactly the same conditions holdif changes in M are due to the Government printing moneywherewith to meet its current expenditure; — in this case alsothe new money accrues as someone’s income. The new levelof income, however, will not continue sufficiently high for the

194

II.

requirements of M , to absorb the whole of the increase in M ;and some portion of the money will seek an outlet in buyingsecurities or other assets until r has fallen so as to bring aboutan increase in the magnitude of M , and at the same time tostimulate a rise in Y to such an extent that the new money isabsorbed either in M2 or in the M1 which corresponds to therise in Y caused by the fall in r. Thus at one remove this casecomes to the same thing as the alternative case, where the newmoney can only be issued in the first instance by a relaxationof the conditions of credit by the banking system, so as to in-duce someone to sell the banks a debt or a bond in exchangefor the new cash.

It will, therefore, be safe for us to take the latter case as typ-ical. A change in M can be assumed to operate by changingr, and a change in r will lead to a new equilibrium partly bychangingM2 and partly by changing Y and thereforeM1. Thedivision of the increment of cash between M1 and M2 in thenew position of equilibrium will depend on the responses ofinvestment to a reduction in the rate of interest and of incometo an increase in investment.1 Since Y partly depends on r,it follows that a given change in M has to cause a sufficientchange in r for the resultant changes in M1 and M2 respec-tively to add up to the given change in M .

(ii) It is not always made clear whether the income-velocityof money is defined as the ratio of Y to M or as the ratio of Yto M1. I propose, however, to take it in the latter sense. Thusif V is the income-velocity of money,

L1(Y ) = Y/V = M1

.There is, of course, no reason for supposing that V is con-

stant. Its value will depend on the character of banking andindustrial organisation, on social habits, on the distribution ofincome between different classes and on the effective cost of

1We must postpone to Book V. the question of what will determine the char-acter of the new equilibrium.

195


holding idle cash. Nevertheless, if we have a short period oftime in view and can safely assume no material change in anyof these factors, we can treat V as nearly enough constant.

(iii) Finally there is the question of the relation between M2

and r. We have seen in Chapter 13 that uncertainty as to thefuture course of the rate of interest is the sole intelligible ex-planation of the type of liquidity-preference L2 which leadsto the holding of cash M2. It follows that a given M2 will nothave a definite quantitative relation to a given rate of inter-est of r; — what matters is not the absolute level of r but thedegree of its divergence from what is considered a fairly safelevel of r, having regard to those calculations of probabilitywhich are being relied on. Nevertheless, there are two reasonsfor expecting that, in any given state of expectation, a fall in rwill be associated with an increase in M2. In the first place, ifthe general view as to what is a safe level of r is unchanged,every fall in r reduces the market rate relatively to the “safe”rate and therefore increases the risk of illiquidity; and, in thesecond place, every fall in r reduces the current earnings fromilliquidity, which are available as a sort of insurance premiumto offset the risk of loss on capital account, by an amount equalto the difference between the squares of the old rate of interestand the new. For example, if the rate of interest on a long-term debt is 4 per cent., it is preferable to sacrifice liquidityunless on a balance of probabilities it is feared that the long-term rate of interest may rise faster than by 4 per cent. of itselfper annum, i.e. by an amount greater than 0.16 per cent. perannum. If, however, the rate of interest is already as low as2 per cent., the running yield will only offset a rise in it of aslittle as 0.04 per cent. per annum. This, indeed, is perhapsthe chief obstacle to a fall in the rate of interest to a very lowlevel. Unless reasons are believed to exist why future experi-ence will be very different from past experience, a long-termrate of interest of (say) 2 per cent. leaves more to fear than tohope, and offers, at the same time, a running yield which isonly sufficient to offset a very small measure of fear.

It is evident, then, that the rate of interest is a highly psy-

196

II.

chological phenomenon. We shall find, in equilibrium at alevel below the rate which corresponds to full employment; be-cause at such a level a state of true inflation will be produced,with the result that M1 will absorb ever-increasing quantitiesof cash. But at a level above the rate which corresponds tofull employment, the long-term market-rate of interest willdepend, not only on the current policy of the monetary au-thority, but also on market expectations concerning its futurepolicy. The short-term rate of interest is easily controlled bythe monetary authority, both because it is not difficult to pro-duce a conviction that its policy will not greatly change in thevery near future, and also because the possible loss is smallcompared with the running yield (unless it is approachingvanishing point). The the long-term rate may be more recal-citrant when once it has fallen to a level which, on the basisof past experience and present expectations of future mone-tary policy, is considered “unsafe” by representative opinion.For example, in a country linked to an international gold stan-dard, a rate of interest lower than prevails elsewhere will beviewed with a justifiable lack of confidence; yet a domesticrate of interest dragged up to a parity with the highest rate(highest after allowing for risk) prevailing in any country be-longing to the international system may be much higher thanis consistent with domestic full employment.

Thus a monetary policy which strikes public opinion as be-ing experimental in character or easily liable to change mayfail in its objective of greatly reducing the long-term rate of in-terest, because M2 may tend to increase almost without limitin response to a reduction of r below a certain figure. Thesame policy, on the other hand, may prove easily successful ifit appeals to public opinion as being reasonable and practica-ble and in the public interest, rooted in strong conviction, andpromoted by an authority unlikely to be superseded.

It might be more accurate, perhaps, to say that the rate ofinterest is a highly conventional, rather than a highly psycho-logical, phenomenon. For its actual value is largely governedby the prevailing view as to what its value is expected to be.Any level of interest which is accepted with sufficient convic-

197


tion as likely to be durable will be durable; subject, of course,in a changing society to fluctuations for all kinds of reasonsround the expected normal. In particular, whenM1 is increas-ing faster than M , the rate of interest will rise, and vice versa.But it may fluctuate for decades about a level which is chron-ically too high for full employment; — particularly if it is theprevailing opinion that the rate of interest is self-adjusting,so that the level established by convention is thought to berooted in objective grounds much stronger than convention,the failure of Employment to attain an optimum level beingin no way associated, in the minds either of the public or ofauthority, with the prevalence of an inappropriate range ofrates of interest.

The difficulties in the way of maintaining effective demandat a level high enough to provide full employment, which en-sue from the association of a conventional and fairly stablelong-term rate of interest with a fickle and highly unstablemarginal efficiency of capital, should be, by now, obvious tothe reader.

Such comfort as we can fairly take from more encourag-ing reflections must be drawn from the hope that, preciselybecause the convention is not rooted in secure knowledge, itwill not be always unduly resistant to a modest measure ofpersistence and consistency of purpose by the monetary au-thority. Public opinion can be fairly rapidly accustomed to amodest fall in the rate of interest and the conventional expec-tation of the future may be modified accordingly; thus prepar-ing the way for a further movement — up to a point. The fallin the long-term rate of interest in Great Britain after her de-parture from the gold standard provides an interesting exam-ple of this; — the major movements were effected by a seriesof discontinuous jumps, as the liquidity function of the pub-lic, having become accustomed to each successive reduction,became ready to respond to some new incentive in the newsor in the policy of the authorities.

198

III.

III

We can sum up the above in the proposition that in any givenstate of expectation there is in the minds of the public a certainpotentiality towards holding cash beyond what is required bythe transactions-motive or the precautionary-motive, whichwill realise itself in actual cash-holdings in a degree which de-pends on the terms on which the monetary authority is will-ing to create cash. It is this potentiality which is summed upin the liquidity function L2.

Corresponding to the quantity of money created by themonetary authority, there will, therefore, be cet. par. a deter-minate rate of interest or, more strictly, a determinate complexof rates of interest for debts of different maturities. The samething, however, would be true of any other factor in the eco-nomic system taken separately. Thus this particular analysiswill only be useful and significant in so far as there is somespecially direct or purposive connection between changes inthe quantity of money and changes in the rate of interest. Ourreason for supposing that there is such a special connectionarises from the fact that, broadly speaking, the banking sys-tem and the monetary authority are dealers in money anddebts and not in assets or consumables.

If the monetary authority were prepared to deal both wayson specified terms in debts of all maturities, and even moreso if it were prepared to deal in debts of varying degrees ofrisk, the relationship between the complex of rates of interestand the quantity of money would be direct. The complex ofrates of interest would simply be an expression of the termson which the banking system is prepared to acquire or partwith debts; and the quantity of money would be the amountwhich can find a home in the possession of individuals who— after taking account of all relevant circumstances — preferthe control of liquid cash to parting with it in exchange fora debt on the terms indicated by the market rate of interest.Perhaps a complex offer by the central bank to buy and sellat stated prices gilt-edged bonds of all maturities, in place ofthe single bank rate for short-term bills, is the most important

199


practical improvement which can be made in the technique ofmonetary management.

To-day, however, in actual practice, the extent to which theprice of debts as fixed by the banking system is “effective”in the market, in the sense that it governs the actual market-price, varies in different systems. Sometimes the price is moreeffective in one direction than in the other; that is to say, thebanking system may undertake to purchase debts at a certainprice but not necessarily to sell them at a figure near enoughto its buying-price to represent no more than a dealer’s turn,though there is no reason why the price should not be madeeffective both ways with the aid of open-market operations.There is also the more important qualification which arises outof the monetary authority not being, as a rule, an equally will-ing dealer in debts of all maturities. The monetary authorityoften tends in practice to concentrate upon short-term debtsand to leave the price of long-term debts to be influenced bybelated and imperfect reactions from the price of short-termdebts; — though here again there is no reason why they needdo so. Where these qualifications operate, the directness ofthe relation between the rate of interest and the quantity ofmoney is correspondingly modified. In Great Britain the fieldof deliberate control appears to be widening. But in applyingthis theory in any particular case allowance must be made forthe special characteristics of the method actually employed bythe monetary authority. If the monetary authority deals onlyin short-term debts, we have to consider what influence theprice, actual and prospective, of short-term debts exercises ondebts of longer maturity.

Thus there are certain limitations on the ability of the mon-etary authority to establish any given complex of rates of in-terest for debts of different terms and risks, which can be summedup as follows:

(1) There are those limitations which arise out of the mon-etary authority’s own practices in limiting its willingness todeal to debts of a particular type.

(2) There is the possibility, for the reasons discussed above,

200

III.

that, after the rate of interest has fallen to a certain level, liquidity-preference may become virtually absolute in the sense that al-most everyone prefers cash to holding a debt which yields solow a rate of interest. In this event the monetary authoritywould have lost effective control over the rate of interest. Butwhilst this limiting case might become practically importantin future, I know of no example of it hitherto. Indeed, ow-ing to the unwillingness of most monetary authorities to dealboldly in debts of long term, there has not been much oppor-tunity for a test. Moreover, if such a situation were to arise,it would mean that the public authority itself could borrowthrough the banking system on an unlimited scale at a nomi-nal rate of interest.

(3) The most striking examples of a complete breakdownof stability in the rate of interest, due to the liquidity functionflattening out in one direction or the other, have occurred invery abnormal circumstances. In Russia and Central Europeafter the war a currency crisis or flight from the currency wasexperienced, when no one could be induced to retain holdingseither of money or of debts on any terms whatever, and evena high and rising rate of interest was unable to keep pace withthe marginal efficiency of capital (especially of stocks of liq-uid goods) under the influence of the expectation of an evergreater fall in the value of money; whilst in the United Statesat certain dates in 1932 there was a crisis of the opposite kind— a financial crisis or crisis of liquidation, when scarcely any-one could be induced to part with holdings of money on anyreasonable terms.

(4) There is, finally, the difficulty discussed in section iv ofChapter 11, p. 144, in the way of bringing the effective rate ofinterest below a certain figure, which may prove important inan era of low interest-rates; namely the intermediate costs ofbringing the borrower and the ultimate lender together, andthe allowance for risk, especially for moral risk, which thelender requires over and above the pure rate of interest. Asthe pure rate of interest declines it does not follow that the al-lowances for expense and risk decline pari passu. Thus the

201


rate of interest which the typical borrower has to pay maydecline more slowly than the pure rate of interest, and maybe incapable of being brought, by the methods of the existingbanking and financial organisation, below a certain minimumfigure. This is particularly important if the estimation of moralrisk is appreciable. For where the risk is due to doubt in themind of the lender concerning the honesty of the borrower,there is nothing in the mind of a borrower who does not in-tend to be dishonest to offset the resultant higher charge. Itis also important in the case of short-term loans (e.g. bankloans) where the expenses are heavy; — a bank may have tocharge its customers 1 1/2 to 2 per cent., even if the pure rateof interest to the lender is nil.

IV

At the cost of anticipating what is more properly the subject ofChapter 21 below it may be interesting briefly at this stage toindicate the relationship of the above to the Quantity Theoryof Money.

In a static society or in a society in which for any other rea-son no one feels any uncertainty about the future rates of in-terest, the Liquidity Function L2 or the propensity to hoard (aswe might term it), will always be zero in equilibrium. Hencein equilibrium M2 = 0 and M = M1; so that any change in Mwill cause the rate of interest to fluctuate until income reachesa level at which the change in M1 is equal to the supposedchange inM . NowM1V = Y , where V is the income-velocityof money as defined above and Y is the aggregate income.Thus if it is practicable to measure the quantity, O, and theprice, P , of current output, we have Y = OP , and, therefore,MY = OP ; which is much the same as the Quantity Theoryof Money in its traditional form.2

2If we had defined V , not as equal to Y/M , but as equal to Y/M , then,of course, the Quantity Theory is a truism which holds in all circumstances,though without significance.

202

IV.

For the purposes of the real world it is a great fault in theQuantity Theory that it does not distinguish between changesin prices which are a function of changes in output, and thosewhich are a function of changes in the wage-unit.3 The ex-planation of this omission is, perhaps, to be found in the as-sumptions that there is no propensity to hoard and that thereis always full employment. For in this case, O being constantand M2 being zero, it follows, if we can take V also as con-stant, that both the wage-unit and the price-level will be di-rectly proportional to the quantity of money.

3This point will be further developed in Chapter 21 below.

203

CHAPTER 16

SUNDRY OBSERVATIONS ON THE NATURE OFCAPITAL

I

An act of individual saving means — so to speak — a decisionnot to have dinner to-day. But it does not necessitate a deci-sion to have dinner or to buy a pair of boots a week hence ora year hence or to consume any specified thing at any speci-fied date. Thus it depresses the business of preparing to-day’sdinner without stimulating the business of making ready forsome future act of consumption. It is not a substitution of fu-ture consumption-demand for present consumption-demand,— it is a net diminution of such demand. Moreover, the ex-pectation of future consumption is so largely based on currentexperience of present consumption that a reduction in the lat-ter is likely to depress the former, with the result that the actof saving will not merely depress the price of consumption-goods and leave the marginal efficiency of existing capital un-affected, but may actually tend to depress the latter also. Inthis event it may reduce present investment-demand as wellas present consumption-demand.

If saving consisted not merely in abstaining from presentconsumption but in placing simultaneously a specific orderfor future consumption, the effect might indeed be different.For in that case the expectation of some future yield frominvestment would be improved, and the resources releasedfrom preparing for present consumption could be turned overto preparing for the future consumption. Not that they nec-essarily would be, even in this case, on a scale equal to theamount of resources released; since the desired interval of de-lay might require a method of production so inconveniently

205

16. SUNDRY OBSERVATIONS ON THE NATURE OF CAPITAL

“roundabout” as to have an efficiency well below the cur-rent rate of interest, with the result that the favourable effecton employment of the forward order for consumption wouldeventuate not at once but at some subsequent date, so that theimmediate effect of the saving would still be adverse to em-ployment. In any case, however, an individual decision tosave does not, in actual fact, involve the placing of any specificforward order for consumption, but merely the cancellation ofa present order. Thus, since the expectation of consumption isthe only raison d’être of employment, there should be nothingparadoxical in the conclusion that a diminished propensity toconsume has cet. par. a depressing effect on employment.

The trouble arises, therefore, because the act of saving im-plies, not a substitution for present consumption of some spe-cific additional consumption which requires for its prepara-tion just as much immediate economic activity as would havebeen required by present consumption equal in value to thesum saved, but a desire for “wealth” as such, that is for a po-tentiality of consuming an unspecified article at an unspeci-fied time. The absurd, though almost universal, idea that anact of individual saving is just as good for effective demand asan act of individual consumption, has been fostered by the fal-lacy, much more specious than the conclusion derived from it,that an increased desire to hold wealth, being much the samething as an increased desire to hold investments, must, by in-creasing the demand for investments, provide a stimulus totheir production; so that current investment is promoted byindividual saving to the same extent as present consumptionis diminished.

It is of this fallacy that it is most difficult to disabuse men’sminds. It comes from believing that the owner of wealth de-sires a capital-asset as such, whereas what he really desires isits prospective yield. Now, prospective yield wholly dependson the expectation of future effective demand in relation tofuture conditions of supply. If, therefore, an act of savingdoes nothing to improve prospective yield, it does nothing tostimulate investment. Moreover, in order that an individualsaver may attain his desired goal of the ownership of wealth,

206

I.

it is not necessary that a new capital-asset should be producedwherewith to satisfy him. The mere act of saving by one indi-vidual, being two-sided as we have shown above, forces someother individual to transfer to him some article of wealth oldor new. Every act of saving involves a “forced” inevitabletransfer of wealth to him who saves, though he in his turnmay suffer from the saving of others. These transfers of wealthdo not require the creation of new wealth — indeed, as wehave seen, they may be actively inimical to it. The creation ofnew wealth wholly depends on the prospective yield of thenew wealth reaching the standard set by the current rate ofinterest. The prospective yield of the marginal new invest-ment is not increased by the fact that someone wishes to in-crease his wealth, since the prospective yield of the marginalnew investment depends on the expectation of a demand fora specific article at a specific date.

Nor do we avoid this conclusion by arguing that what theowner of wealth desires is not a given prospective yield butthe best available prospective yield, so that an increased de-sire to own wealth reduces the prospective yield with whichthe producers of new investment have to be content. For thisoverlooks the fact that there is always an alternative to theownership of real capital-assets, namely the ownership of moneyand debts; so that the prospective yield with which the pro-ducers of new investment have to be content cannot fall be-low the standard set by the current rate of interest. And thecurrent rate of interest depends, as we have seen, not on thestrength of the desire to hold wealth, but on the strengths ofthe desires to hold it in liquid and in illiquid forms respec-tively, coupled with the amount of the supply of wealth inthe one form relatively to the supply of it in the other. If thereader still finds himself perplexed, let him ask himself why,the quantity of money being unchanged, a fresh act of savingshould diminish the sum which it is desired to keep in liquidform at the existing rate of interest.

Certain deeper perplexities, which may arise when we tryto probe still further into the whys and wherefores, will beconsidered in the next chapter.

207


II

It is much preferable to speak of capital as having a yield overthe course of its life in excess of its original cost, than as beingproductive. For the only reason why an asset offers a prospectof yielding during its life services having an aggregate valuegreater than its initial supply price is because it is scarce; andit is kept scarce because of the competition of the rate of inter-est on money. If capital becomes less scarce, the excess yieldwill diminish, without its having become less productive —at least in the physical sense.

I sympathise, therefore, with the pre-classical doctrine thateverything is produced by labour, aided by what used to becalled art and is now called technique, by natural resourceswhich are free or cost a rent according to their scarcity orabundance, and by the results of past labour, embodied in as-sets, which also command a price according to their scarcityor abundance. It is preferable to regard labour, including, ofcourse, the personal services of the entrepreneur and his assis-tants, as the sole factor of production, operating in a given en-vironment of technique, natural resources, capital equipmentand effective demand. This partly explains why we have beenable to take the unit of labour as the sole physical unit whichwe require in our economic system, apart from units of moneyand of time.

It is true that some lengthy or roundabout processes arephysically efficient. But so are some short processes. Lengthyprocesses are not physically efficient because they are long.Some, probably most, lengthy processes would be physicallyvery inefficient, for there are such things as spoiling or wast-ing with time.1 With a given labour force there is a definitelimit to the quantity of labour embodied in roundabout pro-cesses which can be used to advantage. Apart from otherconsiderations, there must be a due proportion between theamount of labour employed in making machines and the amountwhich will be employed in using them. The ultimate quan-

1Cf. Marshall’s note on Böhm-Bawerk, Principles, p. 593.

208

II.

tity of value will not increase indefinitely, relatively to thequantity of labour employed, as the processes adopted be-come more and more roundabout, even if their physical ef-ficiency is still increasing. Only if the desire to postpone con-sumption were strong enough to produce a situation in whichfull employment required a volume of investment so great asto involve a negative marginal efficiency of capital, would aprocess become advantageous merely because it was lengthy;in which event we should employ physically inefficient pro-cesses, provided they were sufficiently lengthy for the gainfrom postponement to outweigh their inefficiency. We shouldin fact have a situation in which short processes would have tobe kept sufficiently scarce for their physical efficiency to out-weigh the disadvantage of the early delivery of their product.A correct theory, therefore, must be reversible so as to be ableto cover the cases of the marginal efficiency of capital corre-sponding either to a positive or to a negative rate of interest;and it is, I think, only the scarcity theory outlined above whichis capable of this.

Moreover there are all sorts of reasons why various kindsof services and facilities are scarce and therefore expensiverelatively to the quantity of labour involved. For example,smelly processes command a higher reward, because peoplewill not undertake them otherwise. So do risky processes.But we do not devise a productivity theory of smelly or riskyprocesses as such. In short, not all labour is accomplished inequally agreeable attendant circumstances; and conditions ofequilibrium require that articles produced in less agreeable at-tendant circumstances (characterised by smelliness, risk or thelapse of time) must be kept sufficiently scarce to command ahigher price. But if the lapse of time becomes an agreeableattendant circumstance, which is a quite possible case and al-ready holds for many individuals, then, as I have said above,it is the short processes which must be kept sufficiently scarce.

Given the optimum amount of roundaboutness, we shall,of course, select the most efficient roundabout processes whichwe can find up to the required aggregate. But the optimumamount itself should be such as to provide at the appropri-

209


ate dates for that part of consumers’ demand which it is de-sired to defer. In optimum conditions, that is to say, produc-tion should be so organised as to produce in the most effi-cient manner compatible with delivery at the dates at whichconsumers’ demand is expected to become effective. It is nouse to produce for delivery at a different date from this, eventhough the physical output could be increased by changingthe date of delivery; — except in so far as the prospect of alarger meal, so to speak, induces the consumer to anticipateor postpone the hour of dinner. If, after hearing full particu-lars of the meals he can get by fixing dinner at different hours,the consumer is expected to decide in favour of 8 o’clock, it isthe business of the cook to provide the best dinner he can forservice at that hour, irrespective of whether 7.30, 8 o’clock or8.30 is the hour which would suit him best if time counted fornothing, one way or the other, and his only task was to pro-duce the absolutely best dinner. In some phases of society itmay be that we could get physically better dinners by dininglater than we do; but it is equally conceivable in other phasesthat we could get better dinners by dining earlier. Our theorymust, as I have said above, be applicable to both contingen-cies.

If the rate of interest were zero, there would be an optimuminterval for any given article between the average date of in-put and the date of consumption, for which labour cost wouldbe a minimum; — a shorter process of production would beless efficient technically, whilst a longer process would alsobe less efficient by reason of storage costs and deterioration.If, however, the rate of interest exceeds zero, a new elementof cost is introduced which increases with the length of theprocess, so that the optimum interval will be shortened, andthe current input to provide for the eventual delivery of thearticle will have to be curtailed until the prospective pricehas increased sufficiently to cover the increased cost — a costwhich will be increased both by the interest charges and alsoby the diminished efficiency of the shorter method of produc-tion. Whilst if the rate of interest falls below zero (assumingthis to be technically possible), the opposite is the case. Given

210

III.

the prospective consumers’ demand, current input to-day hasto compete, so to speak, with the alternative of starting inputat a later date; and, consequently, current input will only beworth while when the greater cheapness, by reason of greatertechnical efficiency or prospective price changes, of producinglater on rather than now, is insufficient to offset the smallerreturn from negative interest. In the case of the great major-ity of articles it would involve great technical inefficiency tostart up their input more than a very modest length of timeahead of their prospective consumption. Thus even if the rateof interest is zero, there is a strict limit to the proportion ofprospective consumers’ demand which it is profitable to beginproviding for in advance; and, as the rate of interest rises, theproportion of the prospective consumers’ demand for whichit pays to produce to-day shrinks pari passu.

III

We have seen that capital has to be kept scarce enough in thelong-period to have a marginal efficiency which is at leastequal to the rate of interest for a period equal to the life ofthe capital, as determined by psychological and institutionalconditions. What would this involve for a society which findsitself so well equipped with capital that its marginal efficiencyis zero and would be negative with any additional invest-ment; yet possessing a monetary system, such that money will“keep” and involves negligible costs of storage and safe cus-tody, with the result that in practice interest cannot be nega-tive; and, in conditions of full employment, disposed to save?

If, in such circumstances, we start from a position of fullemployment, entrepreneurs will necessarily make losses if theycontinue to offer employment on a scale which will utilise thewhole of the existing stock of capital. Hence the stock of cap-ital and the level of employment will have to shrink until thecommunity becomes so impoverished that the aggregate ofsaving has become zero, the positive saving of some individ-uals or groups being offset by the negative saving of others.

211


Thus for a society such as we have supposed, the positionof equilibrium, under conditions of laissez-faire, will be onein which employment is low enough and the standard of lifesufficiently miserable to bring savings to zero. More probablythere will be a cyclical movement round this equilibrium posi-tion. For if there is still room for uncertainty about the future,the marginal efficiency of capital will occasionally rise abovezero leading to a “boom”, and in the succeeding “slump” thestock of capital may fall for a time below the level which willyield a marginal efficiency of zero in the long run. Assumingcorrect foresight, the equilibrium stock of capital which willhave a marginal efficiency of precisely zero will, of course, bea smaller stock than would correspond to full employment ofthe available labour; for it will be the equipment which cor-responds to that proportion of unemployment which ensureszero saving.

The only alternative position of equilibrium would be givenby a situation in which a stock of capital sufficiently great tohave a marginal efficiency of zero also represents an amountof wealth sufficiently great to satiate to the full the aggregatedesire on the part of the public to make provision for the fu-ture, even with full employment, in circumstances where nobonus is obtainable in the form of interest. It would, how-ever, be an unlikely coincidence that the propensity to save inconditions of full employment should become satisfied just atthe point where the stock of capital reaches the level where itsmarginal efficiency is zero. If, therefore, this more favourablepossibility comes to the rescue, it will probably take effect, notjust at the point where the rate of interest is vanishing, but atsome previous point during the gradual decline of the rate ofinterest.

We have assumed so far an institutional factor which pre-vents the rate of interest from being negative, in the shapeof money which has negligible carrying costs. In fact, how-ever, institutional and psychological factors are present whichset a limit much above zero to the practicable decline in therate of interest. In particular the costs of bringing borrowersand lenders together and uncertainty as to the future of the

212

III.

rate of interest, which we have examined above, set a lowerlimit, which in present circumstances may perhaps be as highas 2 or 2 1/2 per cent. on long term. If this should provecorrect, the awkward possibilities of an increasing stock ofwealth, in conditions where the rate of interest can fall no fur-ther under laissez-faire, may soon be realised in actual experi-ence. Moreover if the minimum level to which it is practicableto bring the rate of interest is appreciably above zero, there isless likelihood of the aggregate desire to accumulate wealthbeing satiated before the rate of interest has reached its mini-mum level.

The post-war experiences of Great Britain and the UnitedStates are, indeed, actual examples of how an accumulationof wealth, so large that its marginal efficiency has fallen morerapidly than the rate of interest can fall in the face of the pre-vailing institutional and psychological factors, can interfere,in conditions mainly of laissez-faire, with a reasonable level ofemployment and with the standard of life which the technicalconditions of production are capable of furnishing.

It follows that of two equal communities, having the sametechnique but different stocks of capital, the community withthe smaller stock of capital may be able for the time being toenjoy a higher standard of life than the community with thelarger stock; though when the poorer community has caughtup the rich — as, presumably, it eventually will — then bothalike will suffer the fate of Midas. This disturbing conclusiondepends, of course, on the assumption that the propensity toconsume and the rate of investment are not deliberately con-trolled in the social interest but are mainly left to the influ-ences of laissez-faire.

If — for whatever reason — the rate of interest cannot fallas fast as the marginal efficiency of capital would fall witha rate of accumulation corresponding to what the communitywould choose to save at a rate of interest equal to the marginalefficiency of capital in conditions of full employment, theneven a diversion of the desire to hold wealth towards assets,which will in fact yield no economic fruits whatever, will in-crease economic well-being. In so far as millionaires find their

213


satisfaction in building mighty mansions to contain their bod-ies when alive and pyramids to shelter them after death, or,repenting of their sins, erect cathedrals and endow monaster-ies or foreign missions, the day when abundance of capitalwill interfere with abundance of output may be postponed.“To dig holes in the ground,” paid for out of savings, will in-crease, not only employment, but the real national dividend ofuseful goods and services. It is not reasonable, however, thata sensible community should be content to remain dependenton such fortuitous and often wasteful mitigations when oncewe understand the influences upon which effective demanddepends.

IV

Let us assume that steps are taken to ensure that the rate ofinterest is consistent with the rate of investment which cor-responds to full employment. Let us assume, further, thatState action enters in as a balancing factor to provide thatthe growth of capital equipment shall be such as to approachsaturation-point at a rate which does not put a disproportion-ate burden on the standard of life of the present generation.

On such assumptions I should guess that a properly runcommunity equipped with modern technical resources, of whichthe population is not increasing rapidly, ought to be able tobring down the marginal efficiency of capital in equilibriumapproximately to zero within a single generation; so that weshould attain the conditions of a quasi-stationary communitywhere change and progress would result only from changes intechnique, taste, population and institutions, with the prod-ucts of capital selling at a price proportioned to the labour,etc., embodied in them on just the same principles as governthe prices of consumption-goods into which capital-chargesenter in an insignificant degree.

If I am right in supposing it to be comparatively easy tomake capital-goods so abundant that the marginal efficiencyof capital is zero, this may be the most sensible way of grad-

214

IV.

ually getting rid of many of the objectionable features of cap-italism. For a little reflection will show what enormous socialchanges would result from a gradual disappearance of a rateof return on accumulated wealth. A man would still be free toaccumulate his earned income with a view to spending it at alater date. But his accumulation would not grow. He wouldsimply be in the position of Pope’s father, who, when he re-tired from business, carried a chest of guineas with him to hisvilla at Twickenham and met his household expenses from itas required.

Though the rentier would disappear, there would still beroom, nevertheless, for enterprise and skill in the estimationof prospective yields about which opinions could differ. Forthe above relates primarily to the pure rate of interest apartfrom any allowance for risk and the like, and not to the grossyield of assets including the return in respect of risk. Thusunless the pure rate of interest were to be held at a negativefigure, there would still be a positive yield to skilled invest-ment in individual assets having a doubtful prospective yield.Provided there was some measurable unwillingness to under-take risk, there would also be a positive net yield from theaggregate of such assets over a period of time. But it is notunlikely that, in such circumstances, the eagerness to obtaina yield from doubtful investments might be such that theywould show in the aggregate a negative net yield.

215

CHAPTER 17

THE ESSENTIAL PROPERTIES OF INTEREST ANDMONEY

I

It seems, then, that the rate of interest on money plays a peculiarpart in setting a limit to the level of employment, since it setsa standard to which the marginal efficiency of a capital-assetmust attain if it is to be newly produced. That this should beso, is, at first sight, most perplexing. It is natural to enquirewherein the peculiarity of money lies as distinct from otherassets, whether it is only money which has a rate of interest,and what would happen in a non-monetary economy. Untilwe have answered these questions, the full significance of ourtheory will not be clear.

The money-rate of interest — we may remind the reader —is nothing more than the percentage excess of a sum of moneycontracted for forward delivery, e.g. a year hence, over whatwe may call the “spot” or cash price of the sum thus con-tracted for forward delivery. It would seem, therefore, thatfor every kind of capital-asset there must be an analogue ofthe rate of interest on money. For there is a definite quan-tity of (e.g.) wheat to be delivered a year hence which hasthe same exchange value to-day as 1000 quarters of wheat for“spot” delivery. If the former quantity is 105 quarters, we maysay that the wheat-rate of interest is 5 per cent. per annum;and if it is 95 quarters, that it is minus 5 per cent. per annum.Thus for every durable commodity we have a rate of inter-est in terms of itself, — a wheat-rate of interest, a copper-rateof interest, a house-rate of interest, even a steel-plant-rate ofinterest.

The difference between the “future” and “spot” contracts

217

17. THE ESSENTIAL PROPERTIES OF INTEREST AND MONEY

for a commodity, such as wheat, which are quoted in the mar-ket, bears a definite relation to the wheat-rate of interest, but,since the future contract is quoted in terms of money for for-ward delivery and not in terms of wheat for spot delivery, italso brings in the money-rate of interest. The exact relation-ship is as follows:

Let us suppose that the spot price of wheat is £100 per 100quarters, that the price of the “future” contract for wheat fordelivery a year hence is £107 per 100 quarters, and that themoney-rate of interest is 5 per cent; what is the wheat-rate ofinterest? £100 spot will buy £105 for forward delivery, and£105 for forward delivery will buy (105/107).100 (=98) quar-ters for forward-delivery. Alternatively £100 spot will buy100 quarters of wheat for spot delivery. Thus 100 quarters ofwheat for spot delivery will buy 98 quarters for forward de-livery. It follows that the wheat-rate of interest is minus 2 percent. per annum.1

It follows from this that there is no reason why their rates ofinterest should be the same for different commodities, — whythe wheat-rate of interest should be equal to the copper-rateof interest. For the relation between the “spot” and “future”contracts, as quoted in the market, is notoriously different fordifferent commodities. This, we shall find, will lead us to theclue we are seeking. For it may be that it is the greatest of theown-rates of interest (as we may call them) which rules theroost (because it is the greatest of these rates that the marginalefficiency of a capital-asset must attain if it is to be newly pro-duced); and that there are reasons why it is the money-rate ofinterest which is often the greatest (because, as we shall find,certain forces, which operate to reduce the own-rates of inter-est of other assets, do not operate in the case of money).

