hayek, f. (1937) 'monetary nationalism and international stability

First edition 1937(London: Longmans, Green- The Graduate Instituteof International Studies, Geneva, PublicationNumber 18,1937)

Reprinted 1989 byAUGUSTUS M. KELLEY, PUBLISHERSFairfield NJ 07006-0008

Library of Congress Cataloging-in-Publication Data

Hayek, Friedrich A. von (Friedrich August), 1899-Monetary nationalism and international stability.

(Reprints of economic classics)Reprint. Originally published: 1st ed. London :

Longmans, Green, 1937.1. Money. 2. Currency question. 3. International

finance. I. Title. II. Series.HG221.H345 1989 332.4 87-17244ISBN 0-678-00047-6

Manufactured in the United States of America

CONTENTS

LECTURE I. National Monetary Systems 11. Theoretical character of these lectures, p. 1;

2. Monetary Nationalism and national monetary sys-tems, p. 4; 3. A homogeneous international currency,p. 5; 4. The mixed or national reserve system, p. 8;5. Independent national currencies, p. 14.

LECTURE II. The Function and Mechanism of InternationalFlows of Money 17

1. The functions of redistributions of the world'sstock of money, p. 17; 2. The mechanism under a sys-tem of purely metallic currencies, p. 19; 3. The mecha-nism under a regime of mixed currencies, p. 25; 4. Therile of central bank policy, p. 32.

LECTURE III. Independent Currencies 361. National stabilisation and international shifts of

demand, p. 35; 2. The position in the country adver-sely affected, p. 38; 3. The position in the countryfavourably affected, p. 41; 4, Causes of the recentgrowth of Monetary Nationalism, p. 43; 5. The signi-ficance of the concepts of inflation and deflation appliedto a national area, p. 47.

LECTURE IV. International Capital Movements 541. Definition and classification of capital movements,

p. 54; 2. Their mechanism under a homogeneous inter-national currency, p. 57; 3. Under national reservesystems, p. 60; 4. Under independent national cur-rencies, p. 63; 5. The control of international capitalmovements, p. 68.

LECTURE V. The Problems of a Really International Stan-dard 73

1. Gold as the international standard, p. 73; 2. The" perverse elasticity " of our credit money, p. 76; 3. TheChicago Plan of banking reform, p. 81; 4. Exchangerates and gold reserves, p. 84; 5. Credit policy of thecentral banks, p. 88; 6. Conclusions, p. 92.

NOTE

SINGE its establishment in 1927, the Graduate Institute ofInternational Studies at Geneva has organised, in addi-tion to its regular instruction, certain series of lecturesby experts from outside its own staff. In spite of the ori-ginal character of most of these lectures, the Institutecannot publish all of them.

The Directors believe, however, that they ought tomake certain of them available to a larger public.

The task of the Directors of the Institute is limited tothe selection of their authors. The authors have been leftcomplete freedom of thought, and naturally, therefore,they bear the sole responsibility not only for their opi-nions, but also for the form of their expression.

Paul MANTOUX.

W. E. RAPPARD.

PREFACE

THE five lectures which are here reproduced are necessa-rily confined to certain aspects of the wide subject indi-cated by the title. They are printed essentially as theywere delivered and, as is explained in the first lecture,limitations of time made it necessary to chose betweendiscussing the concrete problems of the present policy ofMonetary Nationalism and concentrating on the broadertheoretical issues on which the decision between an inter-national standard and independent national currenciesmust ultimately be based. The first course would haveinvolved a discussion of such technical questions as theoperations of Exchange Equalization Accounts, ForwardExchanges, the choice and adjustment of parities, coope-ration between central banks, etc., etc. The reader willfind little on these subjects in the following pages. Itappeared to me more important to use the time availableto discuss the general ideas which are mainly responsiblefor the rise of Monetary Nationalism and to which it ismainly due that policies and pratices which not long agowould have been frowned upon by all responsible finan-cial experts, are now generally employed throughout theworld. The immediate influence of the theoretical specu-lation is probably weak, but that it has had a profoundinfluence in shaping those views which to-day dominatemonetary policy is not open to serious question. It see-med to me better therefore to concentrate on these widerissues.

This decision has permitted me a certain freedom inthe discussion of alternative policies. In discussing the

XII PREFACE

merits of various systems I have not felt bound to confinemyself to those which may to-day be considered practicalpolitics. I have no doubt that to those who take thepresent trend of intellectual development for grantedmuch of the discussion in the following pages will appearhighly academic. Yet fundamentally the alternative poli-cies here considered are no more revolutionary or imprac-ticable than the deviations from traditional practice whichhave been widely discussed and which have even beenattempted in recent years—except that at the momentnot so many people believe in them. But while the poli-tician—and the economist when he is advising on con-crete measures—must take the state of opinion for grantedin deciding what changes can be contemplated here andnow, these limitations are not necessary when we areasking what is best for the human race in general. I amprofoundly convinced that it is academic discussion ofthis sort which in the long run forms public opinion andwhich in consequence decides what will be practical poli-tics some time hence. I regard it therefore not only asthe privilege but as the duty of the academic economistto take all alternatives into consideration, however remotetheir realisation may appear at the moment.

And indeed I must confess thatit seems to me in manyrespects the future development of professional and publicopinion on these matters is much more important thanany concrete measure which may be taken in the nearfuture. Whatever the permanent arrangements in mone-tary policy, the spirit in which the existing institutionsare administered is at least as important as these institu-tions themselves. And just as, long before the break-down of the international gold standard in 1931, mone-tary policy all over the world was guided by the ideas ofMonetary Nationalism which eventually brought itsbreakdown, so at the present time there is grave danger

PREFACE xiii

that a restoration of the external apparatus of the goldstandard may not mean a return to a really internationalcurrency. Indeed I must admit that—