It may be added that, just as there are differing commodity-rates of interest at any time, so also exchange dealers are fa-miliar with the fact that the rate of interest is not even the samein terms of two different moneys, e.g. sterling and dollars. For

1This relationship was first pointed out by Mr. Sraffa, Economic Journal,March 1932, p. 50.

218

I.

here also the difference between the “spot” and “future” con-tracts for a foreign money in terms of sterling are not, as a rule,the same for different foreign moneys.

Now each of these commodity standards offers us the samefacility as money for measuring the marginal efficiency of cap-ital. For we can take any commodity we choose, eg. wheat;calculate the wheat-value of the prospective yields of any cap-ital asset; and the rate of discount which makes the presentvalue of this series of wheat annuities equal to the presentsupply price of the asset in terms of wheat gives us the marginalefficiency of the asset in terms of wheat. If no change is ex-pected in the relative value of two alternative standards, thenthe marginal efficiency of a capital-asset will be the same inwhichever of the two standards it is measured, since the nu-merator and denominator of the fraction which leads up tothe marginal efficiency will be changed in the same propor-tion. If, however, one of the alternative standards is expectedto change in value in terms of the other, the marginal efficien-cies of capital-assets will be changed by the same percentage,according to which standard they are measured in. To illus-trate this let us take the simplest case where wheat, one of thealternative standards, is expected to appreciate at a steady rateof a per cent. per annum in terms of money; the marginal ef-ficiency of an asset, which is x per cent. in terms of money,will then be x − a per cent. in terms of wheat. Since themarginal efficiencies of all capital-assets will be altered by thesame amount, it follows that their order of magnitude will bethe same irrespective of the standard which is selected.

If there were some composite commodity which could beregarded strictly speaking as representative, we could regardthe rate of interest and the marginal efficiency of capital interms of this commodity as being, in a sense, uniquely therate of interest and the marginal efficiency of capital. But thereare, of course, the same obstacles in the way of this as thereare to setting up a unique standard of value.

So far, therefore, the money-rate of interest has no unique-ness compared with other rates of interest, but is on preciselythe same footing. Wherein, then, lies the peculiarity of the

219


money-rate of interest which gives it the predominating prac-tical importance attributed to it in the preceding chapters?Why should the volume of output and employment be moreintimately bound up with the money-rate of interest than withthe wheat-rate of interest or the house-rate of interest?

II

Let us consider what the various commodity-rates of interestover a period of (say) a year are likely to be for different typesof assets. Since we are taking each commodity in turn as thestandard, the returns on each commodity must be reckoned inthis context as being measured in terms of itself.

There are three attributes which different types of assetspossess in different degrees; namely, as follows:

(i) Some assets produce a yield or output q, measured interms of themselves, by assisting some process of productionor supplying services to a consumer.

(ii) Most assets, except money, suffer some wastage or in-volve some cost through the mere passage of time (apart fromany change in their relative value), irrespective of their beingused to produce a yield; i.e. they involve a carrying cost cmeasured in terms of themselves. It does not matter for ourpresent purpose exactly where we draw the line between thecosts which we deduct before calculating q and those whichwe include in c, since in what follows we shall be exclusivelyconcerned with q − c.

(iii) Finally, the power of disposal over an asset during a pe-riod may offer a potential convenience or security, which isnot equal for assets of different kinds, though the assets them-selves are of equal initial value. There is, so to speak, nothingto show for this at the end of the period in the shape of output;yet it is something for which people are ready to pay some-thing. The amount (measured in terms of itself) which theyare willing to pay for the potential convenience or securitygiven by this power of disposal (exclusive of yield or carrying

220

II.

cost attaching to the asset), we shall call its liquidity-premiuml.

It follows that the total return expected from the ownershipof an asset over a period is equal to its yield minus its carryingcost plus its liquidity-premium, i.e. to q − c+ l. That is to say,q − c + l is the own-rate of interest of any commodity, whereq, c and l are measured in terms of itself as the standard.

It is characteristic of instrumental capital (eg. a machine)or of consumption capital (eg. a house) which is in use, thatits yield should normally exceed its carrying cost, whilst itsliquidity-premium is probably negligible; of a stock of liquidgoods or of surplus laid-up instrumental or consumption cap-ital that it should incur a carrying cost in terms of itself with-out any yield to set off against it, the liquidity-premium inthis case also being usually negligible as soon as stocks ex-ceed a moderate level, though capable of being significant inspecial circumstances; and of money that its yield is nil, and itscarrying cost negligible, but its liquidity-premium substantial.Different commodities may, indeed, have differing degrees ofliquidity-premium amongst themselves, and money may in-cur some degree of carrying costs, eg. for safe custody. Butit is an essential difference between money and all (or most)other assets that in the case of money its liquidity-premiummuch exceeds its carrying cost, whereas in the case of other as-sets their carrying cost much exceeds their liquidity-premium.Let us, for purposes of illustration, assume that on houses theyield is q1 and the carrying cost and liquidity-premium negli-gible; that on wheat the carrying cost is c2 and the yield andliquidity-premium negligible; and that on money the liquidity-premium is l3 and the yield and carrying cost negligible. Thatis to say, q1 is the house-rate of interest, −c2 the wheat-rate ofinterest, and l3 the money-rate of interest.

To determine the relationships between the expected re-turns on different types of assets which are consistent withequilibrium, we must also know what the changes in rela-tive values during the year are expected to be. Taking money(which need only be a money of account for this purpose, and

221


we could equally well take wheat) as our standard of mea-surement, let the expected percentage appreciation (or depre-ciation) of houses be a1 and of wheat a2. q1, −c2 and l3 wehave called the own-rates of interest of houses, wheat andmoney in terms of themselves as the standard of value; i.e.q1 is the house-rate of interest in terms of houses, −c2 is thewheat-rate of interest in terms of wheat, and l3 is the money-rate of interest in terms of money. It will also be useful tocall a1 + q1, a2 − c2, and l3, which stand for the same quanti-ties reduced to money as the standard of value, the house-rateof money-interest, the wheat-rate of money-interest and themoney-rate of money-interest respectively. With this notationit is easy to see that the demand of wealth-owners will be di-rected to houses, to wheat or to money, according as a1 + q1or a2 − c2, or l3, is greatest. Thus in equilibrium the demand-prices of houses and wheat in terms of money will be suchthat there is nothing to choose in the way of advantage be-tween the alternatives; — i.e. a1 + q1, a2 − c2 and l3 will beequal. The choice of the standard of value will make no differ-ence to this result because a shift from one standard to anotherwill change all the terms equally, i.e. by an amount equal tothe expected rate of appreciation (or depreciation) of the newstandard in terms of the old.

Now those assets of which the normal supply-price is lessthan the demand-price will be newly produced, and thesewill be those assets of which the marginal efficiency wouldbe greater (on the basis of their normal supply-price) than therate of interest (both being measured in the same standard ofvalue whatever it is). As the stock of the assets, which be-gin by having a marginal efficiency at least equal to the rateof interest, is increased, their marginal efficiency (for reasons,sufficiently obvious, already given) tends to fall. Thus a pointwill come at which it no longer pays to produce them, unlessthe rate of interest falls pari passu. When there is no asset ofwhich the marginal efficiency reaches the rate of interest, thefurther production of capital-assets will come to a standstill.

Let us suppose (as a mere hypothesis at this stage of the ar-gument) that there is some asset (eg. money) of which the rate

222

II.

of interest is fixed (or declines more slowly as output increasesthan does any other commodity’s rate of interest); how is theposition adjusted? Since a1 +q1, a2−c2 and l3, are necessarilyequal, and since l3 by hypothesis is either fixed or falling moreslowly than q1 or−c2, it follows that a1 and a2, must be rising.In other words, the present money-price of every commodityother than money tends to fall relatively to its expected futureprice. Hence, if q1 and −c2, continue to fall, a point comes atwhich it is not profitable to produce any of the commodities,unless the cost of production at some future date is expectedto rise above the present cost by an amount which will coverthe cost of carrying a stock produced now to the date of theprospective higher price.

It is now apparent that our previous statement to the effectthat it is the money-rate of interest which sets a limit to therate of output, is not strictly correct. We should have said thatit is that asset’s rate of interest which declines most slowlyas the stock of assets in general increases, which eventuallyknocks out the profitable production of each of the others, —except in the contingency, just mentioned, of a special rela-tionship between the present and prospective costs of produc-tion. As output increases, own-rates of interest decline to lev-els at which one asset after another falls below the standardof profitable production; — until, finally, one or more own-rates of interest remain at a level which is above that of themarginal efficiency of any asset whatever.

If by money we mean the standard of value, it is clear that itis not necessarily the money-rate of interest which makes thetrouble. We could not get out of our difficulties (as some havesupposed) merely by decreeing that wheat or houses shall bethe standard of value instead of gold or sterling. For, it nowappears that the same difficulties will ensue if there continuesto exist any asset of which the own-rate of interest is reluctantto decline as output increases. It may be, for example, thatgold will continue to fill this role in a country which has goneover to an inconvertible paper standard.

223


III

In attributing, therefore, a peculiar significance to the money-rate of interest, we have been tacitly assuming that the kind ofmoney to which we are accustomed has some special charac-teristics which lead to its own rate of interest in terms of itselfas standard being more reluctant to fall as output increasesthan the own-rates of interest of any other assets in terms ofthemselves. Is this assumption justified? Reflection shows, Ithink, that the following peculiarities, which commonly char-acterise money as we know it, are capable of justifying it. Tothe extent that the established standard of value has these pe-culiarities, the summary statement, that it is the money-rateof interest which is the significant rate of interest, will holdgood.

(i) The first characteristic which tends towards the aboveconclusion is the fact that money has, both in the long and inthe short period, a zero, or at any rate a very small, elastic-ity of production, so far as the power of private enterprise isconcerned, as distinct from the monetary authority; — elastic-ity of production2 meaning, in this context, the response of thequantity of labour applied to producing it to a rise in the quan-tity of labour which a unit of it will command. Money, that isto say, cannot be readily produced; — labour cannot be turnedon at will by entrepreneurs to produce money in increasingquantities as its price rises in terms of the wage-unit. In thecase of an inconvertible managed currency this condition isstrictly satisfied. But in the case of a gold-standard currency itis also approximately so, in the sense that the maximum pro-portional addition to the quantity of labour which can be thusemployed is very small, except indeed in a country of whichgold-mining is the major industry.

Now, in the case of assets having an elasticity of produc-tion, the reason why we assumed their own-rate of interestto decline was because we assumed the stock of them to in-

2See Chapter 20.

224

III.

crease as the result of a higher rate of output. In the case ofmoney, however — postponing, for the moment, our consid-eration of the effects of reducing the wage-unit or of a delib-erate increase in its supply by the monetary authority — thesupply is fixed. Thus the characteristic that money cannot bereadily produced by labour gives at once some prima facie pre-sumption for the view that its own-rate of interest will be rel-atively reluctant to fall; whereas if money could be grown likea crop or manufactured like a motor-car, depressions wouldbe avoided or mitigated because, if the price of other assetswas tending to fall in terms of money, more labour wouldbe diverted into the production of money; — as we see to bethe case in gold-mining countries, though for the world as awhole the maximum diversion in this way is almost negligi-ble.

(ii) Obviously, however, the above condition is satisfied, notonly by money, but by all pure rent-factors, the production ofwhich is completely inelastic. A second condition, therefore,is required to distinguish money from other rent elements.

The second differentia of money is that it has an elasticityof substitution equal, or nearly equal, to zero; which meansthat as the exchange value of money rises there is no tendencyto substitute some other factor for it; — except, perhaps, tosome trifling extent, where the money-commodity is also usedin manufacture or the arts. This follows from the peculiarityof money that its utility is solely derived from its exchange-value, so that the two rise and fall pari passu, with the resultthat as the exchange value of money rises there is no motiveor tendency, as in the case of rent-factors, to substitute someother factor for it.

Thus, not only is it impossible to turn more labour on toproducing money when its labour-price rises, but money is abottomless sink for purchasing power, when the demand forit increases, since there is no value for it at which demand isdiverted — as in the case of other rent-factors — so as to slop

225


over into a demand for other things.3

The only qualification to this arises when the rise in thevalue of money leads to uncertainty as to the future mainte-nance of this rise; in which event, a1 and a2 are increased,which is tantamount to an increase in the commodity-rates ofmoney-interest and is, therefore, stimulating to the output ofother assets.

(iii) Thirdly, we must consider whether these conclusionsare upset by the fact that, even though the quantity of moneycannot be increased by diverting labour into producing it, nev-ertheless an assumption that its effective supply is rigidly fixedwould be inaccurate. In particular, a reduction of the wage-unit will release cash from its other uses for the satisfactionof the liquidity-motive; whilst, in addition to this, as money-values fall, the stock of money will bear a higher proportionto the total wealth of the community.

It is not possible to dispute on purely theoretical groundsthat this reaction might be capable of allowing an adequatedecline in the money-rate of interest. There are, however, sev-eral reasons, which taken in combination are of compellingforce, why in an economy of the type to which we are accus-tomed it is very probable that the money-rate of interest willoften prove reluctant to decline adequately:

(a) We have to allow, first of all, for the reactions of a fallin the wage-unit on the marginal efficiencies of other assets interms of money; — for it is the difference between these andthe money-rate of interest with which we are concerned. Ifthe effect of the fall in the wage-unit is to produce an expec-tation that it will subsequently rise again, the result will bewholly favourable. If, on the contrary, the effect is to producean expectation of a further fall, the reaction on the marginalefficiency of capital may offset the decline in the rate of inter-est.

3This is a matter which will be examined in greater detail in Chapter 19below.

226

III.

(b) The fact that wages tend to be sticky in terms of money,the money-wage being more stable than the real wage, tendsto limit the readiness of the wage-unit to fall in terms of money.Moreover, if this were not so, the position might be worserather than better; because, if money-wages were to fall easily,this might often tend to create an expectation of a further fallwith unfavourable reactions on the marginal efficiency of cap-ital. Furthermore, if wages were to be fixed in terms of someother commodity, eg. wheat, it is improbable that they wouldcontinue to be sticky. It is because of money’s other character-istics — those, especially, which make it liquid — that wages,when fixed in terms of it, tend to be sticky.4

(c) Thirdly, we come to what is the most fundamental con-sideration in this context, namely, the characteristics of moneywhich satisfy liquidity-preference. For, in certain circumstancessuch as will often occur, these will cause the rate of interest tobe insensitive, particularly below a certain figure,5 even to asubstantial increase in the quantity of money in proportionto other forms of wealth. In other words, beyond a certainpoint money’s yield from liquidity does not fall in response toan increase in its quantity to anything approaching the extentto which the yield from other types of assets falls when theirquantity is comparably increased.

In this connection the low (or negligible) carrying-costs ofmoney play an essential part. For if its carrying-costs werematerial, they would offset the effect of expectations as to theprospective value of money at future dates. The readiness ofthe public to increase their stock of money in response to acomparatively small stimulus is due to the advantages of liq-uidity (real or supposed) having no offset to contend with inthe shape of carrying-costs mounting steeply with the lapse oftime. In the case of a commodity other than money a modest

4If wages (and contracts) were fixed in terms of wheat, it might be thatwheat would acquire some of money’s liquidity-premium; — we will return tothis question in (IV) below.

5See p. 172 above.

227


stock of it may offer some convenience to users of the com-modity. But even though a larger stock might have some at-tractions as representing a store of wealth of stable value, thiswould be offset by its carrying-costs in the shape of storage,wastage, etc. Hence, after a certain point is reached, there isnecessarily a loss in holding a greater stock.

In the case of money, however, this, as we have seen, isnot so, — and for a variety of reasons, namely, those whichconstitute money as being, in the estimation of the public, parexcellence “liquid.” Thus those reformers, who look for a rem-edy by creating artificial carrying-costs for money through thedevice of requiring legal-tender currency to be periodicallystamped at a prescribed cost in order to retain its quality asmoney, or in analogous ways, have been on the right track;and the practical value of their proposals deserves considera-tion.

The significance of the money-rate of interest arises, there-fore, out of the combination of the characteristics that, throughthe working of the liquidity-motive, this rate of interest maybe somewhat unresponsive to a change in the proportion whichthe quantity of money bears to other forms of wealth mea-sured in money, and that money has (or may have) zero (ornegligible) elasticities both of production and of substitution.The first condition means that demand may be predominantlydirected to money, the second that when this occurs labourcannot be employed in producing more money, and the thirdthat there is no mitigation at any point through some otherfactor being capable, if it is sufficiently cheap, of doing money’sduty equally well. The only relief — apart from changes inthe marginal efficiency of capital — can come (so long as thepropensity towards liquidity is unchanged) from an increasein the quantity of money, or — which is formally the samething — a rise in the value of money which enables a givenquantity to provide increased money-services.

Thus a rise in the money-rate of interest retards the outputof all the objects of which the production is elastic withoutbeing capable of stimulating the output of money (the pro-duction of which is, by hypothesis, perfectly inelastic). The

228

III.

money-rate of interest, by setting the pace for all the othercommodity-rates of interest, holds back investment in the pro-duction of these other commodities without being capable ofstimulating investment for the production of money, whichby hypothesis cannot be produced. Moreover, owing to theelasticity of demand for liquid cash in terms of debts, a smallchange in the conditions governing this demand may not muchalter the money-rate of interest, whilst (apart from official ac-tion) it is also impracticable, owing to the inelasticity of theproduction of money, for natural forces to bring the money-rate of interest down by affecting the supply side. In the caseof an ordinary commodity, the inelasticity of the demand forliquid stocks of it would enable small changes on the demandside to bring its rate of interest up or down with a rush, whilstthe elasticity of its supply would also tend to prevent a highpremium on spot over forward delivery. Thus with other com-modities left to themselves, “natural forces,” i.e. the ordi-nary forces of the market, would tend to bring their rate ofinterest down until the emergence of full employment hadbrought about for commodities generally the inelasticity ofsupply which we have postulated as a normal characteristicof money. Thus in the absence of money and in the absence —we must, of course, also suppose — of any other commoditywith the assumed characteristics of money, the rates of interestwould only reach equilibrium when there is full employment.

Unemployment develops, that is to say, because peoplewant the moon; — men cannot be employed when the objectof desire (i.e. money) is something which cannot be producedand the demand for which cannot be readily choked off. Thereis no remedy but to persuade the public that green cheese ispractically the same thing and to have a green cheese factory(i.e. a central bank) under public control.

It is interesting to notice that the characteristic which hasbeen traditionally supposed to render gold especially suitablefor use as the standard of value, namely, its inelasticity of sup-ply, turns out to be precisely the characteristic which is at thebottom of the trouble.

229


Our conclusion can be stated in the most general form (tak-ing the propensity to consume as given) as follows. No furtherincrease in the rate of investment is possible when the greatestamongst the own-rates of own-interest of all available assetsis equal to the greatest amongst the marginal efficiencies ofall assets, measured in terms of the asset whose own-rate ofown-interest is greatest.

In a position of full employment this condition is neces-sarily satisfied. But it may also be satisfied before full em-ployment is reached, if there exists some asset, having zero(or relatively small) elasticities of production and substitu-tion,6 whose rate of interest declines more slowly, as outputincreases, than the marginal efficiencies of capital-assets mea-sured in terms of it.

IV

We have shown above that for a commodity to be the standardof value is not a sufficient condition for that commodity’s rateof interest to be the significant rate of interest. It is, however,interesting to consider how far those characteristics of moneyas we know it, which make the money-rate of interest the sig-nificant rate, are bound up with money being the standard inwhich debts and wages are usually fixed. The matter requiresconsideration under two aspects.

In the first place, the fact that contracts are fixed, and wagesare usually somewhat stable, in terms of money unquestion-ably plays a large part in attracting to money so high a liquidity-premium. The convenience of holding assets in the same stan-dard as that in which future liabilities may fall due and in astandard in terms of which the future cost of living is expectedto be relatively stable, is obvious. At the same time the expec-tation of relative stability in the future money-cost of outputmight not be entertained with much confidence ff the stan-dard of value were a commodity with a high elasticity of pro-

6A zero elasticity is a more stringent condition than is necessarily required.

230

IV.

duction. Moreover, the low carrying-costs of money as weknow it play quite as large a part as a high liquidity-premiumin making the money-rate of interest the significant rate. Forwhat matters is the difference between the liquidity-premiumand the carrying-costs; and in the case of most commodities,other than such assets as gold and silver and bank-notes, thecarrying-costs are at least as high as the liquidity-premiumordinarily attaching to the standard in which contracts andwages are fixed, so that, even if the liquidity-premium nowattaching to (e.g.) sterling-money were to be transferred to(eg.) wheat, the wheat-rate of interest would still be unlikelyto rise above zero. It remains the case, therefore, that, whilstthe fact of contracts and wages being fixed in terms of moneyconsiderably enhances the significance of the money-rate ofinterest, this circumstance is, nevertheless, probably insuffi-cient by itself to produce the observed characteristics of themoney-rate of interest.

The second point to be considered is more subtle. The nor-mal expectation that the value of output will be more sta-ble in terms of money than in terms of any other commod-ity, depends of course, not on wages being arranged in termsof money, but on wages being relatively sticky in terms ofmoney. What, then, would the position be if wages were ex-pected to be more sticky (i.e. more stable) in terms of someone or more commodities other than money, than in termsof money itself? Such an expectation requires, not only thatthe costs of the commodity in question are expected to berelatively constant in terms of the wage-unit for a greater orsmaller scale of output both in the short and in the long pe-riod, but also that any surplus over the current demand atcost-price can be taken into stock without cost, i.e. that itsliquidity-premium exceeds its carrying-costs (for, otherwise,since there is no hope of profit from a higher price, the carry-ing of a stock must necessarily involve a loss). If a commod-ity can be found to satisfy these conditions, then, assuredly, itmight be set up as a rival to money. Thus it is not logically im-possible that there should be a commodity in terms of whichthe value of output is expected to be more stable than in terms

231


of money. But it does not seem probable that any such com-modity exists.

I conclude, therefore, that the commodity, in terms of whichwages are expected to be most sticky, cannot be one whoseelasticity of production is not least, and for which the excessof carrying-costs over liquidity-premium is not least. In otherwords, the expectation of a relative stickiness of wages in termsof money is a corollary of the excess of liquidity-premiumover carrying-costs being greater for money than for any otherasset.

Thus we see that the various characteristics, which com-bine to make the money-rate of interest significant, interactwith one another in a cumulative fashion. The fact that moneyhas low elasticities of production and substitution and lowcarrying-costs tends to raise the expectation that will be rela-tively stable; and this expectation enhances money’s liquidity-premium and prevents the exceptional correlation betweenthe money-rate of interest and the marginal efficiencies of otherassets which might, if it could exist, rob the money-rate of in-terest of its sting.

Professor Pigou (with others) has been accustomed to as-sume that there is a presumption in favour of real wages be-ing more stable than money-wages. But this could only bethe case if there were a presumption in favour of stability ofemployment. Moreover, there is also the difficulty that wage-goods have a high carrying-cost. If, indeed, some attemptwere made to stabilise real wages by fixing wages in termsof wage-goods, the effect could only be to cause a violent os-cillation of money-prices. For every small fluctuation in thepropensity to consume and the inducement to invest wouldcause money-prices to rush violently between zero and infin-ity. That money-wages should be more stable than real wagesis a condition of the system possessing inherent stability.

Thus the attribution of relative stability to real wages is notmerely a mistake in fact and experience. It is also a mistakein logic, if we are supposing that the system in view is stable,in the sense that small changes in the propensity to consumeand the inducement to invest do not produce violent effects

232

V.

on prices.

V

As a footnote to the above, it may be worth emphasising whathas been already stated above, namely, that “liquidity” and“carrying-costs” are both a matter of degree; and that it is oilyin having the former high relatively to the latter that the pe-culiarity of “money” consists.

Consider, for example, an economy in which there is noasset for which the liquidity-premium is always in excess ofthe carrying-costs; which is the best definition I can give ofa so-called “non-monetary” economy. There exists nothing,that is to say, but particular consumables and particular cap-ital equipments more or less differentiated according to thecharacter of the consumables which they can yield up, or as-sist to yield up, over a greater or a shorter period of time; all ofwhich, unlike cash, deteriorate or involve expense, if they arekept in stock, to a value in excess of any liquidity-premiumwhich may attach to them.

In such an economy capital equipments will differ fromone another (a) in the variety of the consumables in the pro-duction of which they are capable of assisting, (b) in the stabil-ity of value of their output (in the sense in which the value ofbread is more stable through time than the value of fashion-able novelties), and (c) in the rapidity with which the wealthembodied in them can become “liquid”, in the sense of pro-ducing output, the proceeds of which can be re-embodied ifdesired in quite a different form.

The owners of wealth will then weigh the lack of “liquid-ity” of different capital equipments in the above sense as amedium in which to hold wealth against the best available ac-tuarial estimate of their prospective yields after allowing forrisk. The liquidity-premium, it will be observed, is partly sim-ilar to the risk-premium, but partly different; — the differencecorresponding to the difference between the best estimates wecan make of probabilities and the confidence with which we

233


make them.7 When we were dealing, in earlier chapters, withthe estimation of prospective yield, we did not enter into de-tail as to how the estimation is made: and to avoid compli-cating the argument, we did not distinguish differences in liq-uidity from differences in risk proper. It is evident, however,that in calculating the own-rate of interest we must allow forboth.

There is, clearly, no absolute standard of “liquidity” butmerely a scale of liquidity — a varying premium of whichaccount has to be taken, in addition to the yield of use andthe carrying-costs, in estimating the comparative attractionsof holding different forms of wealth. The conception of whatcontributes to “liquidity” is a partly vague one, changing fromtime to time and depending on social practices and institu-tions. The order of preference in the minds of owners of wealthin which at any given time they express their feelings aboutliquidity is, however, definite and is all we require for ouranalysis of the behaviour of the economic system.

It may be that in certain historic environments the pos-session of land has been characterised by a high liquidity-premium in the minds of owners of wealth; and since landresembles money in that its elasticities of production and sub-stitution may be very low,8 it is conceivable that there havebeen occasions in history in which the desire to hold landhas played the same role in keeping up the rate of interestat too high a level which money has played in recent times. Itis difficult to trace this influence quantitatively owing to theabsence of a forward price for land in terms of itself whichis strictly comparable with the rate of interest on a moneydebt. We have, however, something which has, at times, beenclosely analogous, in the shape of high rates of interest on

7Cf. the footnote to p. 148 above.8The attribute of “liquidity” is by no means independent of the presence of

these two characteristics. For it is unlikely that an asset, of which the supply canbe easily increased or the desire for which can be easily diverted by a change inrelative price, will possess the attribute of “liquidity” in the minds of owners ofwealth. Money itself rapidly loses the attribute of “liquidity” if its future supplyis expected to undergo sharp changes.

234

VI.

mortgages.9 The high rates of interest from mortgages onland, often exceeding the probable net yield from cultivat-ing the land, have been a familiar feature of many agricul-tural economies. Usury laws have been directed primarilyagainst encumbrances of this character. And rightly so. Forin earlier social organisations where long-term bonds in themodern sense were non-existent, the competition of a highinterest-rate on mortgages may well have had the same effectin retarding the growth of wealth from current investment innewly produced capital-assets, as high interest rates on long-term debts have had in more recent times.

That the world after several millennia of steady individualsaving, is so poor as it is in accumulated capital-assets, is to beexplained, in my opinion, neither by the improvident propen-sities of mankind, nor even by the destruction of war, but bythe high liquidity-premiums formerly attaching to the owner-ship of land and now attaching to money. I differ in this fromthe older view as expressed by Marshall with an unusual dog-matic force in his Principles of Economics, p. 581: —

Everyone is aware that the accumulation of wealthis held in check, and the rate of interest so far sus-tained, by the preference which the great mass ofhumanity have for present over deferred gratifica-tions, or, in other words, by their unwillingness to“wait”.

VI

In my Treatise on Money I defined what purported to be a uniquerate of interest, which I called the natural rate of interest —

9A mortgage and the interest thereon are, indeed, fixed in terms of money.But the fact that the mortgagor has the option to deliver the land itself in dis-charge of the debt — and must so deliver it if he cannot find the money on de-mand — has sometimes made the mortgage system approximate to a contractof land for future delivery against land for spot delivery. There have been salesof lands to tenants against mortgages effected by them, which, in fact, camevery near to being transactions of this character.

235


namely, the rate of interest which, in the terminology of myTreatise, preserved equality between the rate of saving (as theredefined) and the rate of investment. I believed this to be adevelopment and clarification of Wicksell’s “natural rate ofinterest”, which was, according to him, the rate which wouldpreserve the stability if some, not quite clearly specified, price-level.

I had, however, overlooked the fact that in any given soci-ety there is, on this definition, a different natural rate of interestfor each hypothetical level of employment. And, similarly, forevery rate of interest there is a level of employment for whichthat rate is the “natural” rate, in the sense that the system willbe in equilibrium with that rate of interest and that level ofemployment. Thus it was a mistake to speak of the naturalrate of interest or to suggest that the above definition wouldyield a unique value for the rate of interest irrespective of thelevel of employment. I had not then understood that, in cer-tain conditions, the system could be in equilibrium with lessthan full employment.

I am now no longer of the opinion that the concept of a“natural” rate of interest, which previously seemed to me amost promising idea, has anything very useful or significantto contribute to our analysis. It is merely the rate of interestwhich will preserve the status quo; and, in general, we haveno predominant interest in the status quo as such.

If there is any such rate of interest, which is unique and sig-nificant, it must be the rate which we might term the neutralrate of interest,10 namely, the natural rate in the above sensewhich is consistent with full employment, given the other pa-rameters of the system; though this rate might be better de-scribed, perhaps, as the optimum rate.

The neutral rate of interest can be more strictly defined asthe rate of interest which prevails in equilibrium when outputand employment are such that the elasticity of employment as

10This definition does not correspond to any of the various definitions ofneutral money given by recent writers; though it may, perhaps, have some rela-tion to the objective which these writers have had in mind.

236

VI.

a whole is zero.11

The above gives us, once again, the answer to the questionas to what tacit assumption is required to make sense of theclassical theory of the rate of interest. This theory assumeseither that the actual rate of interest is always equal to theneutral rate of interest in the sense in which we have just de-fined the latter, or alternatively that the actual rate of interestis always equal to the rate of interest which will maintain em-ployment at some specified constant level. If the traditionaltheory is thus interpreted, there is little or nothing in its practi-cal conclusions to which we need take exception. The classicaltheory assumes that the banking authority or natural forcescause the market-rate of interest to satisfy one or other of theabove conditions; and it investigates what laws will governthe application and rewards of the community’s productiveresources subject to this assumption. With this limitation inforce, the volume of output depends solely on the assumedconstant level of employment in conjunction with the currentequipment and technique; and we are safely ensconced in aRicardian world.


237

CHAPTER 18

THE GENERAL THEORY OF EMPLOYMENTRE-STATED

I

We have now reached a point where we can gather togetherthe threads of our argument. To begin with, it may be usefulto make clear which elements in the economic system we usu-ally take as given, which are the independent variables of oursystem and which are the dependent variables.

We take as given the existing skill and quantity of availablelabour, the existing quality and quantity of available equip-ment, the existing technique, the degree of competition, thetastes and habits of the consumer, the disutility of differentintensifies of labour and of the activities of supervision and or-ganisation, as well as the social structure including the forces,other than our variables set forth below, which determine thedistribution of the national income. This does not mean thatwe assume these factors to be constant; but merely that, inthis place and context, we are not considering or taking intoaccount the effects and consequences of changes in them.

Our independent variables are, in the first instance, thepropensity to consume, the schedule of the marginal efficiencyof capital and the rate of interest, though, as we have alreadyseen, these are capable of further analysis.

Our dependent variables are the volume of employmentand the national income (or national dividend) measured inwage-units.

The factors, which we have taken as given, influence ourindependent variables, but do not completely determine them.For example, the schedule of the marginal efficiency of capitaldepends partly on the existing quantity of equipment which

239

18. THE GENERAL THEORY OF EMPLOYMENT RE-STATED

is one of the given factors, but partly on the state of long-termexpectation which cannot be inferred from the given factors.But there are certain other elements which the given factorsdetermine so completely that we can treat these derivatives asbeing themselves given. For example, the given factors allowus to infer what level of national income measured in terms ofthe wage-unit will correspond to any given level of employ-ment; so that, within the economic framework which we takeas given, the national income depends on the volume of em-ployment, i.e. on the quantity of effort currently devoted toproduction, in the sense that there is a unique correlation be-tween the two.1 Furthermore, they allow us to infer the shapeof the aggregate supply functions, which embody the physicalconditions of supply, for different types of products; — that isto say, the quantity of employment which will be devoted toproduction corresponding to any given level of effective de-mand measured in terms of wage-units. Finally, they furnishus with the supply function of labour (or effort); so that theytell us inter alia at what point the employment function2 forlabour as a whole will cease to be elastic.

The schedule of the marginal efficiency of capital depends,however, partly on the given factors and partly on the prospec-tive yield of capital-assets of different kinds; whilst the rateof interest depends partly on the state of liquidity-preference(i.e. on the liquidity function) and partly on the quantity ofmoney measured in terms of wage-units. Thus we can some-times regard our ultimate independent variables as consistingof (1) the three fundamental psychological factors, namely, thepsychological propensity to consume, the psychological atti-tude to liquidity and the psychological expectation of futureyield from capital-assets, (2) the wage-unit as determined bythe bargains reached between employers and employed, and(3) the quantity of money as determined by the action of the

1We are ignoring at this stage certain complications which arise when theemployment functions of different products have different curvatures withinthe relevant range of employment. See Chapter 20 below.

2Defined in Chapter 20 below.

240

II.

central bank; so that, if we take as given the factors speci-fied above, these variables determine the national income (ordividend) and the quantity of employment. But these againwould be capable of being subjected to further analysis, andare not, so to speak, our ultimate atomic independent ele-ments.

The division of the determinants of the economic systeminto the two groups of given factors and independent vari-ables is, of course, quite arbitrary from any absolute stand-point. The division must be made entirely on the basis ofexperience, so as to correspond on the one hand to the fac-tors in which the changes seem to be so slow or so little rel-evant as to have only a small and comparatively negligibleshort-term influence on our quæsitum; and on the other handto those factors in which the changes are found in practice toexercise a dominant influence on our quæsitum. Our presentobject is to discover what determines at any time the nationalincome of a given economic system and (which is almost thesame thing) the amount of its employment; which means in astudy so complex as economics, in which we cannot hope tomake completely accurate generalisations, the factors whosechanges mainly determine our quæsitum. Our final task mightbe to select those variables which can be deliberately controlledor managed by central authority in the kind of system in whichwe actually live.

II

Let us now attempt to summarise the argument of the previ-ous chapters; taking the factors in the reverse order to that inwhich we have introduced them.

There will be an inducement to push the rate of new invest-ment to the point which forces the supply-price of each typeof capital-asset to a figure which, taken in conjunction with itsprospective yield, brings the marginal efficiency of capital ingeneral to approximate equality with the rate of interest. Thatis to say, the physical conditions of supply in the capital-goods

241


industries, the state of confidence concerning the prospectiveyield, the psychological attitude to liquidity and the quantityof money (preferably calculated in terms of wage-units) deter-mine, between them, the rate of new investment.

But an increase (or decrease) in the rate of investment willhave to carry with it an increase (or decrease) in the rate ofconsumption; because the behaviour of the public is, in gen-eral, of such a character that they are only willing to widen(or narrow) the gap between their income and their consump-tion if their income is being increased (or diminished). Thatis to say, changes in the rate of consumption are, in general,in the same direction (though smaller in amount) as changes inthe rate of income. The relation between the increment of con-sumption which has to accompany a given increment of sav-ing is given by the marginal propensity to consume. The ratio,thus determined, between an increment of investment and thecorresponding increment of aggregate income, both measuredin wage-units, is given by the investment multiplier.

Finally, if we assume (as a first approximation) that the em-ployment multiplier is equal to the investment multiplier, wecan, by applying the multiplier to the increment (or decre-ment) in the rate of investment brought about by the factorsfirst described, infer the increment of employment.

An increment (or decrement) of employment is liable, how-ever, to raise (or lower) the schedule of liquidity-preference;there being three ways in which it will tend to increase thedemand for money, inasmuch as the value of output will risewhen employment increases even if the wage-unit and prices(in terms of the wage-unit) are unchanged, but, in addition,the wage-unit itself will tend to rise as employment improves,and the increase in output will be accompanied by a rise ofprices (in terms of the wage-unit) owing to increasing cost inthe short period.

Thus the position of equilibrium will be influenced by theserepercussions; and there are other repercussions also. More-over, there is not one of the above factors which is not liable tochange without much warning, and sometimes substantially.Hence the extreme complexity of the actual course of events.

242

III.

Nevertheless, these seem to be the factors which it is usefuland convenient to isolate. If we examine any actual problemalong the lines of the above schematism, we shall find it moremanageable; and our practical intuition (which can take ac-count of a more detailed complex of facts than can be treatedon general principles) will be offered a less intractable mate-rial upon which to work.

III

The above is a summary of the General Theory. But the ac-tual phenomena of the economic system are also coloured bycertain special characteristics of the propensity to consume,the schedule of the marginal efficiency of capital and the rateof interest, about which we can safely generalise from experi-ence, but which are not logically necessary.

In particular, it is an outstanding characteristic of the eco-nomic system in which we live that, whilst it is subject to se-vere fluctuations in respect of output and employment, it isnot violently unstable. Indeed it seems capable of remain-ing in a chronic condition of sub-normal activity for a consid-erable period without any marked tendency either towardsrecovery or towards complete collapse. Moreover, the evi-dence indicates that full, or even approximately full, employ-ment is of rare and short-lived occurrence. Fluctuations maystart briskly but seem to wear themselves out before they haveproceeded to great extremes, and an intermediate situationwhich is neither desperate nor satisfactory is our normal lot.It is upon the fact that fluctuations tend to wear themselvesout before proceeding to extremes and eventually to reversethemselves, that the theory of business cycles having a regu-lar phase has been founded. The same thing is true of prices,which, in response to an initiating cause of disturbance, seemto be able to find a level at which they can remain, for the timebeing, moderately stable.

Now, since these facts of experience do not follow of logi-cal necessity, one must suppose that the environment and the

243


psychological propensities of the modern world must be ofsuch a character as to produce these results. It is, therefore,useful to consider what hypothetical psychological propensi-ties would lead to a stable system; and, then, whether thesepropensities can be plausibly ascribed, on our general knowl-edge of contemporary human nature, to the world in whichwe live.

The conditions of stability which the foregoing analysissuggests to us as capable of explaining the observed resultsare the following:

(i) The marginal propensity to consume is such that, whenthe output of a given community increases (or decreases)because more (or less) employment is being applied toits capital equipment, the multiplier relating the two isgreater than unity but not very large.

(ii) When there is a change in the prospective yield of capitalor in the rate of interest, the schedule of the marginal ef-ficiency of capital will be such that the change in new in-vestment will not be in great disproportion to the changein the former ; i.e. moderate changes in the prospectiveyield of capital or in the rate of interest will not be asso-ciated with very great changes in the rate of investment.

(iii) When there is a change in employment, money-wagestend to change in the same direction as, but not in greatdisproportion to, the change in employment; i.e. moder-ate changes in employment are not associated with verygreat chances in money-wages. This is a condition of thestability of prices rather than of employment.

(iv) We may add a fourth condition, which provides not somuch for the stability of the system as for the tendencyof a fluctuation in one direction to reverse itself in duecourse; namely, that a rate of investment, higher (or lower)than prevailed formerly, begins to react unfavourably (orfavourably) on the marginal efficiency of capital if it iscontinued for a period which, measured in years, is notvery large.

244

III.

(i) Our first condition of stability, namely, that the multi-plier, whilst greater than unity, is not very great, is highlyplausible as a psychological characteristic of human nature.As real income increases, both the pressure of present needsdiminishes and the margin over the established standard oflife is increased; and as real income diminishes the oppositeis true. Thus it is natural — at any rate on the average of thecommunity — that current consumption should be expandedwhen employment increases) but by less than the full incre-ment of real income; and that it should be diminished whenemployment diminishes, but by less than the full decrementof real income. Moreover, what is true of the average of in-dividuals is likely to be also true of governments especiallyin an age when a progressive increase of unemployment willusually force the State to provide relief out of borrowed funds.

But whether or not this psychological law strikes the readeras plausible a priori, it is certain that experience would be ex-tremely different from what it is if the law did not hold. Forin that case an increase of investment, however small, wouldset moving a cumulative increase of effective demand untila position of full employment had been reached; while a de-crease of investment would set moving a cumulative decreaseof effective demand until no one at all was employed. Yet ex-perience shows that we are generally in an intermediate po-sition. It is not impossible that there may be a range withinwhich instability does in fact prevail. But, if so, it is probablya narrow one, outside of which in either direction our psy-chological law must unquestionably hold good. Furthermore,it is also evident that the multiplier, though exceeding unity,is not, in normal circumstances, enormously large. For, if itwere, a given change in the rate of investment would involvea great change (limited only by full or zero employment) inthe rate of consumption.

(ii) Whilst our first condition provides that a moderate changein the rate of investment will not involve an indefinitely greatchange in the demand for consumption-goods our second con-dition provides that a moderate change in the prospective yield

245


of capital-assets or in the rate of interest will not involve anindefinitely great change in the rate of investment. This islikely to be the case owing to the increasing cost of produc-ing a greatly enlarged output from the existing equipment. If,indeed, we start from a position where there are very largesurplus resources for the production of capital-assets, theremay be considerable instability within a certain range; but thiswill cease to hold good as soon as the surplus is being largelyutilised. Moreover, this condition sets a limit to the instabil-ity resulting from rapid changes in the prospective yield ofcapital-assets due to sharp fluctuations in business psychol-ogy or to epoch-making inventions — though more, perhaps,in the upward than in the downward direction.

(iii) Our third condition accords with our experience of hu-man nature. For although the struggle for money-wages is, aswe have pointed out above, essentially a struggle to maintaina high relative wage, this struggle is likely, as employment in-creases, to be intensified in each individual case both becausethe bargaining position of the worker is improved and be-cause the diminished marginal utility of his wage and his im-proved financial margin make him readier to run risks. Yet, allthe same, these motives will operate within limits, and work-ers will not seek a much greater money-wage when employ-ment improves or allow a very great reduction rather than suf-fer any unemployment at all.

But here again, whether or not this conclusion is plausiblea priori, experience shows that some such psychological lawmust actually hold. For if competition between unemployedworkers always led to a very great reduction of the money-wage, there would be a violent instability in the price-level.Moreover, there might be no position of stable equilibrium ex-cept in conditions consistent with full employment; since thewage-unit might have to fall without limit until it reached apoint where the effect of the abundance of money in terms ofthe wage-unit on the rate of interest was sufficient to restore alevel of full employment. At no other point could there be a

246

III.

resting-place.3

(iv) Our fourth condition, which is a condition not so muchof stability as of alternate recession and recovery, is merelybased on the presumption that capital-assets are of variousages, wear out with time and are not all very long-lived; sothat if the rate of investment falls below a certain minimumlevel, it is merely a question of time (failing large fluctuationsin other factors) before the marginal efficiency of capital risessufficiently to bring about a recovery of investment above thisminimum. And similarly, of course, if investment rises to ahigher figure than formerly, it is only a question of time be-fore the marginal efficiency of capital falls sufficiently to bringabout a recession unless there are compensating changes inother factors.

For this reason, even those degrees of recovery and reces-sion, which can occur within the limitations set by our otherconditions of stability, will be likely, if they persist for a suf-ficient length of time and are not interfered with by changesin the other factors, to cause a reverse movement in the oppo-site direction, until the same forces as before again reverse thedirection.

Thus our four conditions together are adequate to explainthe outstanding features of our actual experience; — namely,that we oscillate, avoiding the gravest extremes of fluctuationin employment and in prices in both directions, round an in-termediate position appreciably below full employment andappreciably above the minimum employment a decline belowwhich would endanger life.

But we must not conclude that the mean position thus de-termined by “natural” tendencies, namely, by those tenden-cies which are likely to persist, failing measures expressly de-signed to correct them, is, therefore, established by laws ofnecessity. The unimpeded rule of the above conditions is a

3The effects of changes in the wage-unit will be considered in detail inChapter 19.

247


fact of observation concerning the world as it is or has been,and not a necessary principle which cannot be changed.

248

Book V

Money-Wages and Prices

CHAPTER 19

CHANGES IN MONEY-WAGES

I

It would have been an advantage if the effects of a change inmoney-wages could have been discussed in an earlier chapter.For the Classical Theory has been accustomed to rest the sup-posedly self-adjusting character of the economic system on anassumed fluidity of money-wages; and, when there is rigidity,to lay on this rigidity the blame of maladjustment.

It was not possible, however, to discuss this matter fullyuntil our own theory had been developed. For the conse-quences of a change in money-wages are complicated. A re-duction in money-wages is quite capable in certain circum-stances of affording a stimulus to output, as the classical the-ory supposes. My difference from this theory is primarily adifference of analysis; so that it could not be set forth clearlyuntil the reader was acquainted with my own method.

The generally accepted explanation is, as I understand it,quite a simple one. It does not depend on roundabout reper-cussions, such as we shall discuss below. The argument sim-ply is that a reduction in money-wages will cet. par. stimu-late demand by diminishing the price of the finished product,and will therefore increase output and employment up to thepoint where the reduction which labour has agreed to acceptin its money-wages is just offset by the diminishing marginalefficiency of labour as output (from a given equipment) is in-creased.

In its crudest form, this is tantamount to assuming thatthe reduction in money-wages will leave demand unaffected.There may be some economists who would maintain that thereis no reason why demand should be affected, arguing that ag-

251

19. CHANGES IN MONEY-WAGES

gregate demand depends on the quantity of money multipliedby the income-velocity of money and that there is no obvi-ous reason why a reduction in money-wages would reduceeither the quantity of money or its income-velocity. Or theymay even argue that profits will necessarily go up becausewages have gone down. But it would, I think, be more usualto degree that the reduction in money-wages may have someeffect on aggregate demand through its reducing the purchas-ing power of some of the workers, but that the real demand ofother factors, whose money incomes have not been reduced,will be stimulated by the fall in prices, and that the aggre-gate demand of the workers themselves will be very likelyincreased as a result of the increased volume of employment,unless the elasticity of demand for labour in response to changesin money-wages is less than unity. Thus in the new equilib-rium there will be more employment than there would havebeen otherwise except, perhaps, in some unusual limiting casewhich has no reality in practice.

It is from this type of analysis that I fundamentally differ;or rather from the analysis which seems to lie behind such ob-servations as the above. For whilst the above fairly represents,I think, the way in which many economists talk and write, theunderlying analysis has seldom been written down in detail.

It appears, however, that this way of thinking is probablyreached as follows. In any given industry we have a demandschedule for the product relating the quantities which can besold to the prices asked; we have a series of supply schedulesrelating the prices which will be asked for the sale of differ-ent quantities on various bases of cost; and these schedulesbetween them lead up to a further schedule which, on the as-sumption that other costs are unchanged (except as a resultof the change in output), gives us the demand schedule forlabour in the industry relating the quantity of employment todifferent levels of wages, the shape of the curve at any pointfurnishing the elasticity of demand for labour. This concep-tion is then transferred without substantial modification to in-dustry as a whole; and it is supposed, by a parity of reason-ing, that we have a demand schedule for labour in industry as

252

I.

a whole relating the quantity of employment to different lev-els of wages. It is held that it makes no material difference tothis argument whether it is in terms of money-wages or of realwages. If we are thinking in terms of money-wages, we must,of course, correct for changes in the value of money; but thisleaves the general tendency of the argument unchanged, sinceprices certainly do not change in exact proportion to changesin money-wages.

If this is the groundwork of the argument (and, if it is not,I do not know what the groundwork is), surely it is fallacious.For the demand schedules for particular industries can onlybe constructed on some fixed assumption as to the nature ofthe demand and supply schedules of other industries and asto the amount of the aggregate effective demand. It is invalid,therefore, to transfer the argument to industry as a whole un-less we also transfer our assumption that the aggregate effec-tive demand is fixed. Yet this assumption reduces the argu-ment to an ignoratio elenchi. For, whilst no one would wishto deny the proposition that a reduction in money-wages ac-companied by the same ggregate effective demand as beforewill be associated with an increase in employment, the precisequestion at issue is whether the reduction in money-wageswill or will not be accompanied by the same aggregate effectivedemand as before measured in money, or, at any rate, by an ag-gregate effective demand which is not reduced in full propor-tion to the reduction in money-wages (i.e. which is somewhatgreater measured in wage-units). But if the classical theoryis not allowed to extend by analogy its conclusions in respectof a particular industry to industry as a whole, it is whollyunable to answer the question what effect on employment areduction in money-wages will have. For it has no method ofanalysis wherewith to tackle the problem. Professor Pigou’sTheory of Unemployment seems to me to get out of the Classi-cal Theory all that can be got out of it; with the result that thebook becomes a striking demonstration that this theory hasnothing to offer, when it is applied to the problem of what

253


determines the volume of actual employment as a whole.1

II

Let us, then, apply our own method of analysis to answeringthe problem. It falls into two parts. (1) Does a reduction inmoney-wages have a direct tendency, cet. par., to increase em-ployment, “cet. par.” being taken to mean that the propensityto consume, the schedule of the marginal efficiency of capitaland the rate of interest are the same as before for the commu-nity as a whole? And (2) does a reduction in money-wageshave a certain or probable tendency to affect employment in aparticular direction through its certain or probable repercus-sions on these three factors?

The first question we have already answered in the nega-tive in the preceding chapters. For we have shown that thevolume of employment is uniquely correlated with the vol-ume of effective demand measured in wage-units, and thatthe effective demand, being the sum of the expected consump-tion and the expected investment, cannot change, if the propen-sity to consume, the schedule of marginal efficiency of capitaland the rate of interest are all unchanged. If, without anychange in these factors, the entrepreneurs were to increaseemployment as a whole, their proceeds will necessarily fallshort of their supply-price.

Perhaps it will help to rebut the crude conclusion that a re-duction in money-wages will increase employment “becauseit reduces the cost of production”, if we follow up the courseof events on the hypothesis most favourable to this view, namelythat at the outset entrepreneurs expect the reduction in money-wages to have this effect. It is indeed not unlikely that theindividual entrepreneur, seeing his own costs reduced, willoverlook at the outset the repercussions on the demand forhis product and will act on the assumption that he will be

1In an appendix to this chapter Professor Pigou’s Theory of Unemploymentis criticised in detail.

254

II.

able to sell at a profit a larger output than before. If, then, en-trepreneurs generally act on this expectation, will they in factsucceed in increasing their profits? Only if the community’smarginal propensity to consume is equal to unity, so that thereis no gap between the increment of income and the incrementof consumption; or if there is an increase in investment, cor-responding to the gap between the increment of income andthe increment of consumption, which will only occur if theschedule of marginal efficiencies of capital has increased rela-tively to the rate of interest. Thus the proceeds realised fromthe increased output will disappoint the entrepreneurs andemployment will fall back again to its previous figure, un-less the marginal propensity to consume is equal to unity orthe reduction in money-wages has had the effect of increasingthe schedule of marginal efficiencies of capital relatively to therate of interest and hence the amount of investment. For if en-trepreneurs offer employment on a scale which, if they couldsell their output at the expected price, would provide the pub-lic with incomes out of which they would save more thanthe amount of current investment, entrepreneurs are boundto make a loss equal to the difference; and this will be the caseabsolutely irrespective of the level of money wages. At thebest, the date of their disappointment can only be delayed forthe interval during which their own investment in increasedworking capital is filling the gap.

Thus the reduction in money-wages will have no lastingtendency to increase employment except by virtue of its reper-cussions either on the propensity to consume for the commu-nity as a whole, or on the schedule of marginal efficienciesof capital, or on the rate of interest. There is no method ofanalysing the effect of a reduction in money-wages, except byfollowing up its possible effects on these three factors.

The most important repercussions on these factors are likely,in practice, to be the following:

(1) A reduction of money-wages will somewhat reduce prices.It will, therefore, involve some redistribution of real income(a) from wage-earners to other factors entering into marginal

255


prime cost whose remuneration has not been reduced, and (b)from entrepreneurs to rentiers to whom a certain income fixedin terms of money has been guaranteed.

What will be the effect of this redistribution on the propen-sity to consume for the community as a whole? The transferfrom wage-earners to other factors is likely to diminish thepropensity to consume. The effect of the transfer from en-trepreneurs to rentiers is more open to doubt. But if rentiersrepresent on the whole the richer section of the communityand those whose standard of life is least flexible, then the ef-fect of this also will be unfavourable. What the net result willbe on a balance of considerations, we can only guess. Proba-bly it is more likely to be adverse than favourable.

(2) If we are dealing with an unclosed system, and the re-duction of money-wages is a reduction relatively to money-wagesabroad when both are reduced to a common unit, it is evidentthat the change will be favourable to investment, since it willtend to increase the balance of trade. This assumes, of course,that the advantage is not offset by a change in tariffs, quotas,etc. The greater strength of the traditional belief in the effi-cacy of a reduction in money-wages as a means of increasingemployment in Great Britain, as compared with the UnitedStates, is probably attributable to the latter being, compara-tively with ourselves, a closed system.

(3) In the case of an unclosed system, a reduction of money-wages, though it increases the favourable balance of trade, islikely to worsen the terms of trade. Thus there will be a reduc-tion in real incomes, except in the case of the newly employed,which may tend to increase the propensity to consume.

(4) If the reduction of money-wages is expected to be a re-duction relatively to money-wages in the future, the change willbe favourable to investment, because as we have seen above,it will increase the marginal efficiency of capital; whilst forthe same reason it may be favourable to consumption. If,on the other hand, the reduction leads to the expectation, oreven to the serious possibility, of a further wage-reduction inprospect, it will have precisely the opposite effect. For it will

256

II.

diminish the marginal efficiency of capital and will lead to thepostponement both of investment and of consumption.

(5) The reduction in the wages-bill, accompanied by somereduction in prices and in money-incomes generally, will di-minish the need for cash for income and business purposes;and it will therefore reduce pro tanto the schedule of liquidity-preference for the community as a whole. Cet. par. this willreduce the rate of interest and thus prove favourable to invest-ment. In this case, however, the effect of expectation concern-ing the future will be of an opposite tendency to those justconsidered under (4). For, if wages and prices are expectedto rise again later on, the favourable reaction will be muchless pronounced in the case of long-term loans than in thatof short-term loans. If, moreover, the reduction in wages dis-turbs political confidence by causing popular discontent, theincrease in Liquidity preference due to this cause may morethan offset the release of cash from the active circulation.

(6) Since a special reduction of money-wages is always ad-vantageous to an individual entrepreneur or industry, a gen-eral reduction (though its actual effects are different) may alsoproduce an optimistic tone in the minds of entrepreneurs, whichmay break through a vicious circle of unduly pessimistic es-timates of the marginal efficiency of capital and set thingsmoving again on a more normal basis of expectation. On theother hand, if the workers make the same mistake as their em-ployers about the effects of a general reduction, labour trou-bles may offset this favourable factor; apart from which, sincethere is, as a rule, no means of securing a simultaneous andequal reduction of money-wages in all industries, it is in theinterest of all workers to resist a reduction in their own partic-ular case. In fact, a movement by employers to revise money-wage bargains downward will be much more strongly resistedthan a gradual and automatic lowering of real wages as a re-sult of rising prices.

(7) On the other hand, the depressing influence on entrepreneursof their greater burden of debt may partly offset any cheer-ful reactions from the reduction of wages. Indeed if the fall

257


of wages and prices goes far, the embarrassment of those en-trepreneurs who are heavily indebted may soon reach the pointof insolvency, — with severely adverse effects on investment.Moreover the effect of the lower price-level on the real burdenof the National Debt and hence on taxation is likely to provevery adverse to business confidence.

This is not a complete catalogue of all the possible reactionsof wage reductions in the complex real world. But the abovecover, I think, those which are usually the most important.

If, therefore, we restrict our argument to the case of a closedsystem, and assume that there is nothing to be hoped, but ifanything the contrary, from the repercussions of the new dis-tribution of real incomes on the community’s propensity tospend, it follows that we must base any hopes of favourableresults to employment from a reduction in money-wages mainlyon an improvement in investment due either to an increasedmarginal efficiency of capital under (4) or a decreased rate ofinterest under (5). Let us consider these two possibilities infurther detail.

The contingency, which is favourable to an increase in themarginal efficiency of capital, is that in which money-wagesare believed to have touched bottom, so that further changesare expected to be in the upward direction. The most un-favourable contingency is that in which money-wages are slowlysagging downwards and each reduction in wages serves to di-minish confidence in the prospective maintenance of wages.When we enter on a period of weakening effective demand,a sudden large reduction of money-wages to a level so lowthat no one believes in its indefinite continuance would be theevent most favourable to a strengthening of effective demand.But this could only be accomplished by administrative decreeand is scarcely practical politics under a system of free wage-bargaining. On the other hand, it would be much better thatwages should be rigidly fixed and deemed incapable of ma-terial changes, than that depressions should be accompaniedby a gradual downward tendency of money-wages, a furthermoderate wage reduction being expected to signalise each in-

258

II.

crease of, say, 1 per cent. in the amount of unemployment.For example, the effect of an expectation that wages are goingto sag by, say, 2 per cent. in the coming year will be roughlyequivalent to the effect of a rise of 2 per cent. in the amount ofinterest payable for the same period. The same observationsapply mutatis mutandis to the case of a boom.

It follows that with the actual practices and institutions ofthe contemporary world it is more expedient to aim at a rigidmoney-wage policy than at a flexible policy responding byeasy stages to changes in the amount of unemployment; —so far, that is to say, as the marginal efficiency of capital isconcerned. But is this conclusion upset when we turn to therate of interest?

It is, therefore, on the effect of a falling wage- and price-level on the demand for money that those who believe in theself-adjusting quality of the economic system must rest theweight of their argument; though I am not aware that theyhave done so. If the quantity of money is itself a function ofthe wage- and price-level, there is indeed, nothing to hope inthis direction. But if the quantity of money is virtually fixed,it is evident that its quantity in terms of wage-units can be in-definitely increased by a sufficient reduction in money-wages;and that its quantity in proportion to incomes generally can belargely increased, the limit to this increase depending on theproportion of wage-cost to marginal prime cost and on the re-sponse of other elements of marginal prime cost to the fallingwage-unit.

We can, therefore, theoretically at least, produce preciselythe same effects on the rate of interest by reducing wages,whilst leaving the quantity of money unchanged, that we canproduce by increasing the quantity of money whilst leavingthe level of wages unchanged. It follows that wage reduc-tions, as a method of securing full employment, are also sub-ject to the same limitations as the method of increasing thequantity of money. The same reasons as those mentionedabove, which limit the efficacy of increases in the quantity ofmoney as a means of increasing investment to the optimumfigure, apply mutatis mutandis to wage reductions. Just as a

259


moderate increase in the quantity of money may exert an in-adequate influence over the long-term rate of interest, whilstan immoderate increase may offset its other advantages by itsdisturbing effect on confidence; so a moderate reduction inmoney-wages may prove inadequate, whilst an immoderatereduction might shatter confidence even if it were practicable.

There is, therefore, no ground for the belief that a flexi-ble wage policy is capable of maintaining a state of contin-uous full employment; — any more than for the belief than anopen-market monetary policy is capable, unaided, of achiev-ing this result. The economic system cannot be made self-adjusting along these lines.

If, indeed, labour were always in a position to take action(and were to do so), whenever there was less than full em-ployment, to reduce its money demands by concerted actionto whatever point was required to make money so abundantrelatively to the wage-unit that the rate of interest would fallto a level compatible with full employment, we should, in ef-fect, have monetary management by the Trade Unions, aimedat full employment, instead of by the banking system.

Nevertheless while a flexible wage policy and a flexiblemoney policy come, analytically, to the same thing, inasmuchas they are alternative means of changing the quantity of moneyin terms of wage-units, in other respects there is, of course, aworld of difference between them. Let me briefly recall to thereader’s mind the three outstanding considerations.

(i) Except in a socialised community where wage-policy issettled by decree, there is no means of securing uniform wagereductions for every class of labour. The result can only bebrought about by a series of gradual, irregular changes, jus-tifiable on no criterion of social justice or economic expedi-ency, and probably completed only after wasteful and disas-trous struggles, where those in the weakest bargaining posi-tion will suffer relatively to the rest. A change in the quantityof money, on the other hand, is already within the power ofmost governments by open-market policy or analogous mea-sures. Having regard to human nature and our institutions, it

260

II.

can only be a foolish person who would prefer a flexible wagepolicy to a flexible money policy, unless he can point to ad-vantages from the former which are not obtainable from thelatter. Moreover, other things being equal, a method which itis comparatively easy to apply should be deemed preferableto a method which is probably so difficult as to be impractica-ble.

(ii) If money-wages are inflexible, such changes in pricesas occur (i.e. apart from “administered” or monopoly priceswhich are determined by other considerations besides marginalcost) will mainly correspond to the diminishing marginal pro-ductivity of the existing equipment as the output from it isincreased. Thus the greatest practicable fairness will be main-tained between labour and the factors whose remuneration iscontractually fixed in terms of money, in particular the ren-tier class and persons with fixed salaries on the permanentestablishment of a firm, an institution or the State. If impor-tant classes are to have their remuneration fixed in terms ofmoney in any case, social justice and social expediency arebest served if the remunerations of all factors are somewhatinflexible in terms of money. Having regard to the large groupsof incomes which are comparatively inflexible in terms of money,it can only be an unjust person who would prefer a flexiblewage policy to a flexible money policy, unless he can point toadvantages from the former which are not obtainable from thelatter.

(iii) The method of increasing the quantity of money in termsof wage-units by decreasing the wage-unit increases propor-tionately the burden of debt; whereas the method of produc-ing the same result by increasing the quantity of money whilstleaving the wage unit unchanged has the opposite effect. Hav-ing regard to the excessive burden of many types of debt, itcan only be an inexperienced person who would prefer theformer.

(iv) If a sagging rate of interest has to be brought about bya sagging wage-level, there is, for the reasons given above, adouble drag on the marginal efficiency of capital and a double

261


reason for putting off investment and thus postponing recov-ery.

III

It follows, therefore, that if labour were to respond to con-ditions of gradually diminishing employment by offering itsservices at a gradually diminishing money-wage, this wouldnot, as a rule, have the effect of reducing real wages and mighteven have the effect of increasing them, through its adverse in-fluence on the volume of output. The chief result of this policywould be to cause a great instability of prices, so violent per-haps as to make business calculations futile in an economicsociety functioning after the manner of that in which we live.To suppose that a flexible wage policy is a right and proper ad-junct of a system which on the whole is one of laissez-faire, isthe opposite of the truth. It is only in a highly authoritarian so-ciety, where sudden, substantial, all-round changes could bedecreed that a flexible wage-policy could function with suc-cess. One can imagine it in operation in Italy, Germany orRussia, but not in France, the United States or Great Britain.

If, as in Australia, an attempt were made to fix real wagesby legislation., then there would be a certain level of employ-ment corresponding to that level of real wages; and the actuallevel of employment would, in a closed system, oscillate vi-olently between that level and no employment at all, accord-ing as the rate of investment was or was not below the ratecompatible with that level; whilst prices would be in unsta-ble equilibrium when investment was at the critical level, rac-ing to zero whenever investment was below it, and to infin-ity whenever it was above it. The element of stability wouldhave to be found, if at all, in the factors controlling the quan-tity of money being so determined that there always existedsome level of money-wages at which the quantity of moneywould be such as to establish a relation between the rate ofinterest and the marginal efficiency of capital which wouldmaintain investment at the critical level. In this event employ-

262

III.

ment would be constant (at the level appropriate to the legalreal wage) with money-wages and prices fluctuating rapidlyin the degree just necessary to maintain this rate of investmentat the appropriate figure. In the actual case of Australia, theescape was found, partly of course in the inevitable inefficacyof the legislation to achieve its object, and partly in Australianot being a closed system, so that the level of money-wageswas itself a determinant of the level of foreign investment andhence of total investment, whilst the terms of trade were animportant influence on real wages.

In the light of these considerations I am now of the opin-ion that the maintenance of a stable general level of money-wages is, on a balance of considerations, the most advisablepolicy for a closed system; whilst the same conclusion willhold good for an open system, provided that equilibrium withthe rest of the world can be secured by means of fluctuatingexchanges. There are advantages in some degree of flexibilityin the wages of particular industries so as to expedite transfersfrom those which are relatively declining to those which arerelatively expanding. But the money-wage level as a wholeshould be maintained as stable as possible, at any rate in theshort period.

This policy will result in a fair degree of stability in theprice-level;-greater stability, at least, than with a flexible wagepolicy. Apart from “administered” or monopoly prices, theprice-level will only change in the short period in response tothe extent that changes in the volume of employment affectmarginal prime costs; whilst in the long period they will onlychange in response to changes in the cost of production dueto new technique and new or increased equipment.

It is true that, if there are, nevertheless, large fluctuationsin employment, substantial fluctuations in the price-level willaccompany them. But the fluctuations will be less, as I havesaid above, than with a flexible wage policy.

Thus with a rigid wage policy the stability of prices will bebound up in the short period with the avoidance of fluctua-tions in employment. In the long period, on the other hand,we are still left with the choice between a policy of allowing

263


prices to fall slowly with the progress of technique and equip-ment whilst keeping wages stable, or of allowing wages to riseslowly whilst keeping prices stable. On the whole my prefer-ence is for the latter alternative, on account of the fact that itis easier with an expectation of higher wages in future to keepthe actual level of employment within a given range of fullemployment than with an expectation of lower wages in fu-ture, and on account also of the social advantages of graduallydiminishing the burden of debt, the greater ease of adjustmentfrom decaying to growing industries, and the psychologicalencouragement likely to be felt from a moderate tendency formoney-wages to increase. But no essential point of principle isinvolved, an it would lead me beyond the scope of my presentpurpose to develop in detail the arguments on either side.

264

19.A. Professor Pigou’s “Theory of Unemployment”

19.A Professor Pigou’s “Theory ofUnemployment”

Professor Pigou in his Theory of Unemployment makes the vol-ume of employment to depend on two fundamental factors,namely (1) the real rates of wages for which workpeople stipu-late, and (2) the shape of the Real Demand Function for Labour.The central sections of his book are concerned with determin-ing the shape of the latter function. The fact that workpeoplein fact stipulate, not for a real rate of wages, but for a money-rate, is not ignored; but, in effect, it is assumed that the actualmoney-rate of wages divided by the price of wage-goods canbe taken to measure the real rate demanded.

The equations which, as he says, “form the starting pointof the enquiry” into the Real Demand Function for Labour aregiven in his Theory of Unemployment, p. 90. Since the tacit as-sumptions, which govern the application of his analysis, slipin near the outset of his argument, I will summarise his treat-ment up to the crucial point.

Professor Pigou divides industries into those “engaged inmaking wage-goods at home and in making exports the saleof which creates claims to wage-goods abroad” and the “other”industries: which it is convenient to call the wage-goods in-dustries and the non-wage-goods industries respectively. Hesupposes x men to be employed in the former and y men inthe latter. The output in value of wage-goods of the x men hecalls F (x); and the general rate of wages F ′(x). This, thoughhe does not stop to mention it, is tantamount to assuming thatmarginal wage-cost is equal to marginal prime cost.[1] Fur-ther, he assumes that x + y = f(x), i.e. that the number ofmen employed in the wage-goods industries is a function oftotal employment. He then shows that the elasticity of thereal demand for labour in the aggregate (which gives us theshape of our quaesitum, namely the Real Demand Functionfor Labour) can be written

Er =φ′(x)

φ(x)× F ′(x)

F ′′(x)

265


So far as notation goes, there is no significant difference be-tween this and my own modes of expression. In so far as wecan identify Professor Pigou’s wage-goods with my consumption-goods, and his “other goods” with my investment-goods, itfollows that his F (x)/F ′(x), being the value of the output ofthe wage-goods industries in terms of the wage-unit, is thesame as my Cw. Furthermore, his function f is (subject tothe identification of wage-goods with consumption-goods) afunction of what I have called above the employment multi-plier k′. For

∆x = k′∆y ,

so thatφ′(x) = 1 +

1

k′ .

Thus Professor Pigou’s “elasticity of the real demand forlabour in the aggregate” is a concoction similar to some of myown, depending partly on the physical and technical condi-tions in industry (as given by his function F ) and partly onthe propensity to consume wage-goods (as given by his func-tion φ); provided always that we are limiting ourselves to thespecial case where marginal labour-cost is equal to marginalprime cost.

To determine the quantity of employment, Professor Pigouthen combines with his “real demand for labour”, a supplyfunction for labour. He assumes that this is a function of thereal wage and of nothing else. But, as he has also assumedthat the real wage is a function of the number of men x whoare employed in the wage-goods industries, this amounts toassuming that the total supply of labour at the existing realwage is a function of x and of nothing else. That is to say,n = χ(x), where n is the supply of labour available at a realwage F ′(x).

Thus, cleared of all complication, Professor Pigou’s anal-ysis amounts to an attempt to discover the volume of actualemployment from the equations

x+ y = f(x)

266


andn = χ(x) .

But there are here three unknowns and only two equa-tions. It seems clear that he gets round this difficulty by takingn = x + y. This amounts, of course, to assuming that there isno involuntary unemployment in the strict sense, i.e. that alllabour available at the existing real wage is in fact employed.In this case x has the value which satisfies the equation

f(x) = χ(x)

and when we have thus found that the value of x is equal to(say) n1, y must be equal to χ(n1)−n1, and total employmentn is equal to χ(n1).

It is worth pausing for a moment to consider what this in-volves. It means that, if the supply function of labour changes,more labour being available at a given real wage (so that n1 +∂n, is now the value of x which satisfies the equation φ(x) =χ(x)), the demand for the output of the non-wage-goods in-dustries is such that employment in these industries is boundto increase by just the amount which will preserve equalitybetween φ(n1 + ∂n1) and χ(n1 + ∂n1). The only other wayin which it is possible for aggregate employment to change isthrough a modification of the propensity to purchase wage-goods and non-wage-goods respectively such that there is anincrease of y accompanied by a greater decrease of x.

The assumption that n = x + y means, of course, thatlabour is always in a position to determine its own real wage.Thus, the assumption that labour is in a position to determineits own real wage, means that the demand for the output ofthe non-wage-goods industries obeys the above laws. In otherwords, it is assumed that the rate of interest always adjusts it-self to the schedule of the marginal efficiency of capital in sucha way as to preserve full employment. Without this assump-tion Professor Pigou’s analysis breaks down and provides nomeans of determining what the volume of employment willbe. It is, indeed, strange that Professor Pigou should havesupposed that he could furnish a theory of unemployment

267


which involves no reference at all to changes in the rate ofinvestment (i.e. to changes in employment in the non-wage-goods industries) due, not to a change in the supply functionof labour, but to changes in (e.g.) either the rate of interest orthe state of confidence.

His title the “Theory of Unemployment” is, therefore, some-thing of a misnomer. His book is not really concerned withthis subject. It is a discussion of how much employment therewill be, given the supply function of labour, when the con-ditions for full employment are satisfied. The purpose of theconcept of the elasticity of the real demand for labour in theaggregate is to show by how much full employment will riseor fall corresponding to a given shift in the supply functionof labour. Or — alternatively and perhaps better — we mayregard his book as a non-causative investigation into the func-tional relationship which determines what level of real wageswill correspond to any given level of employment. But it isnot capable of telling us what determines the actual level ofemployment; and on the problem of involuntary unemploy-ment it has no direct bearing.

If Professor Pigou were to deny the possibility of involun-tary unemployment in the sense in which I have defined itabove, as, perhaps, he would, it is still difficult to see how hisanalysis could be applied. For his omission to discuss whatdetermines the connection between x and y, i.e. between em-ployment in the wage-goods and non-wage-goods industriesrespectively, still remains fatal.

Moreover, he agrees that within certain limits labour in factoften stipulates, not for a given real wage, but for a givenmoney-wage. But in this case the supply function of labouris not a function of F ′(x) alone but also of the money-priceof wage-goods; — with the result that the previous analysisbreaks down and an additional factor has to be introduced,without there being an additional equation to provide for thisadditional unknown. The pitfalls of a pseudo-mathematicalmethod, which can make no progress except by making ev-erything a function of a single variable and assuming that allthe partial differentials vanish, could not be better illustrated.

268


For it is no good to admit later on that there are in fact othervariables, and yet to proceed without re-writing everythingthat has been written up to that point. Thus if (within limits)it is a money-wage for which labour stipulates, we still haveinsufficient data, even if we assume that n = x + y, unlesswe know what determines the money-price of wage-goods.For, the money-price of wage-goods depend on the aggre-gate amount of employment. Therefore, we cannot say whataggregate employment will be, until we know the money-price of wage-goods; and we cannot know the money-price ofwage-goods until we know the aggregate amount of employ-ment. We are, as I have said, one equation short. Yet it mightbe a provisional assumption of a rigidity of money-wages,rather than of real wages, which would bring our theory near-est to the facts. For example, money-wages in Great Britainduring the turmoil and uncertainty and wide price fluctua-tions of the decade 1924-1934 were stable within a range of 6per cent., whereas real wages fluctuated by more than 20 percent. A theory cannot claim to be a general theory, unless itis applicable to the case where (or the range within which)money-wages are fixed, just as much as to any other case.Politicians are entitled to complain that money-wages oughtto be highly flexible; but a theorist must be prepared to dealindifferently with either state of affairs. A scientific theorycannot require the facts to conform to its own assumptions.

When Professor Pigou comes to deal expressly with the ef-fect of a reduction of money-wages, he again, palpably (to mymind), introduces too few data to permit of any definite an-swer being obtainable. He begins by rejecting the argument(op. cit. p. 101) that, if marginal prime cost is equal to marginalwage-cost, non-wage-earners’ incomes will be altered, whenmoney-wages are reduced, in the same proportion as wage-earners’, on the ground that this is only valid, if the quantityof employment remains unaltered — which is the very pointunder discussion. But he proceeds on the next page (op. cit.p. 102) to make the same mistake himself by taking as hisassumption that “at the outset nothing has happened to non-wage-earners’ money-income”, which, as he has just shown,

269


is only valid if the quantity of employment does not remainunaltered — which is the very point under discussion. In fact,no answer is possible, unless other factors are included in ourdata.

The manner in which the admission, that labour in factstipulates for a given money-wage and not for a given realwage (provided that the real wage does not fall below a cer-tain minimum), affects the analysis, can also be shown bypointing out that in this case the assumption that more labouris not available except at a greater real wage, which is funda-mental to most of the argument, breaks down. For example,Professor Pigou rejects (Op. cit. p. 75) the theory of the multi-plier by assuming that the rate of real wages is given, i.e. that,there being already full employment, no additional labour isforthcoming at a lower real wage. Subject to this assumption,the argument is, of course, correct. But in this passage Profes-sor Pigou is criticising a proposal relating to practical policy;and it is fantastically far removed from the facts to assume,at a time when statistical unemployment in Great Britain ex-ceeded 2,000,000 (i.e. when there were 2,000,000 men willingto work at the existing money-wage), that any rise in the costof living, however moderate, relatively to the money-wagewould cause the withdrawal from the labour market of morethan the equivalent of all these 2,000,000 men.

It is important to emphasise that the whole of ProfessorPigou’s book is written on the assumption that any rise in thecost of living, however moderate, relatively to the money-wage willcause the withdrawal from the labour market of a number of workersgreater than that of all the existing unemployed.

Moreover, Professor Pigou does not notice in this passage(Op. cit. p. 75) that the argument, which he advances against“secondary” employment as a result of public works, is, onthe same assumptions, equally fatal to increased “primary”employment from the same policy. For if the real rate of wagesruling in the wage-goods industries is given, no increased em-ployment whatever is possible except, indeed, as a result ofnon-wage-earners reducing their consumption of wage-goods.For those newly engaged in the primary employment will pre-

270


sumably increase their consumption of wage-goods which willreduce the real wage and hence (on his assumptions) leadto a withdrawal of labour previously employed elsewhere.Yet Professor Pigou accepts, apparently, the possibility of in-creased primary employment. The line between primary andsecondary employment seems to be the critical psychologicalpoint at which his good common sense ceases to overbear hisbad theory.

The difference in the conclusions to which the above dif-ferences in assumptions and in analysis lead can be shownby the following important passage in which Professor Pigousums up his point of view: “With perfectly free competitionamong workpeople and labour perfectly mobile, the nature ofthe relation (i.e. between the real wage-rates for which peo-ple stipulate and the demand function for labour) will be verysimple. There will always be at work a strong tendency forwage-rates to be so related to demand that everybody is em-ployed. Hence, in stable conditions everyone will actually beemployed. The implication is that such unemployment as ex-ists at any time is due wholly to the fact that changes in de-mand conditions are continually taking place and that fric-tional resistances prevent the appropriate wage adjustmentsfrom being made instantaneously.”2

He concludes (op. cit. p. 253) that unemployment is pri-marily due to a wage policy which fails to adjust itself suf-ficiently to changes in the real Jemand function for labour.Thus Professor Pigou believes that in the long run unemploy-ment can be cured by wage adjustments;3 whereas I main-tain that the real wage (subject only to a minimum set bythe marginal disutility of employment) is not primarily deter-mined by “wage adjustments” (though these may have reper-cussions) but by the other forces of the system, some of which(in particular the relation between the schedule of the marginalefficiency of capital and the rate of interest) Professor Pigou

2Op. Cit. p. 252.3There is no hint or suggestion that this comes about through reactions on

the rate of interest.

271


has failed, if I am right, to include in his formal scheme.Finally, when Professor Pigou comes to the “Causation of

Unemployment” he speaks, it is true, of fluctuations in thestate of demand, much as I do. But he identifies the state ofdemand with the Real Demand Function for Labour, forget-ful of how narrow a thing the latter is on his definition. Forthe Real Demand Function for Labour depends by definition(as we have seen above) on nothing but two factors, namely(1) the relationship in any given environment between the to-tal number of men employed and the number who have to beemployed in the wage-goods industries to provide them withwhat they consume, and (2) the state of marginal productiv-ity in the wage-goods industries. Yet in Part V. of his Theory ofUnemployment fluctuations in the state of “the real demand forlabour” are given a position of importance. The “real demandfor labour” is regarded as a factor which is susceptible of wideshort-period fluctuations (op. cit. Part V. chaps. vi.-xii.), andthe suggestion seems to be that swings in “the real demandfor labour” are, in combination with the failure of wage pol-icy to respond sensitively to such changes, largely responsiblefor the trade cycle. To the reader all this seems, at first, rea-sonable and familiar. For, unless he goes back to the defini-tion, “fluctuations in the real demand for labour” will conveyto his mind the same sort of suggestion as I mean to conveyby “fluctuations in the state of aggregate demand”. But if wego back to the definition of the “real demand for labour”, allthis loses its plausibility. For we shall find that there is noth-ing in the world less likely to be subject to sharp short-periodswings than this factor.

Professor Pigou’s “real demand for labour” depends bydefinition on nothing but F (x), which represents the physicalconditions of production in the wage-goods industries, andφ(x), which represents the functional relationship betweenemployment in the wage-goods industries and total employ-ment corresponding to any given level of the latter. It is dif-ficult to see a reason why either of these functions shouldchange, except gradually over a long period. Certainly thereseems no reason to suppose that they are likely to fluctuate

272


during a trade cycle. For F (x) can only change slowly, and,in a technically progressive community, only in the forwarddirection; whilst φ(x) will remain stable, unless we suppose asudden outbreak of thrift in the working classes, or, more gen-erally, a sudden shift in the propensity to consume. I shouldexpect, therefore, that the real demand for labour would re-main virtually constant throughout a trade cycle. I repeat thatProfessor Pigou has altogether omitted from his analysis theunstable factor, namely fluctuations in the scale of investment,which is most often at the bottom of the phenomenon of fluc-tuations in employment.

I have criticised at length Professor Pigou’s theory of un-employment not because he seems to me to be more open tocriticism than other economists of the classical school; but be-cause his is the only attempt with which I am acquainted towrite down the classical theory of unemployment, precisely.Thus it has been incumbent on me to raise my objections tothis theory in the most formidable presentment in which ithas been advanced.

273

CHAPTER 20

THE EMPLOYMENT FUNCTION 1

I

In Chapter 3 (p. 23) we have defined the aggregate supplyfunction Z = φ(N), which relates the employmentN with theaggregate supply price of the corresponding output. The em-ployment function only differs from the aggregate supply func-tion in that it is, in effect, its inverse function and is definedin terms of the wage-unit; the object of the employment func-tion being to relate the amount of the effective demand, mea-sured in terms of the wage-unit, directed to a given firm orindustry or to industry as a whole with the amount of employ-ment, the supply price of the output of which will compare tothat amount of effective demand. Thus if an amount of effec-tive demand Dwr , measured in wage-units, directed to a firmor industry calls forth an amount of employment Nr in thatfirm or industry, the employment function is given by Nr =Fr(Dwr). Or, more Generally, if we are entitled to assume thatDwr is a unique function of the total effective demand Dw,the employment function is given by Nr = Fr(Dw). That isto say, Nr men will be employed in industry r when effectivedemand is Dw.

We shall develop in this chapter certain properties of theemployment function. But apart from any interest which thesemay have, there are two reasons why the substitution of theemployment function for the ordinary supply curve is conso-nant with the methods and objects of this book. In the firstplace, it expresses the relevant facts in terms of the units towhich we have decided to restrict ourselves, without intro-

1Those who (rightly) dislike algebra will lose little by omitting the first sec-tion of this chapter.

275

20. THE EMPLOYMENT FUNCTION

ducing any of the units which have a dubious quantitativecharacter. In the second place, it lends itself to the problems ofindustry and output as a whole, as distinct from the problemsof a single industry or firm in a given environment, more eas-ily than does the ordinary supply curve — for the followingreasons.

The ordinary demand curve for a particular commodity isdrawn on some assumption as to the incomes of members ofthe public, and has to be re-drawn if the incomes change. Inthe same way the ordinary supply curve for a particular com-modity is drawn on some assumption as to the output of in-dustry as a whole and is liable to change if the aggregate out-put of industry is changed. When, therefore, we are exam-ining the response of individual industries to changes in ag-gregate employment, we are necessarily concerned, not witha single demand curve for each industry, in conjunction witha single supply curve, but with two families of such curvescorresponding to different assumptions as to the aggregate em-ployment. In the case of the employment function, however,the task of arriving at a function for industry as a whole whichwill reflect changes in employment as a whole is more practi-cable.

For let us assume (to begin with) that the propensity toconsume is given as well as the other factors which we havetaken as given in Chapter 18 above, and that we are consider-ing changes in employment in response to changes in the rateof investment. Subject to this assumption, for every level ofeffective demand in terms of wage-units there will be a cor-responding aggregate employment and this effective demandwill be divided in determinate proportions between consump-tion and investment. Moreover, each level of effective de-mand will correspond to a given distribution of income. It isreasonable, therefore, further to assume that corresponding toa given level of aggregate effective demand there is a uniquedistribution of it between different industries.

This enables us to determine what amount of employmentin each industry will correspond to a given level of aggregateemployment. That is to say, it gives us the amount of employ-

276

I.

ment in each particular industry corresponding to each levelof aggregate effective demand measured in terms of wage-units, so that the conditions are satisfied for the second formof the employment function for the industry, defined above,namelyNr = Fr(Dw). Thus we have the advantage that, inthese conditions, the individual employment functions are ad-ditive in the sense that the employment function for indus-try as a whole, corresponding to a given level of effective de-mand, is equal to the sum of the employment functions foreach separate industry; i.e.

F (Dw) = N = ΣNr = ΣFr(Dw) .

Next, let us define the elasticity of employment. The elas-ticity of employment for a given industry is

eer =∂Nr

∂Dwr× Dwr

Nr,

since it measures the response of the number of labour-unitsemployed in the industry to changes in the number of wage-units which are expected to be spent on purchasing its output.The elasticity of employment for industry as a whole we shallwrite

ee =∂N

∂Dw× ∂Dw

N.

Provided that we can find some sufficiently satisfactorymethod of measuring output, it is also useful to define whatmay be called the elasticity of output or production, whichmeasures the rate at which output in any industry increaseswhen more effective demand in terms of wage-units is di-rected towards it, namely

eor =∂Or

∂Dwr× ∂Dwr

Or.

Provided we can assume that the price is equal to the marginalprime cost, we then have

∆Dwr =1

1− eor∆Pr

277


where Pr is the expected profit.2 It follows from this that ifeor = 0, i.e. if the output of the industry is perfectly inelas-tic, the whole of the increased effective demand (in terms ofwage-units) is expected to accrue to the entrepreneur as profit,i.e. ∆Dwr = ∆Pr ; whilst if eor = 1, i.e. if the elasticity ofoutput is unity, no part of the increased effective demand isexpected to accrue as profit, the whole of it being absorbed bythe elements entering into marginal prime cost.

Moreover, if the output of an industry is a function φ(Nr)

2For, if pwr is the expected price of a unit of output in terms of the wage-unit,

∆Dwr = D(pwrOr) = pwr∆Or +Or∆pwr

=Dwr

Or

×∆Or +Or∆pwr ,

so that

Or∆pwr = ∆Dwr(1− eor)

or

∆Dwr =Or∆pwr

1− eor

.

But

Or∆pwr = ∆Dwr − pwr∆Or

= ∆Dwr − (marginal prime cost)∆Or

= ∆P .

Hence

∆Dwr =1

1− eor

∆Pr .

278

I.

of the labour employed in it, we have3

1− eor

eer=−Nrφ

′′(Nr)

pwr{φ′(Nr)}2 ,

where pw is the expected price of a unit of output in termsof the wage-unit. Thus the condition eor = 1 means thatφ′′(Nr) = 0, i.e. that there are constant returns in responseto increased employment.

Now, in so far as the classical theory assumes that realwages are always equal to the marginal disutility of labourand that, the latter increases when employment increases, sothat the labour supply will fall off, cet. par., if real wagesare reduced, it is assuming that in practice it is impossibleto increase expenditure in terms of wage-units. If this weretrue, the concept of elasticity of employment would have nofield of application. Moreover, it would, in this event, be im-possible to increase employment by increasing expenditure interms of money; for money-wages would rise proportionatelyto the increased money expenditures so that there would beno increase of expenditure in terms of wage-units and con-sequently no increase in employment. But if the classical as-sumption does not hold good, it will be possible to increaseemployment by increasing expenditure in terms of money un-til real wages have fallen to equality with the marginal disu-tility of labour, at which point there will, by definition, be fullemployment.

Ordinarily, of course, eor will have a value intermediate be-tween zero and unity. The extent to which prices (in terms ofwage-units) will rise, i.e. the extent to which real wages will

3 For, sinceDwr = pwrOr , we have

1 = pwr∂Or

∂Dwr

+Or∂pwr

∂Dwr

= eor −Nrφ

′′(Nr)

{φ′(Nr)}2×eor

pwr

.

279


fall, when money expenditure is increased, depends, there-fore, on the elasticity of output in response to expenditure interms of wage-units.

Let the elasticity of the expected price pwr in response tochanges in effective demand Dwr , namely ∂pwr

∂Dwr× Dwr

pwr, be

written e′pr .

Since Or × pwr = Dwr , we have

∂Or

∂Dwr× Dwr

Or+∂pwr

∂Dwr× Dwr

pwr= 1

or

e′pr + eor = 1 .

That is to say, the sum of the elasticities of price and ofoutput in response to changes in effective demand (measuredin terms of wage-units) is equal to unity. Effective demandspends itself, partly in affecting output and partly in affectingprice, according to this law.

If we are dealing with industry as a whole and are pre-pared to assume that we have a unit in which output as awhole can be measured, the same line of argument applies,so that e′

p + eo = 1, where the elasticities without a suffix rapply to industry as a whole.

Let us now measure values in money instead of wage-unitsand extend to this case our conclusions in respect of industryas a whole.

If W stands for the money-wages of a unit of labour and pfor the expected price of a unit of output as a whole in termsof money, we can write ep (= Ddp/pdD) for the elasticityof money-prices in response to changes in effective demandmeasured in terms of money, and ew (= DdW/WdD) for theelasticity of money-wages in response to changes in effectivedemand in terms of money. It is then easily shown that

ep = 1− eo(1− ew) .

280

II.

4

This equation is, as we shall see in the next chapter, a firststep to a generalised Quantity Theory of Money. If eo = 0or if ew = 1, output will be unaltered and prices will rise inthe same proportion as effective demand in terms of money.Otherwise they will rise in a smaller proportion.

II

Let us return to the employment function. We have assumedin the foregoing that to every level of aggregate effective de-mand there corresponds a unique distribution of effective de-mand between the products of each individual industry. Now,as aggregate expenditure changes, the corresponding expen-diture on the products of an individual industry will not, ingeneral, change in the same proportion; — partly because in-dividuals will not, as their incomes rise, increase the amountof the products of each separate industry, which they pur-chase, in the same proportion, and partly because the prices

4 For, since p = pw ×W andD = Dw ×W , we have

∆p = W∆pw +p

W∆W

= W × e′ppw

Dw

∆Dw +p

W∆W

= e′p

p

D∆D −

D

W∆W +

p

W∆W

= e′p

p

D∆D + ∆W

p

W(1− e′p) ,

so that

ep =D∆p

p∆D= e′p +

D

p∆D×W∆p

W)(1− e′p)

= e′p + ew(1− e′p)

= 1− eo(1− ew) .

281


of different commodities will respond in different degrees toincreases in expenditure upon them.

It follows from this that the assumption upon which wehave worked hitherto, that changes in employment dependsolely on changes in aggregate effective demand (in terms ofwage-units), is no better than a first approximation, if we ad-mit that there is more than one way in which an increase ofincome can be spent. For the way in which we suppose theincrease in aggregate demand to be distributed between dif-ferent commodities may considerably influence the volume ofemployment. If, for example, the increased demand is largelydirected towards products which have a high elasticity of em-ployment, the aggregate increase in employment will be greaterthan if it is largely directed towards products which have alow elasticity of employment.

In the same way employment may fall off without therehaving been any change in aggregate demand, if the directionof demand is change in favour of products having a relativelylow elasticity of employment.

These considerations are particularly important if we areconcerned with short-period phenomena in the sense of changesin the amount or direction of demand which are not foreseensome time ahead. Some products take time to produce, sothat it is practically impossible to increase the supply of themquickly. Thus, if additional demand is directed to them with-out notice, they will show a low elasticity of employment; al-though it may be that, given sufficient notice, their elasticityof employment approaches unity.

It is in this connection that I find the principal significanceof the conception of a period of production. A product, Ishould prefer to say5 has a period of production n if n time-units of notice of changes in the demand for it have to begiven if it is to offer its maximum elasticity of employment.Obviously consumption-goods, taken as a whole, have in thissense the longest period of production, since of every produc-

5This is not identical with the usual definition, but it seems to me to em-body what is significant in the idea.

282

II.

tive process they constitute the last stage. Thus if the first im-pulse towards the increase in effective demand comes from anincrease in consumption, the initial elasticity of employmentwill be further below its eventual equilibrium-level than if theimpulse comes from an increase in investment. Moreover, ifthe increased demand is directed to products with a relativelylow elasticity of employment, a larger proportion of it will goto swell the incomes of entrepreneurs and a smaller propor-tion to swell the incomes of wage-earners and other prime-cost factors; with the possible result that the repercussionsmay be somewhat less favourable to expenditure, owing tothe likelihood of entrepreneurs saving more of their incrementof income than wage-earners would. Nevertheless the distinc-tion between the two cases must not be over-stated, since alarge part of the reactions will be much the same in both.6

However long the notice given to entrepreneurs of a prospec-tive change in demand, it is not possible for the initial elastic-ity of employment, in response to a given increase of invest-ment, to be as great as its eventual equilibrium value, unlessthere are surplus stocks and surplus capacity at every stageof production. On the other hand, the depletion of the sur-plus stocks will have an offsetting effect on the amount bywhich investment increases. If we suppose that there are ini-tially some surpluses at every point, the initial elasticity ofemployment may approximate to unity; then after the stockshave been absorbed, but before an increased supply is comingforward at an adequate rate from the earlier stages of produc-tion, the elasticity will fall away; rising again towards unityas the new position of equilibrium is approached. This issubject, however, to some qualification in so far as there arerent factors which absorb more expenditure as employmentincreases, or if the rate of interest increases. For these reasonsperfect stability of prices is impossible in an economy subjectto change — unless, indeed, there is some peculiar mechanismwhich ensures temporary fluctuations of just the right degree

6Some further discussion of the above topic is to be found in my Treatiseon Money, Book IV.

283


in the propensity to consume. But price-instability arising inthis way does not lead to the kind of profit stimulus which isliable to bring into existence excess capacity. For the windfallgain will wholly accrue to those entrepreneurs who happen topossess products at a relatively advanced stage of production,and there is nothing which the entrepreneur, who does notpossess specialised resources of the right kind, can do to at-tract this gain to himself. Thus the inevitable price-instabilitydue to change cannot affect the actions of entrepreneurs, butmerely directs a defacto windfall of wealth into the laps of thelucky ones (mutatis mutandis when the supposed change is inthe other direction). This fact has, I think, been overlooked insome contemporary discussions of a practical policy aimed atstabilising prices. It is true that in a society liable to changesuch a policy cannot be perfectly successful. But it does notfollow that every small temporary departure from price sta-bility necessarily sets up a cumulative disequilibrium.

III

We have shown that when effective demand is deficient thereis under-employment of labour in the sense that there are menunemployed who would be willing to work at less than the ex-isting real wage. Consequently, as effective demand increases,employment increases, though at a real wage equal to or lessthan the existing one, until a point comes at which there isno surplus of labour available at the then existing real wage;i.e. no more men (or hours of labour) available unless money-wages rise (from this point onwards) faster than prices. Thenext problem is to consider what will happen if, when thispoint has been reached, expenditure still continues to increase.

Up to this point the decreasing return from applying morelabour to a given capital equipment has been offset by the ac-quiescence of labour in a diminishing real wage. But afterthis point a unit of labour would require the inducement ofthe equivalent of an increased quantity of product, whereasthe yield from applying a further unit would be a diminished

284

III.

quantity of product. The conditions of strict equilibrium re-quire, therefore, that wages and prices, and consequently prof-its also, should all rise in the same proportion as expendi-ture, the “real” position, including the volume of output andemployment, being left unchanged in all respects. We havereached, that is to say, a situation in which the crude quantitytheory of money (interpreting “velocity” to mean “income-velocity”) is fully satisfied; for output does not alter and pricesrise in exact proportion to MV.

Nevertheless there are certain practical qualifications to thisconclusion which must be borne in mind in applying it to anactual case:

(1) For a time at least, rising prices may delude entrepreneursinto increasing employment beyond the level which maximisestheir individual profits measured in terms of the product. Forthey are so accustomed to regard rising sale-proceeds in termsof money as a signal for expanding production, that they maycontinue to do so when this policy has in fact ceased to beto their best advantage; i.e. they may underestimate theirmarginal user cost in the new price environment.

(2) Since that part of his profit which the entrepreneur hasto hand on to the rentier is fixed in terms of money, risingprices, even though unaccompanied by any change in out-put, will re-distribute incomes to the advantage of the en-trepreneur and to the disadvantage of the rentier, which mayhave a reaction on the propensity to consume. This, how-ever, is not a process which will have only begun when fullemployment has been attained; — it will have been makingsteady progress all the time that the expenditure was increas-ing. If the rentier is less prone to spend than the entrepreneur,the gradual withdrawal of real income from the former willmean that full employment will be reached with a smaller in-crease in the quantity of money and a smaller reduction inthe rate of interest than will be the case if the opposite hy-pothesis holds. After full employment has been reached, afurther rise of prices will, if the first hypothesis continues tohold, mean that the rate of interest will have to rise somewhat

285


to prevent prices from rising indefinitely, and that the increasein the quantity of money will be less than in proportion to theincrease in expenditure; whilst if the second hypothesis holds,the opposite will be the case. It may be that, as the real in-come of the rentier is diminished, a point will come when, asa result of his growing relative impoverishment, there will bea changeover from the first hypothesis to the second, whichpoint may be reached either before or after full employmenthas been attained.

IV

There is, perhaps, something a little perplexing in the appar-ent asymmetry between Inflation and Deflation. For whilst adeflation of effective demand below the level required for fullemployment will diminish employment as well as prices, aninflation of it above this level will merely affect prices. Thisasymmetry is, however, merely a reflection of the fact that,whilst labour is always in a position to refuse to work on ascale involving a real wage which is less than the marginaldisutility of that amount of employment, it is not in a posi-tion to insist on being offered work on a scale involving a realwage which is not greater than the marginal disutility of thatamount of employment.

286

CHAPTER 21

THE THEORY OF PRICES

I

So long as economists are concerned with what is called theTheory of Value, they have been accustomed to teach thatprices are governed by the conditions of supply and demand;and, in particular, changes in marginal cost and the elastic-ity of short-period supply have played a prominent part. Butwhen they pass in volume II, or more often in a separate trea-tise, to the Theory of Money and Prices, we hear no more ofthese homely but intelligible concepts and move into a worldwhere prices are governed by the quantity of money, by itsincome-velocity, by the velocity of circulation relatively to thevolume of transactions, by hoarding, by forced saving, by in-flation and deflation et hoc genus omne; and little or no attemptis made to relate these vaguer phrases to our former notionsof the elasticities of supply and demand. If we reflect on whatwe are being taught and try to rationalise it, in the simplerdiscussions it seems that the elasticity of supply must have be-come zero and demand proportional to the quantity of money;whilst in the more sophisticated we are lost in a haze wherenothing is clear and everything is possible. We have all of usbecome used to finding ourselves sometimes on the one sideof the moon and sometimes on the other, without knowingwhat route or journey connects them, related, apparently, af-ter the fashion of our waking and our dreaming lives.

One of the objects of the foregoing chapters has been to es-cape from this double life and to bring the theory of prices asa whole back to close contact with the theory of value. The di-vision of Economics between the Theory of Value and Distri-bution on the one hand and the Theory of Money on the other

287

21. THE THEORY OF PRICES

hand is, I think, a false division. The right dichotomy is, I sug-gest, between the Theory of the Individual Industry or Firmand of the rewards and the distribution between different usesof a given quantity of resources on the one hand, and the The-ory of Output and Employment as a whole on the other hand.So long as we limit ourselves to the study of the individualindustry or firm on the assumption that the aggregate quan-tity of employed resources is constant, and, provisionally, thatthe conditions of other industries or firms are unchanged, it istrue that we are not concerned with the significant character-istics of money. But as soon as we pass to the problem of whatdetermines output and employment as a whole, we requirethe complete theory of a Monetary Economy.

Or, perhaps, we might make our line of division betweenthe theory of stationary equilibrium and the theory of shift-ing equilibrium-meaning by the latter the theory of a systemin which changing views about the future are capable of in-fluencing the present situation. For the importance of moneyessentially flows from its being a link between the present and thefuture. We can consider what distribution of resources be-tween different uses will be consistent with equilibrium un-der the influence of normal economic motives in a world inwhich our views concerning the future are fixed and reliablein all respects; — with a further division, perhaps, betweenan economy which is unchanging and one subject to change,but where all things are foreseen from the beginning. Or wecan pass from this simplified propaedeutic to the problems ofthe real world in which our previous expectations are liableto disappointment and expectations concerning the future af-fect what we do to-day. It is when we have made this transi-tion that the peculiar properties of money as a link betweenthe present and the future must enter into our calculations.But, although the theory of shifting equilibrium must neces-sarily be pursued in terms of a monetary economy, it remainsa theory of value and distribution and not a separate “theoryof money”. Money in its significant attributes is, above all, asubtle device for linking the present to the future; and we can-not even begin to discuss the effect of changing expectations

288

II.

on current activities except in monetary terms. We cannot getrid of money even by abolishing gold and silver and legal ten-der instruments. So long as there exists any durable asset, it iscapable of possessing monetary attributes1 and, therefore, ofgiving rise to the characteristic problems of a monetary econ-omy.

II

In a single industry its particular price-level depends partlyon the rate of remuneration of the factors of production whichenter into its marginal cost, and partly on the scale of output.There is no reason to modify this conclusion when we pass toindustry as a whole. The general price-level depends partlyon the rate of remuneration of the factors of production whichenter into marginal cost and partly on the scale of output as awhole, i.e. (taking equipment and technique as given) on thevolume of employment. It is true that, when we pass to out-put as a whole, the costs of production in any industry partlydepend on the output of other industries. But the more signif-icant change, of which we have to take account, is the effectof changes in demand both on costs and on volume. It is onthe side of demand that we have to introduce quite new ideaswhen we are dealing with demand as a whole and no longerwith the demand for a single product taken in isolation, withdemand as a whole assumed to be unchanged.

III

If we allow ourselves the simplification of assuming that therates of remuneration of the different factors of productionwhich enter into marginal cost all change in the same propor-tion, i.e. in the same proportion as the wage-unit, it followsthat the general price-level (taking equipment and technique

1Cf. Chapter 17 above.

289


as given) depends partly on the wage-unit and partly on thevolume of employment. Hence the effect of changes in thequantity of money on the price-level can be considered as be-ing compounded of the effect on the wage-unit and the effecton employment.

To elucidate the ideas involved, let us simplify our assump-tions still further, and assume (1) that all unemployed resourcesare homogeneous and interchangeable in their efficiency toproduce what is wanted, and (2) that the factors of productionentering into marginal cost are content with the same money-wage so long as there is a surplus of them unemployed. Inthis case we have constant returns and a rigid wage-unit, solong as there is any unemployment. It follows that an in-crease in the quantity of money will have no effect whateveron prices, so long as there is any unemployment, and that em-ployment will increase in exact proportion to any increase ineffective demand brought about by the increase in the quan-tity of money; whilst as soon as full employment is reached, itwill thenceforward be the wage-unit and prices which will in-crease in exact proportion to the increase in effective demand.Thus if there is perfectly elastic supply so long as there is un-employment, and perfectly inelastic supply so soon as fullemployment is reached, and if effective demand changes inthe same proportion as the quantity of money, the QuantityTheory of Money can be enunciated as follows: “So long asthere is unemployment, employment will change in the sameproportion as the quantity of money; and when there is fullemployment, prices will change in the same proportion as thequantity of money”.

Having, however, satisfied tradition by introducing a suffi-cient number of simplifying assumptions to enable us to enun-ciate a Quantity Theory of Money, let us now consider the pos-sible complications which will in fact influence events:

(1) Effective demand will not change in exact proportion tothe quantity of money.

(2) Since resources are not homogeneous, there will be dimin-

290

III.

ishing, and not constant, returns as employment gradu-ally increases.

(3) Since resources are not interchangeable, some commodi-ties will reach a condition of inelastic supply whilst thereare still unemployed resources available for the produc-tion of other commodities.

(4) The wage-unit will tend to rise, before full employmenthas been reached.

(5) The remunerations of the factors entering into marginalcost will not all change in the same proportion.

Thus we must first consider the effect of changes in thequantity of money on the quantity of effective demand; andthe increase in effective demand will, generally speaking, spenditself partly in increasing the quantity of employment and partlyin raising the level of prices. Thus instead of constant pricesin conditions of unemployment, and of prices rising in pro-portion to the quantity of money in conditions of full employ-ment, we have in fact a condition of prices rising graduallyas employment increases. The Theory of Prices, that is to say,the analysis of the relation between changes in the quantity ofmoney and changes in the price-level with a view to determin-ing the elasticity of prices in response to changes in the quan-tity of money, must, therefore, direct itself to the five compli-cating factors set forth above.

We will consider each of them in turn. But this proceduremust not be allowed to lead us into supposing that they are,strictly speaking, independent. For example, the proportion,in which an increase in effective demand is divided in its effectbetween increasing output and raising prices, may affect theway in which the quantity of money is related to the quantityof effective demand. Or, again, the differences in the propor-tions, in which the remunerations of different factors change,may influence the relation between the quantity of money andthe quantity of effective demand. The object of our analysis is,not to provide a machine, or method of blind manipulation,

291


which will furnish an infallible answer, but to provide our-selves with an organised and orderly method of thinking outparticular problems; and, after we have reached a provisionalconclusion by isolating the complicating factors one by one,we then have to go back on ourselves and allow, as well aswe can, for the probable interactions of the factors amongstthemselves. This is the nature of economic thinking. Anyother way of applying our formal principles of thought (with-out which, however, we shall be lost in the wood) will lead usinto error. It is a great fault of symbolic pseudo-mathematicalmethods of formalising a system of economic analysis, suchas we shall set down in section vi of this chapter, that theyexpressly assume strict independence between the factors in-volved and lose all their cogency and authority if this hypoth-esis is disallowed; whereas, in ordinary discourse, where weare not blindly manipulating but know all the time what weare doing and what the words mean, we can keep “at the backof our heads” the necessary reserves and qualifications andthe adjustments which we shall have to make later on, in away in which we cannot keep complicated partial differen-tials “at the back” of several pages of algebra which assumethat they all vanish. Too large a proportion of recent “mathe-matical” economics are mere concoctions, as imprecise as theinitial assumptions they rest on, which allow the author tolose sight of the complexities and interdependencies of thereal world in a maze of pretentious and unhelpful symbols.

IV

(1) The primary effect of a change in the quantity of moneyon the quantity of effective demand is through its influence onthe rate of interest. If this were the only reaction, the quanti-tative effect could be derived from the three elements — (a)the schedule of liquidity-preference which tells us by howmuch the rate of interest will have to fall in order that thenew money may be absorbed by willing holders, (b) the sched-ule of marginal efficiencies which tells us by how much a

292

IV.

given fall in the rate of interest will increase investment, and(c) the investment multiplier which tells us by how much agiven increase in investment will increase effective demandas a whole.

But this analysis, though it is valuable in introducing or-der and method into our enquiry, presents a deceptive sim-plicity, if we forget that the three elements (a), (b) and (c) arethemselves partly dependent on the complicating factors (2),(3), (4) and (5) which we have not yet considered. For theschedule of liquidity-preference itself depends on how muchof the new money is absorbed into the income and industrialcirculations, which depends in turn on how much effectivedemand increases and how the increase is divided betweenthe rise of prices, the rise of wages, and the volume of outputand employment. Furthermore, the schedule of marginal ef-ficiencies will partly depend on the effect which the circum-stances attendant on the increase in the quantity of moneyhave on expectations of the future monetary prospects. Andfinally the multiplier will be influenced by the way in whichthe new income resulting from the increased effective demandis distributed between different classes of consumers. Nor, ofcourse, is this list of possible interactions complete. Neverthe-less, if we have all the facts before us, we shall have enoughsimultaneous equations to give us a determinate result. Therewill be a determinate amount of increase in the quantity of ef-fective demand which, after taking everything into account,will correspond to, and be in equilibrium with, the increase inthe quantity of money. Moreover, it is only in highly excep-tional circumstances that an increase in the quantity of moneywill be associated with a decrease in the quantity of effectivedemand.

The ratio between the quantity of effective demand and thequantity of money closely corresponds to what is often calledthe “income-velocity of money”; — except that effective de-mand corresponds to the income the expectation of which hasset production moving, not to the actually realised income,and to gross, not net, income. But the “income-velocity ofmoney” is, in itself, merely a name which explains nothing.

293


There is no reason to expect that it will be constant. For itdepends, as the foregoing discussion has shown, on manycomplex and variable factors. The use of this term obscures, Ithink, the real character of the causation, and has led to noth-ing but confusion.

(2) As we have shown above (Chapter 4. s. III), the dis-tinction between diminishing and constant returns partly de-pends on whether workers are remunerated in strict propor-tion to their efficiency. If so, we shall have constant labour-costs (in terms of the wage-unit) when employment increases.But if the wage of a given grade of labourers is uniform ir-respective of the efficiency of the individuals, we shall haverising labour-costs, irrespective of the efficiency of the equip-ment. Moreover, if equipment is non-homogeneous and somepart of it involves a greater prime cost per unit of output, weshall have increasing marginal prime costs over and aboveany increase due to increasing labour-costs.

Hence, in general, supply price will increase as output.from a given equipment is increased. Thus increasing outputwill be associated with rising prices, apart from any change inthe wage-unit.

(3) Under (2) we have been contemplating the possibilityof supply being imperfectly elastic. If there is a perfect bal-ance in the respective quantities of specialised unemployedresources, the point of full employment will be reached for allof them simultaneously. But, in general, the demand for someservices and commodities will reach a level beyond whichtheir supply is, for the time being, perfectly inelastic, whilst inother directions there is still a substantial surplus of resourceswithout employment. Thus as output increases, a series of“bottlenecks” will be successively reached, where the supplyof particular commodities ceases to be elastic and their priceshave to rise to whatever level is necessary to divert demandinto other directions.

It is probable that the general level of prices will not risevery much as output increases, so long as there are availableefficient unemployed resources of every type. But as soon as

294

IV.

output has increased sufficiently to begin to reach the “bottle-necks”, there is likely to be a sharp rise in the prices of certaincommodities.

Under this heading, however, as also under heading (2),the elasticity of supply partly depends on the elapse of time.If we assume a sufficient interval for the quantity of equip-ment itself to change, the elasticities of supply will be decid-edly greater eventually. Thus a moderate change in effectivedemand, coming on a situation where there is widespreadunemployment, may spend itself very little in raising pricesand mainly in increasing employment; whilst a larger change,which, being unforeseen, causes some temporary “bottlenecks”to be reached, will spend itself in raising prices, as distinctfrom employment, to a greater extent at first than subsequently.

(4) That the wage-unit may tend to rise before full employ-ment has been reached, requires little comment or explana-tion. Since each group of workers will gain, cet. par., by a risein its own wages, there is naturally for all groups a pressurein this direction, which entrepreneurs will be more ready tomeet when they are doing better business. For this reason aproportion of any increase in effective demand is likely to beabsorbed in satisfying the upward tendency of the wage-unit.

Thus, in addition to the final critical point of full employ-ment at which money-wages have to rise, in response to anincreasing effective demand in terms of money, fully in pro-portion to the rise in the prices of wage-goods, we have asuccession of earlier semi-critical points at which an increas-ing effective demand tends to raise money-wages though notfully in proportion to the rise in the price of wage-goods; andsimilarly in the case of a decreasing effective demand. In ac-tual experience the wage-unit does not change continuouslyin terms of money in response to every small change in effec-tive demand; but discontinuously. These points of disconti-nuity are determined by the psychology of the workers andby the policies of employers and trade unions. In an opensystem, where they mean a change relatively to wage-costselsewhere, and in a trade cycle, where even in a closed system

295


they may mean a change relatively to expected wage-costs inthe future, they can be of considerable practical significance.These points, where a further increase in effective demand interms of money is liable to cause a discontinuous rise in thewage-unit, might be deemed, from a certain point of view, tobe positions of semi-inflation, having some analogy (thougha very imperfect one) to the absolute inflation (cf. Section Vbelow) which ensues on an increase in effective demand incircumstances of full employment. They have, moreover, agood deal of historical importance. But they do not readilylend themselves to theoretical generalisations.

(5) Our first simplification consisted in assuming that theremunerations of the various factors entering into marginalcost all change in the same proportion. But in fact the rates ofremuneration of different factors in terms of money will showvarying degrees of rigidity and they may also have differentelasticities of supply in response to changes in the money-rewards offered. If it were not for this, we could say that theprice-level is compounded of two factors, the wage-unit andthe quantity of employment.

Perhaps the most important element in marginal cost whichis likely to change in a different proportion from the wage-unit, and also to fluctuate within much wider limits, is marginaluser cost. For marginal user cost may increase sharply whenemployment begins to improve, if (as will probably be thecase) the increasing effective demand brings a rapid changein the prevailing expectation as to the date when the replace-ment of equipment will be necessary.

Whilst it is for many purposes a very useful first approxi-mation to assume that the rewards of all the factors enteringinto marginal prime-cost change in the same proportion asthe wage-unit, it might be better, perhaps, to take a weightedaverage of the rewards of the factors entering into marginalprime-cost, and call this the cost-unit. The cost-unit, or, sub-ject to the above approximation, the wage-unit, can thus beregarded as the essential standard of value; and the price-level, given the state of technique and equipment, will depend

296

V.

partly on the cost-unit and partly on the scale of output, in-creasing, where output increases, more than in proportion toany increase in the cost-unit, in accordance with the princi-ple of diminishing returns in the short period. We have fullemployment when output has risen to a level at which themarginal return from a representative unit of the factors ofproduction has fallen to the minimum figure at which a quan-tity of the factors sufficient to produce this output is available.

V

When a further increase in the quantity of effective demandproduces no further increase in output and entirely spends it-self on an increase in the cost-unit fully proportionate to theincrease in effective demand, we have reached a conditionwhich might be appropriately designated as one of true in-flation. Up to this point the effect of monetary expansion isentirely a question of degree, and there is no previous pointat which we can draw a definite line and declare that condi-tions of inflation have set in. Every previous increase in thequantity of money is likely, in so far as it increases effectivedemand, to spend itself partly in increasing the cost-unit andpartly in increasing output.

It appears, therefore, that we have a sort of asymmetry onthe two sides of the critical level above which true inflationsets in. For a contraction of effective demand below the criticallevel will reduce its amount measured in cost-units; whereasan expansion of effective demand beyond this level will not,in general, have the effect of increasing its amount in terms ofcost-units. This result follows from the assumption that thefactors of production, and in particular the workers, are dis-posed to resist a reduction in their money-rewards, and thatthere is no corresponding motive to resist an increase. Thisassumption is, however, obviously well founded in the facts,due to the circumstance that a change, which is not an all-round change, is beneficial to the special factors affected whenit is upward and harmful when it is downward.

297


If, on the contrary, money-wages were to fall without limitwhenever there was a tendency for less than full employment,the asymmetry would, indeed, disappear. But in that casethere would be no resting-place below full employment un-til either the rate of interest was incapable of falling further orwages were zero. In fact we must have some factor, the valueof which in terms of money is, if not fixed, at least sticky, togive us any stability of values in a monetary system.

The view that any increase in the quantity of money is in-flationary (unless we mean by inflationary merely that pricesare rising) is bound up with the underlying assumption of theclassical theory that we are always in a condition where a re-duction in the real rewards of the factors of production willlead to a curtailment in their supply.

VI

With the aid of the notation introduced in Chapter 20 we can,if we wish, express the substance of the above in symbolicform.

Let us write MV = D where M is the quantity of money,V its income-velocity (this definition differing in the minorrespects indicated above from the usual definition) and D theeffective demand. If, then, V is constant, prices will change inthe same proportion as the quantity of money provided thatep (= Ddp

pdD), is unity. This condition is satisfied (see Chapter

20 §I above) if eo = 0 or if ew = 1. The condition ew = 1means that the wage-unit in terms of money rises in the sameproportion as the effective demand, since ew = DdW

WdD; and

the condition eo = 0 means that output no longer shows anyresponse to a further increase in effective demand, since eo =DdOOdD

. Output in either case will be unaltered.Next, we can deal with the case where income-velocity is

not constant, by introducing yet a further elasticity, namelythe elasticity, of effective demand in response to changes in

298

VI.

the quantity of money,

ed =MdD

DdM.

This gives us

Mdp

pdM= ep × ed where ep = 1− ee · eo(1− ew) ;

so that

e = ed − (1− ew)ed · ee · eo

= ed(1− ee · eo + ee · eo · ew)

where e without suffix (= MdppdM

) stands for the apex of thispyramid and measures the response of money-prices to changesin the quantity of money.

Since this last expression gives us the proportionate changein prices in response to a change in the quantity of money,it can be regarded as a generalised statement of the Quan-tity Theory of Money. I do not myself attach much valueto manipulations of this kind; and I would repeat the warn-ing, which I have given above, that they involve just as muchtacit assumption as to what variables are taken as indepen-dent (partial differentials being ignored throughout) as doesordinary discourse, whilst I doubt if they carry us any fur-ther than ordinary discourse can. Perhaps the best purposeserved by writing them down is to exhibit the extreme com-plexity of the relationship between prices and the quantity ofmoney, when we attempt to express it in a formal manner. Itis, however, worth pointing out that, of the four terms ed, ew,ee and eo upon which the effect on prices of changes in thequantity of money depends, ed stands for the liquidity factorswhich determine the demand for money in each situation, ew

for the labour factors (or, more strictly, the factors enteringinto prime-cost) which determine the extent to which money-wages are raised as employment increases, and ee and eo for

299


the physical factors which determine the rate of decreasingreturns as more employment is applied to the existing equip-ment.

If the public hold a constant proportion of their income inmoney, ed = 1; if money-wages are fixed, ew = 0; if there areconstant returns throughout so that marginal return equalsaverage return, eeeo = 1; and if there is full employment ei-ther of labour or of equipment, eeeo = 0.

Now e = 1, if ed = 1 and ew = 1; or if ed = 1, ew = 0 andeeeo = 1; or if ed = 1 and eo = 0. And obviously there is avariety of other special cases in which e = 1. But in general eis not unity; and it is, perhaps, safe to make the generalisationthat on plausible assumptions relating to the real world, andexcluding the case of a “flight from the currency” in which ed

and ew become large, e is, as a rule, less than unity.

VII

So far, we have been primarily concerned with the way inwhich changes in the quantity of money affect prices in theshort period. But in the long run is there not some simplerrelationship?

This is a question for historical generalisation rather thanfor pure theory. If there is some tendency to a measure oflong-run uniformity in the state of liquidity-preference, theremay well be some sort of rough relationship between the na-tional income and the quantity of money required to satisfyliquidity-preference, taken as a mean over periods of pessimismand optimism together. There may be, for example, somefairly stable proportion of the national income more than whichPeople will not readily keep in the shape of idle balances orlong periods together, provided the rate of interest exceedsa certain psychological minimum; so that if the quantity ofmoney beyond what is required in the active circulation is inexcess of this proportion of the national income, there willbe a tendency sooner or later for the rate of interest to fallto the neighbourhood of this minimum. The falling rate of

300

VII.

interest will then, cet. par., increase effective demand, andthe increasing effective demand will reach one or more of thesemi-critical points at which the wage-unit will tend to showa discontinuous rise, with a corresponding effect on prices.The opposite tendencies will set in if the quantity of surplusmoney is an abnormally low proportion of the national in-come. Thus the net effect of fluctuations over a period of timewill be to establish a mean figure in conformity with the sta-ble proportion between the national income and the quantityof money to which the psychology of the public tends sooneror later to revert.

These tendencies will probably work with less friction inthe upward than in the downward direction. But if the quan-tity of money remains very deficient for a long time, the es-cape will be normally found in changing the monetary stan-dard or the monetary system so as to raise the quantity ofmoney, rather than in forcing down the wage-unit and therebyincreasing the burden of debt. Thus the very long-run courseof prices has almost always been upward. For when money isrelatively abundant, the wage-unit rises; and when money isrelatively scarce, some means is found to increase the effectivequantity of money.

During the nineteenth century, the growth of populationand of invention, the opening-up of new lands, the state ofconfidence and the frequency of war over the average of (say)each decade seem to have been sufficient, taken in conjunc-tion with the propensity to consume, to establish a scheduleof the marginal efficiency of capital which allowed a reason-ably satisfactory average level of employment to be compat-ible with a rate of interest high enough to be psychologicallyacceptable to wealth-owners. There is evidence that for a pe-riod of almost one hundred and fifty years the long-run typ-ical rate of interest in the leading financial centres was about5 pet cent., and the gilt-edged rate between 3 and 3 1/2 percent.; and that these rates of interest were modest enough toencourage a rate of investment consistent with an average ofemployment which was not intolerably low. Sometimes thewage-unit, but more often the monetary standard or the mon-

301


etary system (in particular through the development of bank-money), would be adjusted so as to ensure that the quantityof money in terms of wage-units was sufficient to satisfy nor-mal liquidity-preference at rates of interest which were sel-dom much below the standard rates indicated above. Thetendency of the wage-unit was, as usual, steadily upwardson the whole, but the efficiency of labour was also increasing.Thus the balance of forces was such as to allow a fair measureof stability of prices; — the highest quinquennial average forSauerbeck’s index number between 1820 and 1914 was only50 per cent. above the lowest. This was not accidental. It isrightly described as due to a balance of forces in an age whenindividual groups of employers were strong enough to pre-vent the wage-unit from rising much faster than the efficiencyof production, and when monetary systems were at the sametime sufficiently fluid and sufficiently conservative to providean average supply of money in terms of wage-units whichallowed to prevail the lowest average rate of interest read-ily acceptable by wealth-owners under the influence of theirliquidity-preferences. The average level of employment was,of course, substantially below full employment, but not so in-tolerably below it as to provoke revolutionary changes.

To-day and presumably for the future the schedule of themarginal efficiency of capital is, for a variety of reasons, muchlower than it was in the nineteenth century. The acuteness andthe peculiarity of our contemporary problem arises, therefore,out of the possibility that the average rate of interest whichwill allow a reasonable average level of employment is oneso unacceptable to wealth-owners that it cannot be readily es-tablished merely by manipulating the quantity of money. Solong as a tolerable level of employment could be attained onthe average of one or two or three decades merely by assuringan adequate supply of money in terms of wage-units, eventhe nineteenth century could find a way. If this was our onlyproblem now — if a sufficient degree of devaluation is all weneed — we, to-day, would certainly find a way.

But the most stable, and the least easily shifted, elementin our contemporary economy has been hitherto, and may

302

VII.

prove to be in future, the minimum rate of interest accept-able to the generality of wealth-owners.2 If a tolerable levelof employment requires a rate of interest much below the av-erage rates which ruled in the nineteenth century, it is mostdoubtful whether it can be achieved merely by manipulatingthe quantity of money. From the percentage gain, which theschedule of marginal efficiency of capital allows the borrowerto expect to earn, there has to be deducted (1) the cost of bring-ing borrowers and lenders together, (2) income and surtaxesand (3) the allowance which the lender requires to cover hisrisk and uncertainty, before we arrive at the net yield availableto tempt the wealth-owner to sacrifice his liquidity. If, in con-ditions of tolerable average employment, this net yield turnsout to be infinitesimal, time-honoured methods may proveunavailing.

To return to our immediate subject, the long-run relation-ship between the national income and the quantity of moneywill depend on liquidity-preferences. And the long-run sta-bility or instability of prices will depend on the strength ofthe upward trend or the wage-unit (or, more precisely, of thecost-unit) compared with the rate of increase in the efficiencyof the productive system.

2Cf. the nineteenth-century saying, quoted by Bagehot, that “John Bull canstand many things, but he cannot stand 2 per cent.”

303

Book VI

Short Notes Suggested by theGeneral Theory

CHAPTER 22

NOTES ON THE TRADE CYCLE

Since we claim to have shown in the preceding chapters whatdetermines the volume of employment at any time, it follows,if we are right, that our theory must be capable of explainingthe phenomena of the Trade Cycle.

If we examine the details of any actual instance of the TradeCycle, we shall find that it is highly complex and that everyelement in our analysis will be required for its complete ex-planation. In particular we shall. find that fluctuations inthe propensity to consume, in the state of liquidity-preference,and in the marginal efficiency of capital have all played a part.But I suggest that the essential character of the Trade Cycle,and, especially, the regularity of time-sequence and of dura-tion which justifies us in calling it a cycle, is mainly due to theway in which the marginal efficiency of capital fluctuates. TheTrade Cycle is best regarded, I think, as being occasioned bya cyclical change in the marginal efficiency of capital, thoughcomplicated. and often aggravated by associated changes inthe other significant short-period variables of the economicsystem. To develop this thesis would occupy a book ratherthan a chapter, and would require a close examination of facts.But the following short notes will be sufficient to indicate theline of investigation which our preceding theory suggests.

I

By a cyclical movement we mean that as the system progressesin, e.g., the upward direction, the forces propelling it upwardsat first gather force and have a cumulative effect on one an-other but gradually lose their strength until at a certain pointthey tend to be replaced by forces operating in the opposite

307

22. NOTES ON THE TRADE CYCLE

direction; which in turn gather force for a time and accentuateone another, until they too, having reached their maximumdevelopment, wane and give place to their opposite. We donot, however, merely mean by a cyclical movement that up-ward and downward tendencies, once started, do not persistfor ever in the same direction but are ultimately reversed. Wemean also that there is some recognisable degree of regularityin the time sequence and duration of the upward and down-ward movements.

There is, however, another characteristic of what we callthe Trade Cycle which our explanation must cover if it is to beadequate; namely, the phenomenon of the crisis — the fact thatthe substitution of a downward for an upward tendency oftentakes place suddenly and violently, whereas there is, as a rule,no such sharp turning-point when an upward is substitutedfor a downward tendency.

Any fluctuation in investment not offset by a correspond-ing change in the propensity to consume will, of course, re-sult in a fluctuation in employment. Since, therefore, the vol-ume of investment is subject to highly complex influences, itis highly improbable that all fluctuations either in investmentitself or in the marginal efficiency of capital will be of a cyclicalcharacter. One special case, in particular, namely, that whichis associated with agricultural fluctuations, will be separatelyconsidered in a later section of this chapter. I suggest, how-ever, that there are certain definite reasons why, in the case ofa typical industrial trade cycle in the nineteenth-century en-vironment, fluctuations in the marginal efficiency of capitalshould have had cyclical characteristics. These reasons are byno means unfamiliar either in themselves or as explanationsof the trade cycle. My only purpose here is to link them upwith the preceding theory.

II

I can best introduce what I have to say by beginning with thelater stages of the boom and the onset of the “crisis”.

308

II.

We have seen above that the marginal efficiency of capital1

depends, not only on the existing abundance or scarcity ofcapital-goods and the current cost of production of capital-goods, but also on current expectations as to the future yieldof capital-goods. In the case of durable assets it is, therefore,natural and reasonable that expectations of the future shouldplay a dominant part in determining the scale on which newinvestment is deemed advisable. But, as we have seen, thebasis for such expectations is very precarious. Being based onshifting and unreliable evidence, they are subject to suddenand violent changes.

Now, we have been accustomed in explaining the “crisis2to lay stress on the rising tendency of the rate of interest un-der the influence of the increased demand for money both fortrade and speculative purposes. At times this factor may cer-tainly play an aggravating and, occasionally perhaps, an ini-tiating part. But I suggest that a more typical, and often thepredominant, explanation of the crisis is, not primarily a risein the rate of interest, but a sudden collapse in the marginalefficiency of capital.

The later stages of the boom are characterised by optimisticexpectations as to the future yield of capital-goods sufficientlystrong to offset their growing abundance and their rising costsof production and, probably, a rise in the rate of interest also.It is of the nature of organised investment markets, under theinfluence of purchasers largely ignorant of what they are buy-ing and of speculators who are more concerned with forecast-ing the next shift of market sentiment than with a reasonableestimate of the future yield of capital-assets, that, when dis-illusion falls upon an over-optimistic and over-bought mar-ket, it should fall with sudden and even catastrophic force.2

Moreover, the dismay and uncertainty as to the future which

1It is often convenient in contexts where there is no room for misunder-standing to write “the marginal efficiency of capital”, where “the schedule ofthe marginal efficiency of capital” is meant.

2I have shown above (Chapter 12) that, although the private investor isseldom himself directly responsible for new investment, nevertheless the en-trepreneurs, who are directly responsible, will find it financially advantageous,

309


accompanies a collapse in the marginal efficiency of capitalnaturally precipitates a sharp increase in liquidity-preference— and hence a rise in the rate of interest. Thus the fact that acollapse in the marginal efficiency of capital tends to be associ-ated with a rise in the rate of interest may seriously aggravatethe decline in investment. But the essence of the situation isto be found, nevertheless, in the collapse in the marginal effi-ciency of capital, particularly in the case of those types of cap-ital which have been contributing most to the previous phaseof heavy new investment. Liquidity-preference, except thosemanifestations of it which are associated with increasing tradeand speculation, does not increase until after the collapse inthe marginal efficiency of capital.

It is this, indeed, which renders the slump so intractable.Later on, a decline in the rate of interest will be a great aidto recovery and, probably, a necessary condition of it. But,for the moment, the collapse in the marginal efficiency of cap-ital may be so complete that no practicable reduction in therate of interest will be enough. If a reduction in the rate ofinterest was capable of proving an effective remedy by itself,it might be possible to achieve a recovery without the elapseof any considerable interval of time and by means more orless directly under the control of the monetary authority. But,in fact, this is not usually the case; and it is not so easy torevive the marginal efficiency of capital, determined, as it is,by the uncontrollable and disobedient psychology of the busi-ness world. It is the return of confidence, to speak in ordinarylanguage, which is so insusceptible to control in an economyof individualistic capitalism. This is the aspect of the slumpwhich bankers and business men have been right in empha-sising, and which the economists who have put their faith ina “purely monetary” remedy have underestimated.

This brings me to my point. The explanation of the time-element in the trade cycle, of the fact that an interval of timeof a particular order of magnitude must usually elapse before

and often unavoidable, to fall in with the ideas of the market, even though theythemselves are better instructed.

310

II.

recovery begins, is to be sought in the influences which gov-ern the recovery of the marginal efficiency of capital. Thereare reasons, given firstly by the length of life of durable as-sets in relation to the normal rate of growth in a given epoch,and secondly, by the carrying-costs of surplus stocks, why theduration of the downward movement should have an orderof magnitude which is not fortuitous, which does not fluctu-ate between, say, one year this time and ten years next time,but which shows some regularity of habit between, let us say,three and five years.

Let us recur to what happens at the crisis. So long as theboom was continuing, much of the new investment showeda not unsatisfactory current yield. The disillusion comes be-cause doubts suddenly arise concerning the reliability of theprospective yield, perhaps because the current yield showssigns of failing off, as the stock of newly produced durablegoods steadily increases. If current costs of production arethought to be higher than they will be later on, that will bea further reason for a fall in the marginal efficiency of capital.Once doubt begins it spreads rapidly. Thus at the outset of theslump there is probably much capital of which the marginalefficiency has become negligible or even negative. But the in-terval of time, which will have to elapse before the shortageof capital through use, decay and obsolescence causes a suf-ficiently obvious scarcity to increase the marginal efficiency,may be a somewhat stable function of the average durabilityof capital in a given epoch. If the characteristics of the epochshift, the standard time-interval will change. If, for example,we pass from a period of increasing population into one ofdeclining population, the characteristic phase of the cycle willbe lengthened. But we have in the above a substantial reasonwhy the duration of the slump should have a definite relation-ship to the length of life of durable assets and to the normalrate of growth in a given epoch.

The second stable time-factor is due to the carrying-costsof surplus stocks which force their absorption within a certainperiod, neither very short nor very long. The sudden cessa-tion of new investment after the crisis will probably lead to

311


an accumulation of surplus stocks of unfinished goods. Thecarrying-costs of these stocks will seldom be less than 10 percent. per annum. Thus the fall in their price needs to be suffi-cient to bring about a restriction which provides for their ab-sorption within a period of, say, three to five years at the out-side. Now the process of absorbing the stocks represents neg-ative investment, which is a further deterrent to employment;and, when it is over, a manifest relief will be experienced.

Moreover, the reduction in working capital, which is nec-essarily attendant on the decline in output on the downwardphase, represents a further element of disinvestment, whichmay be large; and, once the recession has begun, this exertsa strong cumulative influence in the downward direction. Inthe earliest phase of a typical slump there will probably bean investment in increasing stocks which helps to offset disin-vestment in working-capital; in the next phase there may be ashort period of disinvestment both in stocks and in working-capital; after the lowest point has been passed there is likelyto be a further disinvestment in stocks which partially offsetsreinvestment in working-capital; and, finally, after the recov-ery is well on its way, both factors will be simultaneouslyfavourable to investment. It is against this background thatthe additional and superimposed effects of fluctuations of in-vestment in durable goods must be examined. When a declinein this type of investment has set a cyclical fluctuation in mo-tion there will be little encouragement to a recovery in suchinvestment until the cycle has partly run its course.3

Unfortunately a serious fall in the marginal efficiency ofcapital also tends to affect adversely the propensity to con-sume. For it involves a severe decline in the market value ofStock Exchange equities. Now, on the class who take an ac-tive interest in their Stock Exchange investments, especiallyif they are employing, borrowed funds, this naturally exertsa very depressing influence. These people are, perhaps, evenmore influenced in their readiness to spend by rises and falls

3Some part of the discussion in my Treatise on Money, Book IV, bears uponthe above.

312

III.

in the value of their investments than by the state of their in-come. With a “stock-minded” public, as in the United Statesto-day, a rising stock-market may be an almost essential con-dition of a satisfactory propensity to consume; and this cir-cumstance, generally overlooked until lately, obviously servesto aggravate still further the depressing effect of a decline inthe marginal efficiency of capital.

When once the recovery has been started, the manner inwhich it feeds on itself and cumulates is obvious. But dur-ing the downward phase, when both fixed capital and stocksof materials are for the time being redundant and working-capital is being reduced, the schedule of the marginal effi-ciency of capital may fall so low that it can scarcely be cor-rected, so as to secure a satisfactory rate of new investment,by any practicable reduction in the rate of interest. Thus withmarkets organised and influenced as they are at present, themarket estimation of the marginal efficiency of capital maysuffer such enormously wide fluctuations that it cannot besufficiently offset by corresponding fluctuations in the rateof interest. Moreover, the corresponding movements in thestock-market may, as we have seen above, depress the propen-sity to consume just when it is most needed. In conditionsof laissez-faire the avoidance of wide fluctuations in employ-ment may, therefore, prove impossible without a far-reachingchange in the psychology of investment markets such as thereis no reason to expect. I conclude that the duty of ordering thecurrent volume of investment cannot safely be left in privatehands.

III

The preceding analysis may appear to be in conformity withthe view of those who hold that over-investment is the charac-teristic of the boom, that the avoidance of this over-investmentis the only possible remedy for the ensuing slump, and that,whilst for the reasons given above the slump cannot be pre-vented by a low rate of interest, nevertheless the boom can be

313


avoided by a high rate of interest. There is, indeed, force inthe argument that a high rate of interest is much more effec-tive against a boom than a low rate of interest against a slump.

To infer these conclusions from the above would, however,misinterpret my analysis; and would, according to my way ofthinking, involve serious error. For the term over-investmentis ambiguous. It may refer to investments which are destinedto disappoint the expectations which prompted them or forwhich there is no use in conditions of severe unemployment,or it may indicate a state of affairs where every kind of capital-goods is so abundant that there is no new investment which isexpected, even in conditions of full employment, to earn in thecourse of its life more than its replacement cost. It is only thelatter state of affairs which is one of over-investment, strictlyspeaking, in the sense that any further investment would be asheer waste of resources.4 Moreover, even if over-investmentin this sense was a normal characteristic of the boom, the rem-edy would not lie in clapping on a high rate of interest whichwould probably deter some useful investments and might fur-ther diminish the propensity to consume, but in taking drasticsteps, by redistributing incomes or otherwise, to stimulate thepropensity to consume.

According to my analysis, however, it is only in the formersense that the boom can be said to be characterised by over-investment. The situation, which I am indicating as typical, isnot one in which capital is so abundant that the communityas a whole has no reasonable use for any more, but whereinvestment is being made in conditions which are unstableand cannot endure, because it is prompted by expectationswhich are destined to disappointment.

It may, of course, be the case — indeed it is likely to be —that the illusions of the boom cause particular types of capital-assets to be produced in such excessive abundance that some

4On certain assumptions, however, as to the distribution of the propensityto consume through time, investment which yielded a negative return might beadvantageous in the sense that, for the community as a whole, it would max-imise satisfaction.

314

III.

part of the output is, on any criterion, a waste of resources;— which sometimes happens, we may add, even when thereis no boom. It leads, that is to say, to misdirected investment.But over and above this it is an essential characteristic of theboom that investments which will in fact yield, say, 2 per cent.in conditions of full employment are made in the expecta-tion of a yield of, say, 6 per cent., and are valued accordingly.When the disillusion comes, this expectation is replaced bya contrary “error of pessimism”, with the result that the in-vestments, which would in fact yield 2 per cent. in condi-tions of full employment, are expected to yield less than noth-ing; and the resulting collapse of new investment then leadsto a state of unemployment in which the investments, whichwould have yielded 2 per cent. in conditions of full employ-ment, in fact yield less than nothing. We reach a conditionwhere there is a shortage of houses, but where neverthelessno one can afford to live in the houses that there are.

Thus the remedy for the boom is not a higher rate of in-terest but a lower rate of interest!5 For that may enable theso-called boom to last. The right remedy for the trade cycleis not to be found in abolishing booms and thus keeping uspermanently in a semi-slump; but in abolishing slumps andthus keeping us permanently in a quasi-boom.

The boom which is destined to end in a slump is caused,therefore, by the combination of a rate of interest, which in acorrect state of expectation would be too high for full employ-ment, with a misguided state of expectation which, so long asit lasts, prevents this rate of interest from being in fact deter-rent. A boom is a situation in which over-optimism triumphsover a rate of interest which, in a cooler light, would be seento be excessive.

Except during the war, I doubt if we have any recent expe-rience of a boom so strong that it led to full employment. In

5See below (p. 327) for some arguments which can be urged on the otherside. For, if we are precluded from making large changes in our present meth-ods, I should agree that to raise the rate of interest during a boom may be, inconceivable circumstances, the lesser evil.

315


the United States employment was very satisfactory in 1928-29 on normal standards; but I have seen no evidence of ashortage of labour, except, perhaps, in the case of a few groupsof highly specialised workers. Some “bottle-necks” were reached,but output as a whole was still capable of further expansion.Nor was there over-investment in the sense that the standardand equipment of housing was so high that everyone, assum-ing full employment, had all he wanted at a rate which wouldno more than cover the replacement cost, without any allowancefor interest, over the life of the house; and that transport, pub-lic services and agricultural improvement had been carried toa point where further additions could not reasonably be ex-pected to yield even their replacement cost. Quite the con-trary. It would be absurd to assert of the United States in1929 the existence of over-investment in the strict sense. Thetrue state of affairs was of a different character. New invest-ment during the previous five years had been, indeed, on soenormous a scale in the aggregate that the prospective yieldof further additions was, coolly considered, falling rapidly.Correct foresight would have brought down the marginal ef-ficiency of capital to an unprecedentedly low figure; so thatthe “boom” could not have continued on a sound basis exceptwith a very low long-term rate of interest, and an avoidanceof misdirected investment in the particular directions whichwere in danger of being over-exploited. In fact, the rate ofinterest was high enough to deter new investment except inthose particular directions which were under the influence ofspeculative excitement and, therefore, in special danger of be-ing over-exploited; and a rate of interest, high enough to over-come the speculative excitement, would have checked, at thesame time, every kind of reasonable new investment. Thusan increase in the rate of interest, as a remedy for the state ofaffairs arising out of a prolonged period of abnormally heavynew investment, belongs to the species of remedy which curesthe disease by killing the patient.

It is, indeed, very possible that the prolongation of approx-imately full employment over a period of years would be as-sociated in countries so wealthy as Great Britain or the United

316

IV.

States with a volume of new investment, assuming the exist-ing propensity to consume, so great that it would eventuallylead to a state of full investment in the sense that an aggre-gate gross yield in excess of replacement cost could no longerbe expected on a reasonable calculation from a further incre-ment of durable goods of any type whatever. Moreover, thissituation might be reached comparatively soon — say withintwenty-five years or less. I must not be taken to deny this, be-cause I assert that a state of full investment in the strict sensehas never yet occurred, not even momentarily.

Furthermore, even if we were to suppose that contempo-rary booms are apt to be associated with a momentary condi-tion of full investment or over-investment in the strict sense, itwould still be absurd to regard a higher rate of interest as theappropriate remedy. For in this event the case of those whoattribute the disease to under-consumption would be whollyestablished. The remedy would lie in various measures de-signed to increase the propensity to consume by the redistri-bution of incomes or otherwise; so that a given level of em-ployment would require a smaller volume of current invest-ment to support it.

IV

It may be convenient at this point to say a word about theimportant schools of thought which maintain, from variouspoints of view, that the chronic tendency of contemporary so-cieties to under-employment is to be traced to under-consumption;— that is to say, to social practices and to a distribution ofwealth which result in a propensity to consume which is un-duly low.

In existing conditions — or, at least, in the conditions whichexisted until lately — where the volume of investment is un-planned and uncontrolled, subject to the vagaries of the marginalefficiency of capital as determined by the private judgment ofindividuals ignorant or speculative, and to a long-term rateof interest which seldom or never falls below a conventional

317


level, these schools of thought are, as guides to practical pol-icy, undoubtedly in the right. For in such conditions there isno other means of raising the average level of employment toa more satisfactory level. If it is impracticable materially toincrease investment, obviously there is no means of securinga higher level of employment except by increasing consump-tion.

Practically I only differ from these schools of thought inthinking that they may lay a little too much emphasis on in-creased consumption at a time when there is still much socialadvantage to be obtained from increased investment. Theo-retically, however, they are open to the criticism of neglectingthe fact that there are two ways to expand output. Even if wewere to decide that it would be better to increase capital moreslowly and to concentrate effort on increasing consumption,we must decide this with open eyes, after well consideringthe alternative. I am myself impressed by the great social ad-vantages of increasing the stock of capital until it ceases to bescarce. But this is a practical judgment, not a theoretical im-perative.

Moreover, I should readily concede that the wisest courseis to advance on both fronts at once. Whilst aiming at a so-cially controlled rate of investment with a view to a progres-sive decline in the marginal efficiency of capital, I should sup-port at the same time all sorts of policies for increasing thepropensity to consume. For it is unlikely that full employ-ment can be maintained, whatever we may do about invest-ment, with the existing propensity to consume. There is room,therefore, for both policies to operate together; — to promoteinvestment and, at the same time, to promote consumption,not merely to the level which with the existing propensity toconsume would correspond to the increased investment, butto a higher level still.

If — to take round figures for the purpose of illustration— the average level of output of to-day is 15 per cent. be-low what it would be with continuous full employment, andif 10 per cent. of this output represents net investment andgo per cent. of it consumption — if, furthermore, net invest-

318

V.

ment would have to rise 50 per cent. in order to secure fullemployment with the existing propensity to consume, so thatwith full employment output would rise from 100 to 115, con-sumption from go to 100 and net investment from 10 to 15: —then we might aim, perhaps, at so modifying the propensityto consume that with full employment consumption wouldrise from go to 103 and net investment from 10 to 12.

V

Another school of thought finds the solution of the trade cy-cle, not in increasing either consumption or investment, butin diminishing the supply of labour seeking employment; i.e.by redistributing the existing volume of employment withoutincreasing employment or output.

This seems to me to be a premature policy — much moreclearly so than the plan of increasing consumption. A pointcomes where every individual weighs the advantages of in-creased leisure against increased income. But at present theevidence is, I think, strong that the great majority of individu-als would prefer increased income to increased leisure; and Isee no sufficient reason for compelling those who would pre-fer more income to enjoy more leisure.

VI

It may appear extraordinary that a school of thought shouldexist which finds the solution for the trade cycle in checkingthe boom in its early stages by a higher rate of interest. Theonly line of argument, along which any justification for thispolicy can be discovered, is that put forward by Mr. D. H.Robertson, who assumes, in effect, that full employment is animpracticable ideal and that the best that we can hope for isa level of employment much more stable than at present andaveraging, perhaps, a little higher.

319


If we rule out major changes of policy affecting either thecontrol of investment or the propensity to consume, and as-sume, broadly speaking, a continuance of the existing state ofaffairs, it is, I think, arguable that a more advantageous av-erage state of expectation might result from a banking policywhich always nipped in the bud an incipient boom by a rateof interest high enough to deter even the most misguided op-timists. The disappointment of expectation, characteristic ofthe slump, may lead to so much loss and waste that the aver-age level of useful investment might be higher if a deterrent isapplied. It is difficult to be sure whether or not this is correcton its own assumptions; it is a matter for practical judgmentwhere detailed evidence is wanting. It may be that it over-looks the social advantage which accrues from the increasedconsumption which attends even on investment which provesto have been totally misdirected, so that even such investmentmay be more beneficial than no investment at all. Neverthe-less, the most enlightened monetary control might find itselfin difficulties, faced with a boom of the 1929 type in Amer-ica, and armed with no other weapons than those possessedat that time by the Federal Reserve System; and none of the al-ternatives within its power might make much difference to theresult. However this may be, such an outlook seems to me tobe dangerously and unnecessarily defeatist. It recommends,or at least assumes, for permanent acceptance too much thatis defective in our existing economic scheme.

The austere view, which would employ a high rate of inter-est to check at once any tendency in the level of employmentto rise appreciably above the average of, say, the previousdecade, is, however, more usually supported by argumentswhich have no foundation at all apart from confusion of mind.It flows, in some cases, from the belief that in a boom invest-ment tends to outrun saving, and that a higher rate of interestwill restore equilibrium by checking investment on the onehand and stimulating savings on the other. This implies thatsaving and investment can be unequal, and has, therefore, nomeaning until these terms have been defined in some specialsense. Or it is sometimes suggested that the increased sav-

320

VI.

ing which accompanies increased investment is undesirableand unjust because it is, as a rule, also associated with risingprices. But if this were so, any upward change in the existinglevel of output and employment is to be deprecated. For therise in prices is not essentially due to the increase in invest-ment; — it is due to the fact that in the short period supplyprice usually increases with increasing output, on account ei-ther of the physical fact of diminishing return or of the ten-dency of the cost-unit to rise in terms of money when out-put increases. If the conditions were those of constant supply-price, there would, of course, be no rise of prices; yet, all thesame, increased saving would accompany increased invest-ment. It is the increased output which produces the increasedsaving; and the rise of prices is, merely a by-product of theincreased output, which will occur equally if there is no in-creased saving but, instead, an increased propensity to con-sume. No one has a legitimate vested interest in being able tobuy at prices which are only low because output is low.

Or, again, the evil is supposed to creep in if the increasedinvestment has been promoted by a fall in the rate of interestengineered by an increase in the quantity of money. Yet thereis no special virtue in the pre-existing rate of interest, and thenew money is not “forced” on anyone; — it is created in or-der to satisfy the increased liquidity-preference which corre-sponds to the lower rate of interest or the increased volumeof transactions, and it is held by those individuals who pre-fer to hold money rather than to lend it at the lower rate ofinterest. Or, once more, it is suggested that a boom is charac-terised by “capital consumption”, which presumably meansnegative net investment, i.e. by an excessive propensity toconsume. Unless the phenomena of the trade cycle have beenconfused with those of a flight from the currency such as oc-curred during the post-war European currency collapses, theevidence is wholly to the contrary. Moreover, even if it wereso, a reduction in the rate of interest would be a more plausi-ble remedy than a rise in the rate of interest for conditions ofunder-investment. I can make no sense at all of these schoolsof thought; except, perhaps, by supplying a tacit assumption

321


that aggregate output is incapable of change. But a theorywhich assumes constant output is obviously not very service-able for explaining the trade cycle.

VII

In the earlier studies of the trade cycle, notably by Jevons, anexplanation was found in agricultural fluctuations due to theseasons, rather than in the phenomena of industry. In the lightof the above theory this appears as an extremely plausibleapproach to the problem. For even to-day fluctuation in thestocks of agricultural products as between one year and an-other is one of the largest individual items amongst the causesof changes in the rate of current investment; whilst at the timewhen Jevons wrote — and more particularly over the periodto which most of his statistics applied — this factor must havefar outweighed all others.

Jevons’s theory, that the trade cycle was primarily due tothe fluctuations in the bounty of the harvest, can be re-statedas follows. When an exceptionally large harvest is gathered in,an important addition is usually made to the quantity carriedover into later years. The proceeds of this addition are addedto the current incomes of the farmers and are treated by themas income; whereas the increased carry-over involves no drainon the income-expenditure of other sections of the communitybut is financed out of savings. That is to say, the addition tothe carry-over is an addition to current investment. This con-clusion is not invalidated even if prices fall sharply. Similarlywhen there is a poor harvest, the carry-over is drawn uponfor current consumption, so that a corresponding part of theincome-expenditure of the consumers creates no current in-come for the farmers. That is to say, what is taken from thecarry-over involves a corresponding reduction in current in-vestment. Thus, if investment in other directions is taken tobe constant, the difference in aggregate investment between ayear in which there is a substantial addition to the carry-overand a year in which there is a substantial subtraction from it

322

VII.

may be large; and in a community where agriculture is thepredominant industry it will be overwhelmingly large com-pared with any other usual cause of investment fluctuations.Thus it is natural that we should find the upward turning-point to be marked by bountiful harvests and the downwardturning-point by deficient harvests. The further theory, thatthere are physical causes for a regular cycle of good and badharvests, is, of course, a different matter with which we arenot concerned here.

More recently, the theory has been advanced that it is badharvests, not good harvests, which are good for trade, eitherbecause bad harvests make the population ready to work fora smaller real reward or because the resulting redistributionof purchasing-power is held to be favourable to consumption.Needless to say, it is not these theories which I have in mindin the above description of harvest phenomena as an explana-tion of the trade cycle.

The agricultural causes of fluctuation are, however, muchless important in the modern world for two reasons. In thefirst place agricultural output is a much smaller proportionof total output. And in the second place the developmentof a world market for most agricultural products, drawingupon both hemispheres, leads to an averaging out of the ef-fects of good and bad seasons, the percentage fluctuation inthe amount of the world harvest being far less than the per-centage fluctuations in the harvests of individual countries.But in old days, when a country was mainly dependent on itsown harvest, it is difficult to see any possible cause of fluc-tuations in investment, except war, which was in any waycomparable in magnitude with changes in the carry-over ofagricultural products.

Even to-day it is important to pay close attention to thepart played by changes in the stocks of raw materials, bothagricultural and mineral, in the determination of the rate ofcurrent investment. I should attribute the slow rate of recov-ery from a slump, after the turning-point has been reached,mainly to the deflationary effect of the reduction of redundantstocks to a normal level. At first the accumulation of stocks,

323


which occurs after the boom has broken, moderates the rateof the collapse; but we have to pay for this relief later on inthe damping-down of the subsequent rate of recovery. Some-times, indeed, the reduction of stocks may have to be virtuallycompleted before any measurable degree of recovery can bedetected. For a rate of investment in other directions, whichis sufficient to produce an upward movement when there isno current disinvestment in stocks to set off against it, maybe quite inadequate so long as such disinvestment is still pro-ceeding.

We have seen, I think, a signal example of this in the earlierphases of America’s “New Deal”. When President Roosevelt’ssubstantial loan expenditure began, stocks of all kinds — andparticularly of agricultural products — still stood at a veryhigh level. The “New Deal” partly consisted in a strenuousattempt to reduce these stocks — by curtailment of currentoutput and in all sorts of ways. The reduction of stocks to anormal level was a necessary process — a phase which hadto be endured. But so long as it lasted, namely, about twoyears, it constituted a substantial offset to the loan expendi-ture which was being incurred in other directions. Only whenit had been completed was the way prepared for substantialrecovery.

Recent American experience has also afforded good exam-ples of the part played by fluctuations in the stocks of finishedand unfinished goods — “inventories” as it is becoming usualto call them — in causing the minor oscillations within themain movement of the Trade Cycle. Manufacturers, setting in-dustry in motion to provide for a scale of consumption whichis expected to prevail some months later, are apt to make mi-nor miscalculations, generally in the direction of running a lit-tle ahead of the facts. When they discover their mistake theyhave to contract for a short time to a level below that of cur-rent consumption so as to allow for the absorption of the ex-cess inventories; and the difference of pace between runninga little ahead and dropping back again has proved sufficientin its effect on the current rate of investment to display itselfquite clearly against the background of the excellently com-

324

VII.

plete statistics now available in the United States.

325

CHAPTER 23

NOTES ON MERCANTILISM, THE USURY LAWS,STAMPED MONEY AND THEORIES OF

UNDER-CONSUMPTION

I

For some two hundred years both economic theorists and prac-tical men did not doubt that there is a peculiar advantage to acountry in a favourable balance of trade, and grave danger inan unfavourable balance, particularly if it results in an effluxof the precious metals. But for the past one hundred yearsthere has been a remarkable divergence of opinion. The ma-jority of statesmen and practical men in most countries, andnearly half of them even in Great Britain, the home of the op-posite view, have remained faithful to the ancient doctrine;whereas almost all economic theorists have held that anxietyconcerning such matters is absolutely groundless except on avery short view, since the mechanism of foreign trade is self-adjusting and attempts to interfere with it are not only futile,but greatly impoverish those who practise them because theyforfeit the advantages of the international division of labour. Itwill be convenient, in accordance with tradition, to designatethe older opinion as Mercantilism and the newer as Free Trade,though these terms, since each of them has both a broader anda narrower signification, must be interpreted with reference tothe context.

Generally speaking, modern economists have maintainednot merely that there is, as a rule, a balance of gain from theinternational division of labour sufficient to outweigh suchadvantages as mercantilist practice can fairly claim, but thatthe mercantilist argument is based, from start to finish, on anintellectual confusion.

327

23. NOTES ON MERCANTILISM, THE USURY LAWS, STAMPEDMONEY AND THEORIES OF UNDER-CONSUMPTION

Marshall,1 for example, although his references to Mercan-tilism are not altogether unsympathetic, had no regard fortheir central theory as such and does not even mention thoseelements of truth in their contentions which I shall examinebelow.2 In the same way, the theoretical concessions whichfree-trade economists have been ready to make in contem-porary controversies, relating, for example, to the encourage-ment of infant industries or to the improvement of the termsof trade, are not concerned with the real substance of the mer-cantilist case. During the fiscal controversy of the first quarterof the present century I do not remember that any concessionwas ever allowed by economists to the claim that Protectionmight increase domestic employment. It will be fairest, per-haps, to quote, as an example, what I wrote myself. So latelyas 1923, as a faithful pupil of the classical school who did notat that time doubt what he had been taught and entertainedon this matter no reserves at all, I wrote: “If there is one thingthat Protection can not do, it is to cure Unemployment. ...There are some arguments for Protection, based upon its se-curing possible but improbable advantages, to which thereis no simple answer. But the claim to cure Unemploymentinvolves the Protectionist fallacy in its grossest and crudestform.”3 As for earlier mercantilist theory, no intelligible ac-count was available; and we were brought up to believe thatit was little better than nonsense. So absolutely overwhelmingand complete has been the domination of the classical school.

1Vide his Industry and Trade, Appendix D; Money, Credit and Commerce, p.130; and Principles of Economics, Appendix I.

2His view of them is well summed up in a footnote to the first edition of hisPrinciples, p. 51: “Much study has been given both in England and Germany tomedieval opinions as to the relation of money to national wealth. On the wholethey are to be regarded as confused through want of a clear understanding ofthe functions of money, rather than as wrong in consequence of a deliberateassumption that the increase in the net wealth of a nation can be effected onlyby an increase of the stores of the precious metals in her.”

3The Nation and the Athenaeum, November 24, 1923.

328

II.

II

Let me first state in my own terms what now seems to me to bethe element of scientific truth in mercantilist doctrine. We willthen compare this with the actual arguments of the mercan-tilists. It should be understood that the advantages claimedare avowedly national advantages and are unlikely to benefitthe world as a whole.

When a country is growing in wealth somewhat rapidly,the further progress of this happy state of affairs is liable to beinterrupted, in conditions of laissez-faire, by the insufficiencyof the inducements to new investment. Given the social andpolitical environment and the national characteristics whichdetermine the propensity to consume, the well-being of a pro-gressive state essentially depends, for the reasons we havealready explained, on the sufficiency of such inducements.They may be found either in home investment or in foreigninvestment (including in the latter the accumulation of theprecious metals), which, between them, make up aggregateinvestment. In conditions in which the quantity of aggregateinvestment is determined by the profit motive alone, the op-portunities for home investment will be governed, in the longrun, by the domestic rate of interest; whilst the volume of for-eign investment is necessarily determined by the size of thefavourable balance of trade. Thus, in a society where there isno question of direct investment under the aegis of public au-thority, the economic objects, with which it is reasonable forthe government to be preoccupied, are the domestic rate ofinterest and the balance of foreign trade.

Now, if the wage-unit is somewhat stable and not liableto spontaneous changes of significant magnitude (a conditionwhich is almost always satisfied), if the state of liquidity-preferenceis somewhat stable, taken as an average of its short-periodfluctuations, and if banking conventions are also stable, therate of interest will tend to be governed by the quantity of theprecious metals, measured in terms of the wage-unit, avail-able to satisfy the community’s desire for liquidity. At thesame time, in an age in which substantial foreign loans and

329


the outright ownership of wealth located abroad are scarcelypracticable, increases and decreases in the quantity of the pre-cious metals will largely depend on whether the balance oftrade is favourable or unfavourable.

Thus, as it happens, a preoccupation on the part of the au-thorities with a favourable balance of trade served both pur-poses; and was, furthermore, the only available means of pro-moting them. At a time when the authorities had no directcontrol over the domestic rate of interest or the other induce-ments to home investment, measures to increase the favourablebalance of trade were the only direct means at their disposalfor increasing foreign investment; and, at the same time, theeffect of a favourable balance of trade on the influx of theprecious metals was their only indirect means of reducing thedomestic rate of interest and so increasing the inducement tohome investment.

There are, however, two limitations on the success of thispolicy which must not be overlooked. If the domestic rateof interest falls so low that the volume of investment is suffi-ciently stimulated to raise employment to a level which breaksthrough some of the critical points at which the wage-unitrises, the increase in the domestic level of costs will begin toreact unfavourably on the balance of foreign trade, so that theeffort to increase the latter will have overreached and defeateditself. Again, if the domestic rate of interest falls so low rela-tively to rates of interest elsewhere as to stimulate a volumeof foreign lending which is disproportionate to the favourablebalance, there may ensue an efflux of the precious metals suf-ficient to reverse the advantages previously obtained. The riskof one or other of these limitations becoming operative is in-creased in the case of a country which is large and internation-ally important by the fact that, in conditions where the currentoutput of the precious metals from the mines is on a relativelysmall scale, an influx of money into one country means an ef-flux from another; so that the adverse effects of rising costsand falling rates of interest at home may be accentuated (ifthe mercantilist policy is pushed too far) by falling costs andrising rates of interest abroad.

330

II.

The economic history of Spain in the latter part of the fif-teenth and in the sixteenth centuries provides an example ofa country whose foreign trade was destroyed by the effect onthe wage-unit of an excessive abundance of the precious met-als. Great Britain in the pre-war years of the twentieth centuryprovides an example of a country in which the excessive facil-ities for foreign lending and the purchase of properties abroadfrequently stood in the way of the decline in the domestic rateof interest which was required to ensure full employment athome. The history of India at all times has provided an ex-ample of a country impoverished by a preference for liquidityamounting to so strong a passion that even an enormous andchronic influx of the precious metals has been insufficient tobring down the rate of interest to a level which was compati-ble with the growth of real wealth.

Nevertheless, if we contemplate a society with a somewhatstable wage-unit, with national characteristics which deter-mine the propensity to consume and the preference for liq-uidity, and with a monetary system which rigidly links thequantity of money to the stock of the precious metals, it willbe essential for the maintenance of prosperity that the author-ities should pay close attention to the state of the balance oftrade. For a favourable balance, provided it is not too large,will prove extremely stimulating; whilst an unfavourable bal-ance may soon produce a state of persistent depression.

It does not follow from this that the maximum degree ofrestriction of imports will promote the maximum favourablebalance of trade. The earlier mercantilists laid great emphasison this and were often to be found opposing trade restrictionsbecause on a long view they were liable to operate adverselyto a favourable balance. It is, indeed, arguable that in the spe-cial circumstances of mid-nineteenth-century Great Britain analmost complete freedom of trade was the policy most con-ducive to the development of a favourable balance. Contem-porary experience of trade restrictions in post-war Europe of-fers manifold examples of ill-conceived impediments on free-dom which, designed to improve the favourable balance, hadin fact a contrary tendency.

331


For this and other reasons the reader must not reach a pre-mature conclusion as to the practical policy to which our ar-gument leads up. There are strong presumptions of a generalcharacter against trade restrictions unless they can be justifiedon special grounds. The advantages of the international divi-sion of labour are real and substantial, even though the classi-cal school greatly overstressed them. The fact that the advan-tage which our own country gains from a favourable balanceis liable to involve an equal disadvantage to some other coun-try (a point to which the mercantilists were fully alive) meansnot only that great moderation is necessary, so that a countrysecures for itself no larger a share of the stock of the preciousmetals than is fair and reasonable, but also that an immoder-ate policy may lead to a senseless international competitionfor a favourable balance which injures all alike.4 And finally,a policy of trade restrictions is a treacherous instrument evenfor the attainment of its ostensible object, since private inter-est, administrative incompetence and the intrinsic difficulty ofthe task may divert it into producing results directly oppositeto those intended.

Thus, the weight of my criticism is directed against the in-adequacy of the theoretical foundations of the laissez-faire doc-trine upon which I was brought up and which for many yearsI taught;— against the notion that the rate of interest and thevolume of investment are self-adjusting at the optimum level,so that preoccupation with the balance of trade is a waste oftime. For we, the faculty of economists, prove to have beenguilty of presumptuous error in treating as a puerile obses-sion what for centuries has been a prime object of practicalstatecraft.

Under the influence of this faulty theory the City of Lon-don gradually devised the most dangerous technique for themaintenance of equilibrium which can possibly be imagined,namely, the technique of bank rate coupled with a rigid par-

4The remedy of an elastic wage-unit, so that a depression is met by a re-duction of wages, is liable, for the same reason, to be a means of benefitingourselves at the expense of our neighbours.

332

II.

ity of the foreign exchanges. For this meant that the objec-tive of maintaining a domestic rate of interest consistent withfull employment was wholly ruled out. Since, in practice, itis impossible to neglect the balance of payments, a means ofcontrolling it was evolved which, instead of protecting the do-mestic rate of interest, sacrificed it to the operation of blindforces. Recently, practical bankers in London have learnt much,and one can almost hope that in Great Britain the technique ofbank rate will never be used again to protect the foreign bal-ance in conditions in which it is likely to cause unemploymentat home.

Regarded as the theory of the individual firm and of thedistribution of the product resulting from the employment ofa given quantity of resources, the classical theory has made acontribution to economic thinking which cannot be impugned.It is impossible to think clearly on the subject without this the-ory as a part of one’s apparatus of thought. I must not besupposed to question this in calling attention to their neglectof what was valuable in their predecessors. Nevertheless, asa contribution to statecraft, which is concerned with the eco-nomic system as whole and with securing the optimum em-ployment of the system’s entire resources, the methods of theearly pioneers of economic thinking in the sixteenth and sev-enteenth centuries may have attained to fragments of practi-cal wisdom which the unrealistic abstractions of Ricardo firstforgot and then obliterated. There was wisdom in their in-tense preoccupation with keeping down the rate of interestby means of usury laws (to which we will return later in thischapter), by maintaining the domestic stock of money and bydiscouraging rises in the wage-unit; and in their readiness inthe last resort to restore the stock of money by devaluation, ifit had become plainly deficient through an unavoidable for-eign drain, a rise in the wage-unit5, or any other cause.

5Experience since the age of Solon at least, and probably, if we had thestatistics, for many centuries before that, indicates what a knowledge of humannature would lead us to expect, namely, that there is a steady tendency for thewage-unit to rise over long periods of time and that it can be reduced only

333


III

The early pioneers of economic thinking may have hit upontheir maxims of practical wisdom without having had muchcognisance of the underlying theoretical grounds. Let us, there-fore, examine briefly the reasons they gave as well as whatthey recommended. This is made easy by reference to Pro-fessor Heckscher’s great work on Mercantilism, in which theessential characteristics of economic thought over a period oftwo centuries are made available for the first time to the gen-eral economic reader. The quotations which follow are mainlytaken from his pages.6

(a) Mercantilist thought never supposed that there was aself-adjusting tendency by which the rate of interest wouldbe established at the appropriate level. On the contrary theywere emphatic that an unduly high rate of interest was themain obstacle to the growth of wealth; and they were evenaware that the rate of interest depended on liquidity-preferenceand the quantity of money. They were concerned both withdiminishing liquidity-preference and with increasing the quan-tity of money, and several of them made it clear that their pre-occupation with increasing the quantity of money was due totheir desire to diminish the rate of interest. Professor Heckschersums up this aspect of their theory as follows:

The position of the more perspicacious mercan-tilists was in this respect, as in many others, per-fectly clear within certain limits. For them, moneywas—to use the terminology of to-day—a factor ofproduction, on the same footing as land, sometimesregarded as “artificial” wealth as distinct from the

amidst the decay and dissolution of economic society. Thus, apart altogetherfrom progress and increasing population, a gradually increasing stock of moneyhas proved imperative.

6They are the more suitable for my purpose because Prof. Heckscher ishimself an adherent, on the whole, of the classical theory and much less sympa-thetic to the mercantilist theories than I am. Thus there is no risk that his choiceof quotations has been biased in any way by a desire to illustrate their wisdom.

334

III.

“natural” wealth; interest on capital was the pay-ment for the renting of money similar to rent forland. In so far as mercantilists sought to discoverobjective reasons for the height of the rate of interest—and they did so more and more during this period—they found such reasons in the total quantity of money.From the abundant material available, only the mosttypical examples will be selected, so as to demon-strate first and foremost how lasting this notion was,how deep-rooted and independent of practical con-siderations.

Both of the protagonists in the struggle over mon-etary policy and the East India trade in the early1620’s in England were in entire agreement on thispoint. Gerard Malynes stated, giving detailed rea-son for his assertion, that “Plenty of money decreasethusury in price or rate” (Lex Mercatoria and Mainte-nance of Free Trade, 1622). His truculent and ratherunscrupulous adversary, Edward Misselden, repliedthat “The remedy for Usury may be plenty of money”(Free Trade or the Meanes to make Trade Florish, sameyear). Of the leading writers of half a century later,Child, the omnipotent leader of the East India Com-pany and its most skilful advocate, discussed (1668)the question of how far the legal maximum rate ofinterest, which he emphatically demanded, wouldresult in drawing “the money” of the Dutch awayfrom England. He found a remedy for this dreadeddisadvantage in the easier transference of bills ofdebt, if these were used as currency, for this, he said,“will certainly supply the defect of at least one-halfof all the ready money we have in use in the na-tion”. Petty, the other writer, who was entirely un-affected by the clash of interests, was in agreementwith the rest when he explained the “natural” fall inthe rate of interest from 10 per cent to 6 per cent bythe increase in the amount of money (Political Arith-metick, 1676), and advised lending at interest as an

335


appropriate remedy for a country with too much“Coin” (Quantulumcunque concerning Money, 1682).

This reasoning, naturally enough, was by no meansconfined to England. Several years later (1701 and1706), for example, French merchants and states-men complained of the prevailing scarcity of coin(disette des espèces) as the cause of the high interestrates, and they were anxious to lower the rate ofusury by increasing the circulation of money.7

The great Locke was, perhaps, the first to express in ab-stract terms the relationship between the rate of interest andthe quantity of money in his controversy with Petty.8 He wasopposing Petty’s proposal of a maximum rate of interest onthe ground that it was as impracticable as to fix a maximumrent for land, since “the natural Value of Money, as it is aptto yield such an yearly Income by Interest, depends on thewhole quantity of the then passing Money of the Kingdom,in proportion to the whole Trade of the Kingdom (i.e. thegeneral Vent of all the commodities)”.9 Locke explains thatMoney has two values: (1) its value in use which is given bythe rate of interest “and in this it has the Nature of Land, theIncome of one being called Rent, of the other, Use10”, and (2)its value in exchange “and in this it has the Nature of a Com-modity”, its value in exchange “depending only on the Plentyor Scarcity of Money in proportion to the Plenty or Scarcity ofthose things and not on what Interest shall be”. Thus Lockewas the parent of twin quantity theories. In the first place heheld that the rate of interest depended on the proportion ofthe quantity of money (allowing for the velocity of circula-tion) to the total value of trade. In the second place he held

7Heckscher, Mercantilism, vol. ii. pp. 200, 201, very slightly abridged.8Some Considerations of the Consequences of the Lowering of Interest and Raising

the Value of Money, 1692, but written some years previously.9He adds: “not barely on the quantity of money but the quickness of its

circulation”.10“Use” being, of course, old-fashioned English for “interest”.

336

III.

that the value of money in exchange depended on the propor-tion of the quantity of money to the total volume of goods inthe market. But—standing with one foot in the mercantilistworld and with one foot in the classical world11—he was con-fused concerning the relation between these two proportions,and he overlooked altogether the possibility of fluctuations inliquidity-preference. He was, however, eager to explain thata reduction in the rate of interest has no direct effect on theprice-level and affects prices “only as the Change of Interestin Trade conduces to the bringing in or carrying out Moneyor Commodity, and so in time varying their Proportion herein England from what it was before”, i.e. if the reduction inthe rate of interest leads to the export of cash or an increase inoutput. But he never, I think, proceeds to a genuine synthe-sis.12

How easily the mercantilist mind distinguished betweenthe rate of interest and the marginal efficiency of capital is il-lustrated by a passage (printed in 1621) which Locke quotesfrom A Letter to a Friend concerning Usury: “High Interest de-

11Hume a little later had a foot and a half in the classical world. ForHume began the practice amongst economists of stressing the importance ofthe equilibrium position as compared with the ever-shifting transition towardsit, though he was still enough of a mercantilist not to overlook the fact that it isin the transition that we actually have our being: “It is only in this interval orintermediate situation, between the acquisition of money and a rise of prices,that the increasing quantity of gold and silver is favourable to industry. ... It isof no manner of consequence, with regard to the domestic happiness of a state,whether money be in a greater or less quantity. The good policy of the mag-istrate consists only in keeping it, if possible, still increasing; because by thatmeans he keeps alive a spirit of industry in the nation, and increases the stateof labour in which consists all real power and riches. A nation, whose moneydecreases, is actually, at that time, weaker and more miserable than another na-tion, which possesses no more money but is on the increasing trend.” (Essay OnMoney, 1752).

12It illustrates the completeness with which the mercantilist view, that in-terest means interest on money (the view which is, as it now seems to me,indubitably correct), has dropt out, that Prof. Heckscher, as a good classicaleconomist, sums up his account of Locke’s theory with the comment—“Locke’sargument would be irrefutable ... if interest really were synonymous with theprice for the loan of money; as this is not so, it is entirely irrelevant” (op. cit. vol.ii. p. 204).

337


cays Trade. The advantage from Interest is greater than theProfit from Trade, which makes the rich Merchants give over,and put out their Stock to Interest, and the lesser MerchantsBreak.” Fortrey (England’s Interest and Improvement, 1663) af-fords another example of the stress laid on a low rate of inter-est as a means of increasing wealth.

The mercantilists did not overlook the point that, if an ex-cessive liquidity-preference were to withdraw the influx ofprecious metals into hoards, the advantage to the rate of in-terest would be lost. In some cases (e.g. Mun) the object of en-hancing the power of the State led them, nevertheless, to ad-vocate the accumulation of state treasure. But others franklyopposed this policy:

Schrötter, for instance, employed the usual mer-cantilist arguments in drawing a lurid picture ofhow the circulation in the country would be robbedof all its money through a greatly increasing statetreasury ... he, too, drew a perfectly logical par-allel between the accumulation of treasure by themonasteries and the export surplus of precious met-als, which, to him, was indeed the worst possiblething which he could think of. Davenant explainedthe extreme poverty of many Eastern nations—whowere believed to have more gold and silver thanany other countries in the world—by the fact thattreasure “is suffered to stagnate in the Princes’ Cof-fers”. ... If hoarding by the state was considered,at best, a doubtful boon, and often a great danger,it goes without saying that private hoarding was tobe shunned like the pest. It was one of the tenden-cies against which innumerable mercantilist writersthundered, and I do not think it would be possibleto find a single dissentient voice.13

(b) The mercantilists were aware of the fallacy of cheapnessand the danger that excessive competition may turn the terms

13Heckscher, op. cit. vol. ii. pp. 210, 21I.

338

III.

of trade against a country. Thus Malynes wrote in his LexMereatoria (1622): “Strive not to undersell others to the hurt ofthe Commonwealth, under colour to increase trade: for tradedoth not increase when commodities are good cheap, becausethe cheapness proceedeth of the small request and scarcity ofmoney, which maketh things cheap: so that the contrary aug-menteth trade, when there is plenty of money, and commodi-ties become dearer being in request”.14 Professor Heckschersums up as follows this strand in mercantilist thought:

In the course of a century and a half the stand-point was formulated again and again in this way,that a country with relatively less money than othercountries must “sell cheap and buy dear...”

Even in the original edition of the Discourse of theCommon Weal, that is in the middle of the 16th cen-tury, this attitude was already manifested. Halessaid, in fact, “And yet if strangers should be con-tent to take but our wares for theirs, what should letthem to advance the price of other things (meaning:among others, such as we buy from them), thoughours were good cheap unto them? And then shallwe be still losers, and they at the winning handwith us, while they sell dear and yet buy ours goodcheap, and consequently enrich themselves and im-poverish us. Yet had I rather advance our wares inprice, as they advance theirs, as we now do; thoughsome be losers thereby, and yet not so many as shouldbe the other way.” On this point he had the Unqual-ified approval of his editor several decades later (1581).In the 17th century, this attitude recurred again with-out any fundamental change in significance. Thus,Malynes believed this unfortunate position to be theresult of what he dreaded above all things, i.e. aforeign under-valuation of the English exchange....The same conception then recurred continually. In

14Heckscher, op. cit. vol. ii. p. 228.

339


his Verbum Sapienti (written 1665, published 1691),Petty believed that the violent efforts to increase thequantity of money could only cease “when we havecertainly more money than any of our NeighbourStates (though never so little), both in Arithmeti-cal and Geometrical proportion”. During the pe-riod between the writing and the publication of thiswork, Coke declared, “If our Treasure were morethan our Neighbouring Nations, I did not care whetherwe had one fifth part of the Treasure we now have”(1675).15

(c) The mercantilists were the originals of “the fear of goods”and the scarcity of money as causes of unemployment whichthe classicals were to denounce two centuries later as an ab-surdity:

One of the earliest instances of the applicationof the Unemployment argument as a reason for theprohibition of imports is to be found in Florencein the year 1426.... The English legislation on thematter goes back to at least 1455 .... An almost con-temporary French decree of 1466, forming the basisof the silk industry of Lyons, later to become so fa-mous, was less interesting in so far as it was notactually directed against foreign goods. But it, too,mentioned the possibility of giving work to tens ofthousands of unemployed men and women. It isseen how very much this argument was in the airat the time...

The first great discussion of this matter, as ofnearly all social and economic problems, occurredin England in the middle of the 16th century or ratherearlier, during the reigns of Henry VIII and EdwardVI. In this connection we cannot but mention a se-ries of writings, written apparently at the latest inthe 1530’s, two of which at any rate are believed

15Heckscher, op. cit. vol. ii. p. 235.

340

III.

to have been by Clement Armstrong.... He formu-lates it, for example, in the following terms: “Byreason of great abundance of strange merchandisesand wares brought yearly into England hath notonly caused scarcity of money, but hath destroyedall handicrafts, whereby great number of commonpeople should have works to get money to pay fortheir meat and drink, which of very necessity mustlive idly and beg and steal”

The best instance to my knowledge of a typicallymercantilist discussion of a state of affairs of thiskind is the debates in the English House of Com-mons concerning the scarcity of money, which oc-curred in 1621, when a serious depression had setin, particularly in the cloth export. The conditionswere described very clearly by one of the most in-fluential members of parliament, Sir Edwin Sandys.He stated that the farmer and the artificer had tosuffer almost everywhere; that looms were stand-ing idle for want of money in the country, and thatpeasants were forced to repudiate their contracts,“not (thanks be to God) for want of fruits of theearth, but for want of money”. The situation ledto detailed enquiries into where the money couldhave got to, the want of which was felt so bitterly.Numerous attacks were directed against all personswho were supposed to have contributed either toan export (export surplus) of precious metals, orto their disappearance on account of correspondingactivities within the country.16

Mercantilists were conscious that their policy, as ProfessorHeckscher puts it, “killed two birds with one stone”. “On theone hand the country was rid of an unwelcome surplus ofgoods, which was believed to result in unemployment, whileon the other the total stock of money in the country was in-

16Heckscher, op. cit. vol. ii. P. 223.

341


creased”17 with the resulting advantages of a fall in the rate ofinterest.

It is impossible to study the notions to which the mercan-tilists were led by their actual experiences, without perceiv-ing that there has been a chronic tendency throughout humanhistory for the propensity to save to be stronger than the in-ducement to invest. The weakness of the inducement to investhas been at all times the key to the economic problem. To-day the explanation of the weakness of this inducement maychiefly lie in the extent of existing accumulations; whereas,formerly, risks and hazards of all kinds may have played alarger part. But the result is the same. The desire of the in-dividual to augment his personal wealth by abstaining fromconsumption has usually been stronger than the inducementto the entrepreneur to augment the national wealth by em-ploying labour on the construction of durable assets.

(d) The mercantilists were under no illusions as to the na-tionalistic character of their policies and their tendency to pro-mote war. It was national advantage and relative strength atwhich they were admittedly aiming.18

We may criticise them for the apparent indifference withwhich they accepted this inevitable consequence of an inter-national monetary system. But intellectually their realism ismuch preferable to the confused thinking of contemporaryadvocates of an international fixed gold standard and laissez-faire in international lending, who believe that it is preciselythese policies which will best promote peace.

For in an economy subject to money contracts and customsmore or less fixed over an appreciable period of time, where

17Heckscher, op. cit. vol. ii. p. 178.18“Within the state, mercantilism pursued thoroughgoing dynamic ends.

But the important thing is that this was bound up with a static conception ofthe total economic resources in the world; for this it was that created that fun-damental disharmony which sustained the endless commercial wars.... Thiswas the tragedy of mercantilism. Both the Middle Ages with their universalstatic ideal and laissez-faire with its universal dynamic ideal avoided this conse-quence” (Heckscher, op. cit. vol. ii. pp. 25, 26).

342

III.

the quantity of the domestic circulation and the domestic rateof interest are primarily determined by the balance of pay-ments, as they were in Great Britain before the war, there is noorthodox means open to the authorities for countering unem-ployment at home except by struggling for an export surplusand an import of the monetary metal at the expense of theirneighbours. Never in history was there a method devised ofsuch efficacy for setting each country’s advantage at variancewith its neighbours’ as the international gold (or, formerly, sil-ver) standard. For it made domestic prosperity directly de-pendent on a competitive pursuit of markets and a competi-tive appetite for the precious metals. When by happy accidentthe new supplies of gold and silver were comparatively abun-dant, the struggle might be somewhat abated. But with thegrowth of wealth and the diminishing marginal propensityto consume, it has tended to become increasingly internecine.The part played by orthodox economists, whose common sensehas been insufficient to check their faulty logic, has been dis-astrous to the latest act. For when in their blind struggle for anescape, some countries have thrown off the obligations whichhad previously rendered impossible an autonomous rate ofinterest, these economists have taught that a restoration of theformer shackles is a necessary first step to a general recovery.

In truth the opposite holds good. It is the policy of an au-tonomous rate of interest, unimpeded by international preoc-cupations, and of a national investment programme directedto an optimum level of domestic employment which is twiceblessed in the sense that it helps ourselves and our neighboursat the same time. And it is the simultaneous pursuit of thesepolicies by all countries together which is capable of restor-ing economic health and strength internationally, whether wemeasure it by the level of domestic employment or by the vol-ume of international trade.19

19The consistent appreciation of this truth by the International Labour Of-fice, first under Albert Thomas and subsequently under Mr. H. B. Butler, hasstood out conspicuously amongst the pronouncements of the numerous post-war international bodies.

343


IV

The mercantilists perceived the existence of the problem with-out being able to push their analysis to the point of solvingit. But the classical school ignored the problem, as a conse-quence of introducing into their premisses conditions whichinvolved its non-existence; with the result of creating a cleav-age between the conclusions of economic theory and those ofcommon sense. The extraordinary achievement of the classi-cal theory was to overcome the beliefs of the “natural man”and, at the same time, to be wrong. As Professor Heckscherexpresses it:

If, then, the underlying attitude towards moneyand the material from which money was createddid not alter in the period between the Crusadesand the 18th century, it follows that we are deal-ing with deep-rooted notions. Perhaps the samenotions have persisted even beyond the 500 yearsincluded in that period, even though not nearly tothe same degree as the “fear of goods”. ... With theexception of the period of laissez-faire, no age hasbeen free from these ideas. It was only the uniqueintellectual tenacity of laissez-faire that for a timeovercame the beliefs of the “natural man” on thispoint.20

It required the unqualified faith of doctrinairelaissez-faire to wipe out the “fear of goods” ... [which]is the most natural attitude of the “natural man”in a money economy. Free Trade denied the ex-istence of factors which appeared to be obvious,and was doomed to be discredited in the eyes ofthe man in the street as soon as laissez-faire couldno longer hold the minds of men enchained in itsideology.21

20Heckscher, op. cit. vol. ii. pp. 176-7.21Op. cit. vol. ii. p. 335.

344

V.

I remember Bonar Law’s mingled rage and perplexity inface of the economists, because they were denying what wasobvious. He was deeply troubled for an explanation. One re-curs to the analogy between the sway of the classical schoolof economic theory and that of certain religions. For it is afar greater exercise of the potency of an idea to exorcise theobvious than to introduce into men’s common notions the re-condite and the remote.

V

There remains an allied, but distinct, matter where for cen-turies, indeed for several millenniums, enlightened opinionheld for certain and obvious a doctrine which the classicalschool has repudiated as childish, but which deserves reha-bilitation and honour. I mean the doctrine that the rate of in-terest is not self-adjusting at a level best suited to the socialadvantage but constantly tends to rise too high, so that a wiseGovernment is concerned to curb it by statute and custom andeven by invoking the sanctions of the moral law.

Provisions against usury are amongst the most ancient eco-nomic practices of which we have record. The destruction ofthe inducement to invest by an excessive liquidity-preferencewas the outstanding evil, the prime impediment to the growthof wealth, in the ancient and medieval worlds. And naturallyso, since certain of the risks and hazards of economic life di-minish the marginal efficiency of capital whilst others serveto increase the preference for liquidity. In a world, therefore,which no one reckoned to be safe, it was almost inevitable thatthe rate of interest, unless it was curbed by every instrumentat the disposal of society, would rise too high to permit of anadequate inducement to invest.

I was brought up to believe that the attitude of the Me-dieval Church to the rate of interest was inherently absurd,and that the subtle discussions aimed at distinguishing thereturn on money-loans from the return to active investmentwere merely Jesuitical attempts to find a practical escape from

345


a foolish theory. But I now read these discussions as an hon-est intellectual effort to keep separate what the classical theoryhas inextricably confused together, namely, the rate of interestand the marginal efficiency of capital. For it now seems clearthat the disquisitions of the schoolmen were directed towardsthe elucidation of a formula which should allow the scheduleof the marginal efficiency of capital to be high, whilst usingrule and custom and the moral law to keep down the rate ofinterest.

Even Adam Smith was extremely moderate in his attitudeto the usury laws. For he was well aware that individual sav-ings may be absorbed either by investment or by debts, andthat there is no security that they will find an outlet in theformer. Furthermore, he favoured a low rate of interest as in-creasing the chance of savings finding their outlet in new in-vestment rather than in debts; and for this reason, in a passagefor which he was severely taken to task by Bentham,22 he de-fended a moderate application of the usury laws.23 Moreover,Bentham’s criticisms were mainly on the ground that AdamSmith’s Scotch caution was too severe on “projectors” andthat a maximum rate of interest would leave too little mar-gin for the reward of legitimate and socially advisable risks.For Bentham understood by projectors “all such persons, as, inthe pursuit of wealth, or even of any other object, endeavour,by the assistance of wealth, to strike into any channel of in-vention ... upon all such persons as, in the line of any of theirpursuits, aim at anything that can be called improvement. ... Itfalls, in short, upon every application of the human powers, inwhich ingenuity stands in need of wealth for its assistance.”Of course Bentham is right in protesting against laws whichstand in the way of taking legitimate risks. “A prudent man”,Bentham continues, “will not, in these circumstances, pick outthe good projects from the bad, for he will not meddle withprojects at all.”24

22In his Letter to Adam Smith appended to his Defence of Usury.23Wealth of Nations, Book II, chap. 4.24Having started to quote Bentham in this context, I must remind the reader

346

VI.

It may be doubted, perhaps, whether the above is just whatAdam Smith intended by his term. Or is it that we are hear-ing in Bentham (though writing in March 1787 from “Crichoffin White Russia”) the voice of nineteenth-century Englandspeaking to the eighteenth? For nothing short of the exuber-ance of the greatest age of the inducement to investment couldhave made it possible to lose sight of the theoretical possibilityof its insufficiency.

VI

It is convenient to mention at this point the strange, undulyneglected prophet Silvio Gesell (1862-1930), whose work con-tains flashes of deep insight and who only just failed to reachdown to the essence of the matter. In the post-war years hisdevotees bombarded me with copies of his works; yet, owingto certain palpable defects in the argument, I entirely failedto discover their merit. As is often the case with imperfectlyanalysed intuitions, their significance only became apparentafter I had reached my own conclusions in my own way. Mean-while, like other academic economists, I treated his profoundlyoriginal strivings as being no better than those of a crank.Since few of the readers of this book are likely to be well ac-quainted with the significance of Gesell, I will give to himwhat would be otherwise a disproportionate space.

Gesell was a successful German25 merchant in Buenos Aireswho was led to the study of monetary problems by the cri-sis of the late ’eighties, which was especially violent in theArgentine, his first work, Die Reformation im Münzwesen alsBrücke zum socialen Staat, being published in Buenos Aires in

of his finest passage: “The career of art, the great road which receives the foot-steps of projectors, may be considered as a vast, and perhaps un-bounded,plain, bestrewed with gulphs, such as Curtius was swallowed up in. Each re-quires a human victim to fall into it ere it can close, but when it once closes, itcloses to open no more, and so much of the path is safe to those who follow.”

25Born near the Luxembourg frontier of a German father and a Frenchmother.

347


1891. His fundamental ideas on money were published inBuenos Aires in the same year under the title Nervus rerum,and many books and pamphlets followed until he retired toSwitzerland in 1906 as a man of some means, able to devotethe last decades of his life to the two most delightful occu-pations open to those who do not have to earn their living,authorship and experimental farming.

The first section of his standard work was published in1906 at Les Hauts Geneveys, Switzerland, under the title DieVerwirklichung des Rechtes auf dem vollen Arbeitsertrag, and thesecond section in 1911 at Berlin under the title Die neue Lehrevom Zins. The two together were published in Berlin and inSwitzerland during the war (1916) and reached a sixth editionduring his lifetime under the title Die natürliche Wirtschaftsor-dnung durch Freiland und Freigeld, the English version (trans-lated by Mr. Philip Pye) being called The Natural EconomicOrder. In April 1919 Gesell joined the short-lived Soviet cabi-net of Bavaria as their Minister of Finance, being subsequentlytried by court-martial. The last decade of his life was spent inBerlin and Switzerland and devoted to propaganda. Gesell,drawing to himself the semi-religious fervour which had for-merly centred round Henry George, became the revered prophetof a cult with many thousand disciples throughout the world.The first international convention of the Swiss and GermanFreiland-Freigeld Bund and similar organisations from manycountries was held in Basle in 1923. Since his death in 1930much of the peculiar type of fervour which doctrines such ashis are capable of exciting has been diverted to other (in myopinion less eminent) prophets. Dr. BÃ 1

4chi is the leader of

the movement in England, but its literature seems to be dis-tributed from San Antonio, Texas, its main strength lying to-day in the United States, where Professor Irving Fisher, aloneamongst academic economists, has recognised its significance.

In spite of the prophetic trappings with which his devoteeshave decorated him, Gesell’s main book is written in cool, sci-entific language; though it is suffused throughout by a morepassionate, a more emotional devotion to social justice than

348

VI.

some think decent in a scientist. The part which derives fromHenry George,26 though doubtless an important source of themovement’s strength, is of altogether secondary interest. Thepurpose of the book as a whole may be described as the es-tablishment of an anti-Marxian socialism, a reaction againstlaissez-faire built on theoretical foundations totally unlike thoseof Marx in being based on a repudiation instead of on an ac-ceptance of the classical hypotheses, and on an unfetteringof competition instead of its abolition. I believe that the fu-ture will learn more from the spirit of Gesell than from that ofMarx. The preface to The Natural Economic Order will indicateto the reader, if he will refer to it, the moral quality of Gesell.The answer to Marxism is, I think, to be found along the linesof this preface.

Gesell’s specific contribution to the theory of money andinterest is as follows. In the first place, he distinguishes clearlybetween the rate of interest and the marginal efficiency of cap-ital, and he argues that it is the rate of interest which setsa limit to the rate of growth of real capital. Next, he pointsout that the rate of interest is a purely monetary phenomenonand that the peculiarity of money, from which flows the sig-nificance of the money rate of interest, lies in the fact that itsownership as a means of storing wealth involves the holder innegligible carrying charges, and that forms of wealth, such asstocks of commodities which do involve carrying charges, infact yield a return because of the standard set by money. Hecites the comparative stability of the rate of interest through-out the ages as evidence that it cannot depend on purely phys-ical characters, inasmuch as the variation of the latter fromone epoch to another must have been incalculably greater thanthe observed changes in the rate of interest; i.e. (in my termi-nology) the rate of interest, which depends on constant psy-chological characters, has remained stable, whilst the widelyfluctuating characters, which primarily determine the sched-ule of the marginal efficiency of capital, have determined not

26Gesell differed from George in recommending the payment of compensa-tion when the land is nationalised.

349


the rate of interest but the rate at which the (more or less)given rate of interest allows the stock of real capital to grow.

But there is a great defect in Gesell’s theory. He shows howit is only the existence of a rate of money interest which allowsa yield to be obtained from lending out stocks of commodities.His dialogue between Robinson Crusoe and a stranger27 is amost excellent economic parable—as good as anything of thekind that has been written—to demonstrate this point. But,having given the reason why the money-rate of interest unlikemost commodity rates of interest cannot be negative, he alto-gether overlooks the need of an explanation why the money-rate of interest is positive, and he fails to explain why themoney-rate of interest is not governed (as the classical schoolmaintains) by the standard set by the yield on productive cap-ital. This is because the notion of liquidity-preference had es-caped him. He has constructed only half a theory of the rateof interest.

The incompleteness of his theory is doubtless the expla-nation of his work having suffered neglect at the hands ofthe academic world. Nevertheless he had carried his theoryfar enough to lead him to a practical recommendation, whichmay carry with it the essence of what is needed, though it isnot feasible in the form in which he proposed it. He arguesthat the growth of real capital is held back by the money-rateof interest, and that if this brake were removed the growth ofreal capital would be, in the modern world, so rapid that azero money-rate of interest would probably be justified, notindeed forthwith, but within a comparatively short period oftime. Thus the prime necessity is to reduce the money-rate ofinterest, and this, he pointed out, can be effected by causingmoney to incur carrying-costs just like other stocks of barrengoods. This led him to the famous prescription of “stamped”money, with which his name is chiefly associated and whichhas received the blessing of Professor Irving Fisher. Accord-ing to this proposal currency notes (though it would clearlyneed to apply as well to some forms at least of bank-money)

27The Natural Economic Order, pp. 297 et seq.

350

VII.

would only retain their value by being stamped each month,like an insurance card, with stamps purchased at a post office.The cost of the stamps could, of course, be fixed at any ap-propriate figure. According to my theory it should be roughlyequal to the excess of the money-rate of interest (apart fromthe stamps) over the marginal efficiency of capital correspond-ing to a rate of new investment compatible with full employ-ment. The actual charge suggested by Gesell was 1 per mil.per month, equivalent to 5.4 per cent. per annum. This wouldbe too high in existing conditions, but the correct figure, whichwould have to be changed from time to time, could only bereached by trial and error.

The idea behind stamped money is sound. It is, indeed,possible that means might be found to apply it in practice ona modest scale. But there are many difficulties which Geselldid not face. In particular, he was unaware that money wasnot unique in having a liquidity-premium attached to it, butdiffered only in degree from many other articles, deriving itsimportance from having a greater liquidity-premium than anyother article. Thus if currency notes were to be deprived oftheir liquidity-premium by the stamping system, a long se-ries of substitutes would step into their shoes—bank-money,debts at call, foreign money, jewellery and the precious met-als generally, and so forth. As I have mentioned above, therehave been times when it was probably the craving for theownership of land, independently of its yield, which servedto keep up the rate of interest;—though under Gesell’s systemthis possibility would have been eliminated by land national-isation.

VII

The theories which we have examined above are directed, insubstance, to the constituent of effective demand which de-pends on the sufficiency of the inducement to invest. It is nonew thing, however, to ascribe the evils of unemployment tothe insufficiency of the other constituent, namely, the insuffi-

351


ciency of the propensity to consume. But this alternative ex-planation of the economic evils of the day—equally unpopu-lar with the classical economists—played a much smaller partin sixteenth- and seventeenth-century thinking and has onlygathered force in comparatively recent times.

Though complaints of under-consumption were a very sub-sidiary aspect of mercantilist thought, Professor Heckscherquotes a number of examples of what he calls “the deep-rootedbelief in the utility of luxury and the evil of thrift. Thrift, infact, was regarded as the cause of unemployment, and fortwo reasons: in the first place, because real income was be-lieved to diminish by the amount of money which did not en-ter into exchange, and secondly, because saving was believedto withdraw money from circulation.”28 In 1598 Laffemas (LesTré2sors et richesses pour mettre l’Estat en Splendeur) denouncedthe objectors to the use of French silks on the ground that allpurchasers of French luxury goods created a livelihood for thepoor, whereas the miser caused them to die in distress.29 In1662 Petty justified “entertainments, magnificent shews, tri-umphal arches, etc.”, on the ground that their costs flowedback into the pockets of brewers, bakers, tailors, shoemak-ers and so forth. Fortrey justified “excess of apparel”. VonSchrötter (1686) deprecated sumptuary regulations and de-clared that he would wish that display in clothing and the likewere even greater. Barbon (1690) wrote that “Prodigality is avice that is prejudicial to the Man, but not to trade. ... Cov-etousness is a Vice, prejudicial both to Man and Trade.” 30 In1695 Cary argued that if everybody spent more, all would ob-tain larger incomes “and might then live more plentifully”.31

But it was by Bernard Mandeville’s Fable of the Bees thatBarbon’s opinion was mainly popularised, a book convictedas a nuisance by the grand jury of Middlesex in 1723, whichstands out in the history of the moral sciences for its scan-

28Heckscher, op. cit. vol. ii. p. 208.29Op. cit. vol. ii. p. 290.30Op. cit. vol. ii. p. 209.31Op. cit. vol. ii. p. 291.

352

VII.

dalous reputation. Only one man is recorded as having spo-ken a good word for it, namely Dr. Johnson, who declared thatit did not puzzle him, but “opened his eyes into real life verymuch”. The nature of the book’s wickedness can be best con-veyed by Leslie Stephen’s summary in the Dictionary of Na-tional Biography:

Mandeville gave great offence by this book, inwhich a cynical system of morality was made at-tractive by ingenious paradoxes. ... His doctrinethat prosperity was increased by expenditure ratherthan by saving fell in with many current economicfallacies not yet extinct.32 Assuming with the as-cetics that human desires were essentially evil andtherefore produced “private vices” and assumingwith the common view that wealth was a “publicbenefit”, he easily showed that all civilisation im-plied the development of vicious propensities....

The text of the Fable of the Bees is an allegorical poem—“The Grumbling Hive, or Knaves turned honest”, in whichis set forth the appalling plight of a prosperous community inwhich all the citizens suddenly take it into their heads to aban-don luxurious living, and the State to cut down armaments, inthe interests of Saving:

No Honour now could be content,To live and owe for what was spent,Liv’ries in Broker’s shops are hung;They part with Coaches for a song;Sell stately Horses by whole sets;And Country-Houses to pay debts.Vain cost is shunn’d as moral Fraud;

32In his History of English Thought in the Eighteenth Century Stephen wrote(p. 297) in speaking of “the fallacy made celebrated by Mandeville” that ..thecomplete confutation of it lies in the doctrine—so rarely understood that itscomplete apprehension is, perhaps, the best test of an economist—that demandfor commodities is not demand for labour”.

353


They have no Forces kept Abroad;Laugh at th’ Esteem of Foreigners,And empty Glory got by Wars;They fight, but for their Country’s sake,When Right or Liberty’s at Stake.

The haughty Chloe

Contracts th’ expensive Bill of Fare,And wears her strong Suit a whole Year.

And what is the result?—

Now mind the glorious Hive, and seeHow Honesty and Trade agree:The Shew is gone, it thins apace;And looks with quite another Face,For ’twas not only they that went,By whom vast sums were yearly spent;But Multitudes that lived on them,Were daily forc’d to do the same.In vain to other Trades they’d fly;All were o’er-stocked accordingly.The price of Land and Houses falls;Mirac’lous Palaces whose Walls,Like those of Thebes, were rais’d by Play,Are to be let ...The Building Trade is quite destroy’d,Artificers are not employ’d;No limner for his Art is fam’d,Stone-cutters, Carvers are not nam’d.

So “The Moral” is:

are Virtue can’t make Nations liveIn Splendour. They that would reviveA Golden Age, must be as free,For Acorns as for Honesty.

354

VII.

Two extracts from the commentary which follows the alle-gory will show that the above was not without a theoreticalbasis:

As this prudent economy, which some peoplecall Saving, is in private families the most certainmethod to increase an estate, so some imagine that,whether a country be barren or fruitful, the samemethod if generally pursued (which they think prac-ticable) will have the same effect upon a whole na-tion, and that, for example, the English might bemuch richer than they are, if they would be as fru-gal as some of their neighbours. This, I think, is anerror.33

On the contrary, Mandeville concludes:

The great art to make a nation happy, and whatwe call flourishing, consists in giving everybodyan opportunity of being employed; which to com-pass, let a Government’s first care be to promote asgreat a variety of Manufactures, Arts and Handi-crafts as human wit can invent; and the secondto encourage Agriculture and Fishery in all theirbranches, that the whole Earth may be forced toexert itself as well as Man. It is from this Policyand not from the trifling regulations of Lavishnessand Frugality that the greatness and felicity of Na-tions must be expected; for let the value of Goldand Silver rise or fall, the enjoyment of all Soci-eties will ever depend upon the Fruits of the Earthand the Labour of the People; both which joinedtogether are a more certain, a more inexhaustible

33Compare Adam Smith, the forerunner of the classical school, who wrote,“What is prudence in the conduct of every private family can scarce be folly inthat of a great Kingdom”—probably with reference to the above passage fromMandeville.

355


and a more real Treasure than the Gold of Brazilor the Silver of Potosi.

No wonder that such wicked sentiments called down theopprobrium of two centuries of moralists and economists whofelt much more virtuous in possession of their austere doc-trine that no sound remedy was discoverable except in the ut-most of thrift and economy both by the individual and by thestate. Petty’s “entertainments, magnificent shews, triumphalarches, etc.” gave place to the penny-wisdom of Gladstonianfinance and to a state system which “could not afford” hospi-tals, open spaces, noble buildings, even the preservation of itsancient monuments, far less the splendours of music and thedrama, all of which were consigned to the private charity ormagnanimity of improvident individuals.

The doctrine did not reappear in respectable circles for an-other century, until in the later phase of Malthus the notion ofthe insufficiency of effective demand takes a definite place as ascientific explanation of unemployment. Since I have alreadydealt with this somewhat fully in my essay on Malthus,34 itwill be sufficient if I repeat here one or two characteristic pas-sages which I have already quoted in my essay:

We see in almost every part of the world vastpowers of production which are not put into ac-tion, and I explain this phenomenon by saying thatfrom the want of a proper distribution of the ac-tual produce adequate motives are not furnishedto continued production. ... I distinctly maintainthat an attempt to accumulate very rapidly, whichnecessarily implies a considerable diminution ofunproductive consumption, by greatly impairingthe usual motives to production must prematurelycheck the progress of wealth. ... But if it be truethat an attempt to accumulate very rapidly will

34Essays in Biography, pp. 139-47.

356

VII.

occasion such a division between labour and prof-its as almost to destroy both the motive and thepower of future accumulation and consequentlythe power of maintaining and employing an in-creasing population, must it not be acknowledgedthat such an attempt to accumulate, or that savingtoo much, may be really prejudicial to a country?35

The question is whether this stagnation of cap-ital, and subsequent stagnation in the demand forlabour arising from increased production withoutan adequate proportion of unproductive consump-tion on the part of the landlords and capitalists,could take place without prejudice to the country,without occasioning a less degree both of happi-ness and wealth than would have occurred if theunproductive consumption of the landlords andcapitalists had been so proportioned to the nat-ural surplus of the society as to have continueduninterrupted the motives to production, and pre-vented first an unnatural demand for labour andthen a necessary and sudden diminution of suchdemand. But if this be so, how can it be said withtruth that parsimony, though it may be prejudicialto the producers, cannot be prejudicial to the state;or that an increase of unproductive consumptionamong landlords and capitalists may not some-times be the proper remedy for a state of thingsin which the motives to production fails?36

Adam Smith has stated that capitals are increasedby parsimony, that every frugal man is a publicbenefactor, and that the increase of wealth dependsupon the balance of produce above consumption.That these propositions are true to a great extentis perfectly unquestionable. ... But it is quite ob-

35A letter from Malthus to Ricardo, dated July 7, 1821.36A letter from Malthus to Ricardo, dated July 16, 1822.

357


vious that they are not true to an indefinite ex-tent, and that the principles of saving, pushed toexcess, would destroy the motive to production.If every person were satisfied with the simplestfood, the poorest clothing and the meanest houses,it is certain that no other sort of food, clothing,and lodging would be in existence. ... The two ex-tremes are obvious; and it follows that there mustbe some intermediate point, though the resourcesof political economy may not be able to ascertainit, where, taking into consideration both the powerto produce and the will to consume, the encour-agement to the increase of wealth is the greatest.37

Of all the opinions advanced by able and inge-nious men, which I have ever met with, the opin-ion of M. Say, which states that: un product con-sommé ou détruit est un débouché fermé (I. i. ch. 15),appears to me to be the most directly opposed tojust theory, and the most uniformly contradictedby experience. Yet it directly follows from the newdoctrine, that commodities are to be consideredonly in their relation to each other,—not to the con-sumers. What, I would ask, would become of thedemand for commodities, if all consumption ex-cept bread and water were suspended for the nexthalf-year? What an accumulation of commodities!Quels débouchés! What a prodigious market wouldthis event occasion!38

Ricardo, however, was stone-deaf to what Malthus wassaying. The last echo of the controversy is to be found in JohnStuart Mill’s discussion of his Wages-Fund Theory,39 which in

37Preface to Malthus’s Principles of Political Economy, pp. 8, 9.38Malthus’s Principles of Political Economy, p. 363, footnote.39J. S. Mill, Political Economy, Book I. chapter v. Them is a most important

and penetrating discussion of this aspect of Mill’s theory in Mummery andHobson’s Physiology of Industry? pp. 38 et seq., and, in particular, of his doc-

358

VII.

his own mind played a vital part in his rejection of the laterphase of Malthus, amidst the discussions of which he had, ofcourse, been brought up. Mill’s successors rejected his Wages-Fund Theory but overlooked the fact that Mill’s refutation ofMalthus depended on it. Their method was to dismiss theproblem from the corpus of Economics not by solving it but bynot mentioning it. It altogether disappeared from controversy.Mr. Cairncross, searching recently for traces of it amongst theminor Victorians, 40 has found even less, perhaps, than mighthave been expected. 41 Theories of under-consumption hiber-nated until the appearance in 1889 of The Physiology of Indus-try, by J. A. Hobson and A. F. Mummery, the first and mostsignificant of many volumes in which for nearly fifty yearsMr. Hobson has flung himself with unflagging, but almost un-availing, ardour and courage against the ranks of orthodoxy.Though it is so completely forgotten to-day, the publication ofthis book marks, in a sense, an epoch in economic thought.42

The Physiology of Industry was written in collaboration withA. F. Mummery. Mr. Hobson has told how the book came tobe written as follows: 43

It was not until the middle ’eighties that myeconomic heterodoxy began to take shape. Thoughthe Henry George campaign against land valuesand the early agitation of various socialist groupsagainst the visible oppression of the working classes,

trine (which Marshall, in his very unsatisfactory discussion of the Wages-FundTheory, endeavoured to explain away) that “a demand for commodities is not ademand for labour”.

40“The Victorians and Investment”, Economic History, 1936.41Fullarton’s tract On the Regulation of Currencies (1844) is the most interest-

ing of his references.42J. M. Robertson’s The Fallacy of Saving, published in I892, supported the

heresy of Mummery and Hobson. But it is not a book of much value or signif-icance, being entirety lacking in the penetrating intuitions of The Physiology ofIndustry.

43In an address called “Confessions of an Economic Heretic”, delivered be-fore the London Ethical Society at Conway Hall on Sunday, July 14, 1935. Ireproduce it here by Mr. Hobson’s permission.

359


coupled with the revelations of the two Booths re-garding the poverty of London, made a deep im-pression on my feelings, they did not destroy myfaith in Political Economy. That came from whatmay be called an accidental contact. While teach-ing at a school in Exeter I came into personal rela-tions with a business man named Mummery, knownthen and afterwards as a great mountaineer whohad discovered another way up the Matterhornand who, in 1895, was killed in an attempt to climbthe famous Himalayan mountain Nanga Parbat.My intercourse with him, I need hardly say didnot lie on this physical plane. But he was a mentalclimber as well, with a natural eye for a path of hisown finding and a sublime disregard of intellec-tual authority. This man entangled me in a contro-versy about excessive saving, which he regardedas responsible for the under-employment of capi-tal and labour in periods of bad trade. For a longtime I sought to counter his arguments by the useof the orthodox economic weapons. But at lengthhe convinced me and I went in with him to elab-orate the over-saving argument in a book entitledThe Physiology of Industry, which was published in1889. This was the first open step in my hereti-cal career, and I did not in the least realise its mo-mentous consequences. For just at that time I hadgiven up my scholastic post and was opening anew line of work as University Extension Lecturerin Economics and Literature. The first shock camein a refusal of the London Extension Board to al-low me to offer courses of Political Economy. Thiswas due, I learned, to the intervention of an Eco-nomic Professor who had read my book and con-sidered it as equivalent in rationality to an attemptto prove the flatness of the earth. How could therebe any limit to the amount of useful saving whenevery item of saving went to increase the capital

360

VII.

structure and the fund for paying wages? Soundeconomists could not fail to view with horror anargument which sought to check the source of allindustrial progress.44 Another interesting personalexperience helped to bring home to me the senseof my iniquity. Though prevented from lectur-ing on economics in London, I had been allowedby the greater liberality of the Oxford UniversityExtension Movement to address audiences in theProvinces, confining myself to practical issues re-lating to working-class life. Now it happened atthis time that the Charity Organisation Society wasplanning a lecture campaign upon economic sub-jects and invited me to prepare a course. I had ex-pressed m willingness to undertake this new lec-ture work, when suddenly, without explanation,the invitation was withdrawn. Even then I hardlyrealised that in appearing to question the virtueof unlimited thrift I had committed the unpardon-able sin.

In this early work Mr. Hobson with his collaborator ex-pressed himself with more direct reference to the classical eco-nomics (in which he had been brought up) than in his laterwritings; and for this reason, as well as because it is the firstexpression of his theory, I will quote from it to show how sig-nificant and well-founded were the authors’ criticisms and in-tuitions. They point out in their preface as follows the natureof the conclusions which they attack:

Saving enriches and spending impoverishes thecommunity along with the individual, and it may

44Hobson had written disrespectfully in The Physiology of Industry, p. 26:“Thrift is the source of national wealth, and the more thrifty a nation is the morewealthy it becomes. Such is the common teaching of almost all economists;many of them assume a tone of ethical dignity as they plead the infinite valueof thrift; this note alone in all their dreary song has caught the favour of thepublic ear.”

361


be generally defined as an assertion that the effec-tive love of money is the root of all economic good.Not merely does it enrich the thrifty individualhimself, but it raises wages, gives work to the un-employed, and scatters blessings on every side.From the daily papers to the latest economic trea-tise, from the pulpit to the House of Commons,this conclusion is reiterated and re-stated till it ap-pears positively impious to question it. Yet the ed-ucated world, supported by the majority of eco-nomic thinkers, up to the publication of Ricardo’swork strenuously denied this doctrine, and its ul-timate acceptance was exclusively due to their in-ability to meet the now exploded wages-fund doc-trine. That the conclusion should have survivedthe argument on which it logically stood, can beexplained on no other hypothesis than the com-manding authority of the great men who assertedit. Economic critics have ventured to attack thetheory in detail, but they have shrunk appalledfrom touching its main conclusions. Our purposeis to show that these conclusions are not tenable,that an undue exercise of the habit of saving ispossible, and that such undue exercise impover-ishes the Community, throws labourers out of work,drives down wages, and spreads that gloom andprostration through the commercial world whichis known as Depression in Trade. ...

The object of production is to provide “utilitiesand conveniences” for consumers, and the pro-cess is a continuous one from the first handling ofthe raw material to the moment when it is finallyconsumed as a utility or a convenience. The onlyuse of Capital being to aid the production of theseutilities and conveniences, the total used will nec-essarily vary with the total of utilities and conve-niences daily or weekly consumed. Now saving,while it increases the existing aggregate of Capi-

362

VII.

tal, simultaneously reduces the quantity of utili-ties and conveniences consumed; any undue ex-ercise of this habit must, therefore, cause an ac-cumulation of Capital in excess of that which isrequired for use, and this excess will exist in theform of general over-production.45

In the last sentence of this passage there appears the rootof Hobson’s mistake, namely, his supposing that it is a case ofexcessive saving causing the actual accumulation of capital inexcess of what is required, which is, in fact, a secondary evilwhich only occurs through mistakes of foresight; whereas theprimary evil is a propensity to save in conditions of full em-ployment more than the equivalent of the capital which is re-quired, thus preventing full employment except when there isa mistake of foresight. A page or two later, however, he putsone half of the matter, as it seems to me, with absolute pre-cision, though still overlooking the possible role of changesin the rate of interest and in the state of business confidence,factors which he presumably takes as given:

We are thus brought to the conclusion that thebasis on which all economic teaching since AdamSmith has stood, viz. that the quantity annuallyproduced is determined by the aggregates of Nat-ural Agents, Capital, and Labour available, is er-roneous, and that, on the contrary, the quantityproduced, while it can never exceed the limits im-posed by these aggregates, may be, and actuallyis, reduced far below this maximum by the checkthat undue saving and the consequent accumula-tion of over-supply exerts on production; i.e. thatin the normal state of modern industrial Commu-nities, consumption limits production and not pro-duction consumption.46

45Hobson and Mummery, Physiology of Industry, pp. iii-v.46Hobson and Mummery, Physiology of Industry, p. vi.

363


Finally he notices the bearing of his theory on the validityof the orthodox Free Trade arguments:

We also note that the charge of commercial im-becility, so freely launched by orthodox economistsagainst our American cousins and other Protec-tionist Communities, can no longer be maintainedby any of the Free Trade arguments hitherto ad-duced, since all these are based on the assumptionthat over-supply is impossible.47

The subsequent argument is, admittedly, incomplete. But itis the first explicit statement of the fact that capital is broughtinto existence not by the propensity to save but in response tothe demand resulting from actual and prospective consump-tion. The following portmanteau quotation indicates the lineof thought:

It should be clear that the capital of a com-munity cannot be advantageously increased with-out a subsequent increase in consumption of com-modities. ... Every increase in saving and in capi-tal requires, in order to be effectual, a correspond-ing increase in immediately future consumption.48

... And when we say future consumption, we donot refer to a future of ten, twenty, or fifty yearshence, but to a future that is but little removed hepresent. ... If increased thrift or caution inducespeople to save more in the present, they must con-sent to consume more in the future.49 ... No morecapital can economically exist at any point in theproductive process than is required to furnish com-modities for the current rate of consumption.50 ...

47Op. cit. p. ix.48Op. cit. p. 2749Op. cit. pp. 50, 5150Op. cit. p. 69

364

VII.

It is clear that my thrift in no wise affects the to-tal economic thrift of the community, but only de-termines whether a particular portion of the totalthrift shall have been exercised by myself or bysomebody else. We shall show how the thrift ofone part of the community has power to force an-other part to live beyond their income.51 ... Mostmodern economists deny that consumption couldby any possibility be insufficient. Can we find anyeconomic force at work which might incite a com-munity to this excess, and if there be any suchforces are there not efficient checks provided bythe mechanism of commerce? It will be shown,firstly, that in every highly organised industrialsociety there is constantly at work a force whichnaturally operates to induce excess of thrift; sec-ondly, that the checks alleged to be provided bythe mechanism of commerce are either wholly in-operative or are inadequate to prevent grave com-mercial evil.52 ... The brief answer which Ricardogave to the contentions of Malthus and Chalmersseems to have been accepted as sufficient by mostlater economists. “Productions are always boughtby productions or by services; money is only themedium by which the exchange is effected. Hencethe increased production being always accompa-nied by a correspondingly increased ability to getand consume, there is no possibility of Overpro-duction” (Ricardo, Prin. of Pol. Econ. p. 362).53

Hobson and Mummery were aware that interest was noth-ing whatever except payment for the use of money.54 Theyalso knew well enough that their opponents would claim that

51Op. cit. p. 11352Op. cit. p. 10053Op. cit. p. 10154Op. cit. p. 79

365


there would be “such a fall in the rate of interest (or profit)as will act as a check upon Saving, and restore the proper re-lation between production and consumption”.55 They pointout in reply that “if a fall of Profit is to induce people to saveless, it must operate in one of two ways, either by inducingthem to spend more or by inducing them to produce less”.56

As regards the former they argue that when profits fall theaggregate income of the community is reduced, and “we can-not suppose that when the average rate of incomes is falling,individuals will be induced to increase their rate of consump-tion by the fact that the premium upon thrift is correspond-ingly diminished”; whilst as for the second alternative, “it isso far from being our intention to deny that a fall of profit,due to over-supply, will check production, that the admissionof the operation of this check forms the very centre of our ar-gument”.57 Nevertheless, their theory failed of completeness,essentially on account of their having no independent theoryof the rate of interest; with the result that Mr. Hobson laidtoo much emphasis (especially in his later books) on under-consumption leading to over-investment, in the sense of un-profitable investment, instead of explaining that a relativelyweak propensity to consume helps to cause unemploymentby requiring and not receiving the accompaniment of a com-pensating volume of new investment, which, even if it maysometimes occur temporarily through errors of optimism, isin general prevented from happening at all by the prospectiveprofit falling below the standard set by the rate of interest.

Since the war there has been a spate of heretical theories ofunder-consumption, of which those of Major Douglas are themost famous. The strength of Major Douglas’s advocacy has,of course, largely depended on orthodoxy having no valid re-ply to much of his destructive criticism. On the other hand,

55Op. cit. p. 117.56Op. cit. P. 130.57Hobson and Mummery, Physiology of Industry, p. 131.1. Vide his Industry

and Trade, Appendix D; Money, Credit and Commerce, p. 130; and Principles ofEconomics, Appendix I.

366

VII.

the detail of his diagnosis, in particular the so-called A + Btheorem, includes much mere mystification. If Major Douglashad limited his B-items to the financial provisions made by en-trepreneurs to which no current expenditure on replacementsand renewals corresponds, he would be nearer the truth. Buteven in that case it is necessary to allow for the possibilityof these provisions being offset by new investment in otherdirections as well as by increased expenditure on consump-tion. Major Douglas is entitled to claim, as against some ofhis orthodox adversaries, that he at least has not been whollyoblivious of the outstanding problem of our economic system.Yet he has scarcely established an equal claim to rank—a pri-vate, perhaps, but not a major in the brave army of heretics—with Mandeville, Malthus, Gesell and Hobson, who, follow-ing their intuitions, have preferred to see the truth obscurelyand imperfectly rather than to maintain error, reached indeedwith clearness and consistency and by easy logic, but on hy-potheses inappropriate to the facts.

367

CHAPTER 24

CONCLUDING NOTES ON THE SOCIALPHILOSOPHY TOWARDS WHICH THE GENERAL

THEORY MIGHT LEAD

I

The outstanding faults of the economic society in which welive are its failure to provide for full employment and its arbi-trary and inequitable distribution of wealth and incomes. Thebearing of the foregoing theory on the first of these is obvi-ous. But there are also two important respects in which it isrelevant to the second.

Since the end of the nineteenth century significant progresstowards the removal of very great disparities of wealth and in-come has been achieved through the instrument of direct tax-ation — income tax and surtax and death duties — especiallyin Great Britain. Many people would wish to see this processcarried much further, but they are deterred by two considera-tions; partly by the fear of making skilful evasions too muchworth while and also of diminishing unduly the motive to-wards risk-taking, but mainly, I think, by the belief that thegrowth of capital depends upon the strength of the motivetowards individual saving and that for a large proportion ofthis growth we are dependent on the savings of the rich outof their superfluity. Our argument does not affect the first ofthese considerations. But it may considerably modify our at-titude towards the second. For we have seen that, up to thepoint where full employment prevails, the growth of capitaldepends not at all on a low propensity to consume but is, onthe contrary, held back by it; and only in conditions of fullemployment is a low propensity to consume conducive to thegrowth of capital. Moreover, experience suggests that in ex-

369

24. CONCLUDING NOTES ON THE SOCIAL PHILOSOPHY TOWARDSWHICH THE GENERAL THEORY MIGHT LEAD

isting conditions saving by institutions and through sinkingfunds is more than adequate, and that measures for the re-distribution of incomes in a way likely to raise the propensityto consume may prove positively favourable to the growth ofcapital.

The existing confusion of the public mind on the matter iswell illustrated by the very common belief that the death du-ties are responsible for a reduction in the capital wealth of thecountry. Assuming that the State applies the proceeds of theseduties to its ordinary outgoings so that taxes on incomes andconsumption are correspondingly reduced or avoided, it is, ofcourse, true that a fiscal policy of heavy death duties has theeffect of increasing the community’s propensity to consume.But inasmuch as an increase in the habitual propensity to con-sume will in general (i.e. except in conditions of full employ-ment) serve to increase at the same time the inducement toinvest, the inference commonly drawn is the exact opposite ofthe truth.

Thus our argument leads towards the conclusion that incontemporary conditions the growth of wealth, so far frombeing dependent on the abstinence of the rich, as is commonlysupposed, is more likely to be impeded by it. One of the chiefsocial justifications of great inequality of wealth is, therefore,removed. I am not saying that there are no other reasons, un-affected by our theory, capable of justifying some measure ofinequality in some circumstances. But it does dispose of themost important of the reasons why hitherto we have thoughtit prudent to move carefully. This particularly affects our atti-tude towards death duties: for there are certain justificationsfor inequality of incomes which do not apply equally to in-equality of inheritances.

For my own part, I believe that there is social and psy-chological justification for significant inequalities of incomesand wealth, but not for such large disparities as exist today.There are valuable human activities which require the mo-tive of money-making and the environment of private wealth-ownership for their full fruition. Moreover, dangerous hu-man proclivities can be canalised into comparatively harmless

370

II.

channels by the existence of opportunities for money-makingand private wealth, which, if they cannot be satisfied in thisway, may find their outlet in cruelty, the reckless pursuit ofpersonal power and authority, and other forms of self-aggrandisement.It is better that a man should tyrannise over his bank balancethan over his fellow-citizens; and whilst the former is some-times denounced as being but a means to the latter, sometimesat least it is an alternative. But it is not necessary for the stim-ulation of these activities and the satisfaction of these procliv-ities that the game should be played for such high stakes asat present. Much lower stakes will serve the purpose equallywell, as soon as the players are accustomed to them. The taskof transmuting human nature must not be confused with thetask of managing it. Though in the ideal commonwealth menmay have been taught or inspired or bred to take no interest inthe stakes, it may still be wise and prudent statesmanship toallow the game to be played, subject to rules and limitations,so long as the average man, or even a significant section of thecommunity, is in fact strongly addicted to the money-makingpassion.

II

There is, however, a second, much more fundamental infer-ence from our argument which has a bearing on the future ofinequalities of wealth; namely, our theory of the rate of inter-est. The justification for a moderately high rate of interest hasbeen found hitherto in the necessity of providing a sufficientinducement to save. But we have shown that the extent of ef-fective saving is necessarily determined by the scale of invest-ment and that the scale of investment is promoted by a lowrate of interest, provided that we do not attempt to stimulateit in this way beyond the point which corresponds to full em-ployment. Thus it is to our best advantage to reduce the rate ofinterest to that point relatively to the schedule of the marginalefficiency of capital at which there is full employment.

There can be no doubt that this criterion will lead to a much

371


lower rate of interest than has ruled hitherto; and, so far as onecan guess at the schedules of the marginal efficiency of capi-tal corresponding to increasing amounts of capital, the rate ofinterest is likely to fall steadily, if it should be practicable tomaintain conditions of more or less continuous full employ-ment unless, indeed, there is an excessive change in the ag-gregate propensity to consume (including the State).

I feel sure that the demand for capital is strictly limited inthe sense that it would not be difficult to increase the stock ofcapital up to a point where its marginal efficiency had fallento a very low figure. This would not mean that the use of cap-ital instruments would cost almost nothing, but only that thereturn from them would have to cover little more than theirexhaustion by wastage and obsolescence together with somemargin to cover risk and the exercise of skill and judgment. Inshort, the aggregate return from durable goods in the courseof their life would, as in the case of short-lived goods, justcover their labour costs of production plus an allowance forrisk and the costs of skill and supervision.

Now, though this state of affairs would be quite compatiblewith some measure of individualism, yet it would mean theeuthanasia of the rentier, and, consequently, the euthanasia ofthe cumulative oppressive power of the capitalist to exploitthe scarcity-value of capital. Interest today rewards no gen-uine sacrifice, any more than does the rent of land. The ownerof capital can obtain interest because capital is scarce, just asthe owner of land can obtain rent because land is scarce. Butwhilst there may be intrinsic reasons for the scarcity of land,there are no intrinsic reasons for the scarcity of capital. Anintrinsic reason for such scarcity, in the sense of a genuine sac-rifice which could only be called forth by the offer of a rewardin the shape of interest, would not exist, in the long run, exceptin the event of the individual propensity to consume provingto be of such a character that net saving in conditions of fullemployment comes to an end before capital has become suf-ficiently abundant. But even so, it will still be possible forcommunal saving through the agency of the State to be main-tained at a level which will allow the growth of capital up to

372

II.

the point where it ceases to be scarce.I see, therefore, the rentier aspect of capitalism as a transi-

tional phase which will disappear when it has done its work.And with the disappearance of its rentier aspect much else init besides will suffer a sea-change. It will be, moreover, a greatadvantage of the order of events which I am advocating, thatthe euthanasia of the rentier, of the functionless investor, willbe nothing sudden, merely a gradual but prolonged continu-ance of what we have seen recently in Great Britain, and willneed no revolution.

Thus we might aim in practice (there being nothing in thiswhich is unattainable) at an increase in the volume of capitaluntil it ceases to be scarce, so that the functionless investorwill no longer receive a bonus; and at a scheme of direct taxa-tion which allows the intelligence and determination and ex-ecutive skill of the financier, the entrepreneur et hoc genus omne(who are certainly so fond of their craft that their labour couldbe obtained much cheaper than at present), to be harnessed tothe service of the community on reasonable terms of reward.

At the same time we must recognise that only experiencecan show how far the common will, embodied in the policyof the State, ought to be directed to increasing and supple-menting the inducement to invest; and how far it is safe tostimulate the average propensity to consume, without forego-ing our aim of depriving capital of its scarcity-value withinone or two generations. It may turn out that the propensityto consume will be so easily strengthened by the effects of afalling rate of interest, that full employment can be reachedwith a rate of accumulation little greater than at present. Inthis event a scheme for the higher taxation of large incomesand inheritances might be open to the objection that it wouldlead to full employment with a rate of accumulation whichwas reduced considerably below the current level. I must notbe supposed to deny the possibility, or even the probability, ofthis outcome. For in such matters it is rash to predict how theaverage man will react to a changed environment. If, how-ever, it should prove easy to secure an approximation to fullemployment with a rate of accumulation not much greater

373


than at present, an outstanding problem will at least havebeen solved. And it would remain for separate decision onwhat scale and by what means it is right and reasonable tocall on the living generation to restrict their consumption, soas to establish in course of time, a state of full investment fortheir successors.

III

In some other respects the foregoing theory is moderately con-servative in its implications. For whilst it indicates the vitalimportance of establishing certain central controls in matterswhich are now left in the main to individual initiative, thereare wide fields of activity which are unaffected. The State willhave to exercise a guiding influence on the propensity to con-sume partly through its scheme of taxation, partly by fixingthe rate of interest, and partly, perhaps, in other ways. Fur-thermore, it seems unlikely that the influence of banking pol-icy on the rate of interest will be sufficient by itself to deter-mine an optimum rate of investment. I conceive, therefore,that a somewhat comprehensive socialisation of investmentwill prove the only means of securing an approximation tofull employment; though this need not exclude all manner ofcompromises and of devices by which public authority willco-operate with private initiative. But beyond this no obvi-ous case is made out for a system of State Socialism whichwould embrace most of the economic life of the community. Itis not the ownership of the instruments of production whichit is important for the State to assume. If the State is able todetermine the aggregate amount of resources devoted to aug-menting the instruments and the basic rate of reward to thosewho own them, it will have accomplished all that is neces-sary. Moreover, the necessary measures of socialisation canbe introduced gradually and without a break in the generaltraditions of society.

Our criticism of the accepted classical theory of economicshas consisted not so much in finding logical flaws in its anal-

374

III.

ysis as in pointing out that its tacit assumptions are seldomor never satisfied, with the result that it cannot solve the eco-nomic problems of the actual world. But if our central con-trols succeed in establishing an aggregate volume of outputcorresponding to full employment as nearly as is practicable,the classical theory comes into its own again from this pointonwards. If we suppose the volume of output to be given,i.e. to be determined by forces outside the classical scheme ofthought, then there is no objection to be raised against the clas-sical analysis of the manner in which private self-interest willdetermine what in particular is produced, in what proportionsthe factors of production will be combined to produce it, andhow the value of the final product will be distributed betweenthem. Again, if we have dealt otherwise with the problem ofthrift, there is no objection to be raised against the modernclassical theory as to the degree of consilience between pri-vate and public advantage in conditions of perfect and imper-fect competition respectively. Thus, apart from the necessityof central controls to bring about an adjustment between thepropensity to consume and the inducement to invest, there isno more reason to socialise economic life than there was be-fore.

To put the point concretely, I see no reason to suppose thatthe existing system seriously misemploys the factors of pro-duction which are in use. There are, of course, errors of fore-sight; but these would not be avoided by centralising deci-sions. When 9,000,000 men are employed out of 10,000,000willing and able to work, there is no evidence that the labourof these 9,000,000 men is misdirected. The complaint againstthe present system is not that these 9,000,000 men ought to beemployed on different tasks, but that tasks should be availablefor the remaining 1,000,000 men. It is in determining the vol-ume, not the direction, of actual employment that the existingsystem has broken down.

Thus I agree with Gesell that the result of filling in the gapsin the classical theory is not to dispose of the ‘Manchester Sys-tem’, but to indicate the nature of the environment which thefree play of economic forces requires if it is to realise the full

375


potentialities of production. The central controls necessary toensure full employment will, of course, involve a large exten-sion of the traditional functions of government. Furthermore,the modern classical theory has itself called attention to vari-ous conditions in which the free play of economic forces mayneed to be curbed or guided. But there will still remain a widefield for the exercise of private initiative and responsibility.Within this field the traditional advantages of individualismwill still hold good.

Let us stop for a moment to remind ourselves what theseadvantages are. They are partly advantages of efficiency —the advantages of decentralisation and of the play of self-interest.The advantage to efficiency of the decentralisation of deci-sions and of individual responsibility is even greater, perhaps,than the nineteenth century supposed; and the reaction againstthe appeal to self-interest may have gone too far. But, aboveall, individualism, if it can be purged of its defects and itsabuses, is the best safeguard of personal liberty in the sensethat, compared with any other system, it greatly widens thefield for the exercise of personal choice. It is also the best safe-guard of the variety of life, which emerges precisely from thisextended field of personal choice, and the loss of which is thegreatest of all the losses of the homogeneous or totalitarianstate. For this variety preserves the traditions which embodythe most secure and successful choices of former generations;it colours the present with the diversification of its fancy; and,being the handmaid of experiment as well as of tradition andof fancy, it is the most powerful instrument to better the fu-ture.

Whilst, therefore, the enlargement of the functions of gov-ernment, involved in the task of adjusting to one another thepropensity to consume and the inducement to invest, wouldseem to a nineteenth-century publicist or to a contemporaryAmerican financier to be a terrific encroachment on individu-alism. I defend it, on the contrary, both as the only practicablemeans of avoiding the destruction of existing economic formsin their entirety and as the condition of the successful func-tioning of individual initiative.

376

IV.

For if effective demand is deficient, not only is the pub-lic scandal of wasted resources intolerable, but the individualenterpriser who seeks to bring these resources into action isoperating with the odds loaded against him. The game of haz-ard which he plays is furnished with many zeros, so that theplayers as a whole will lose if they have the energy and hopeto deal all the cards. Hitherto the increment of the world’swealth has fallen short of the aggregate of positive individualsavings; and the difference has been made up by the lossesof those whose courage and initiative have not been supple-mented by exceptional skill or unusual good fortune. But ifeffective demand is adequate, average skill and average goodfortune will be enough.

The authoritarian state systems of today seem to solve theproblem of unemployment at the expense of efficiency and offreedom. It is certain that the world will not much longer tol-erate the unemployment which, apart from brief intervals ofexcitement, is associated and in my opinion, inevitably asso-ciated with present-day capitalistic individualism. But it maybe possible by a right analysis of the problem to cure the dis-ease whilst preserving efficiency and freedom.

IV

I have mentioned in passing that the new system might bemore favourable to peace than the old has been. It is worthwhile to repeat and emphasise that aspect.

War has several causes. Dictators and others such, to whomwar offers, in expectation at least, a pleasurable excitement,find it easy to work on the natural bellicosity of their peoples.But, over and above this, facilitating their task of fanning thepopular flame, are the economic causes of war, namely, thepressure of population and the competitive struggle for mar-kets. It is the second factor, which probably played a predom-inant part in the nineteenth century, and might again, that isgermane to this discussion.

I have pointed out in the preceding chapter that, under the

377


system of domestic laissez-faire and an international gold stan-dard such as was orthodox in the latter half of the nineteenthcentury, there was no means open to a government wherebyto mitigate economic distress at home except through the com-petitive struggle for markets. For all measures helpful to astate of chronic or intermittent under-employment were ruledout, except measures to improve the balance of trade on in-come account.

Thus, whilst economists were accustomed to applaud theprevailing international system as furnishing the fruits of theinternational division of labour and harmonising at the sametime the interests of different nations, there lay concealed aless benign influence; and those statesmen were moved bycommon sense and a correct apprehension of the true courseof events, who believed that if a rich, old country were to ne-glect the struggle for markets its prosperity would droop andfail. But if nations can learn to provide themselves with fullemployment by their domestic policy (and, we must add, ifthey can also attain equilibrium in the trend of their popula-tion), there need be no important economic forces calculatedto set the interest of one country against that of its neigh-bours. There would still be room for the international divi-sion of labour and for international lending in appropriateconditions. But there would no longer be a pressing motivewhy one country need force its wares on another or repulsethe offerings of its neighbour, not because this was necessaryto enable it to pay for what it wished to purchase, but withthe express object of upsetting the equilibrium of paymentsso as to develop a balance of trade in its own favour. Interna-tional trade would cease to be what it is, namely, a desperateexpedient to maintain employment at home by forcing saleson foreign markets and restricting purchases, which, if suc-cessful, will merely shift the problem of unemployment to theneighbour which is worsted in the struggle, but a willing andunimpeded exchange of goods and services in conditions ofmutual advantage.

378

V.

V

Is the fulfilment of these ideas a visionary hope? Have theyinsufficient roots in the motives which govern the evolutionof political society? Are the interests which they will thwartstronger and more obvious than those which they will serve?

I do not attempt an answer in this place. It would needa volume of a different character from this one to indicateeven in outline the practical measures in which they mightbe gradually clothed. But if the ideas are correct — an hy-pothesis on which the author himself must necessarily basewhat he writes — it would be a mistake, I predict, to disputetheir potency over a period of time. At the present momentpeople are unusually expectant of a more fundamental diag-nosis; more particularly ready to receive it; eager to try it out,if it should be even plausible. But apart from this contem-porary mood, the ideas of economists and political philoso-phers, both when they are right and when they are wrong,are more powerful than is commonly understood. Indeed theworld is ruled by little else. Practical men, who believe them-selves to be quite exempt from any intellectual influences, areusually the slaves of some defunct economist. Madmen in au-thority, who hear voices in the air, are distilling their frenzyfrom some academic scribbler of a few years back. I am surethat the power of vested interests is vastly exaggerated com-pared with the gradual encroachment of ideas. Not, indeed,immediately, but after a certain interval; for in the field of eco-nomic and political philosophy there are not many who areinfluenced by new theories after they are twenty-five or thirtyyears of age, so that the ideas which civil servants and politi-cians and even agitators apply to current events are not likelyto be the newest. But, soon or late, it is ideas, not vested inter-ests, which are dangerous for good or evil.

379

Documents

The General Theory of Employment Interest and Money by John Maynard Keynes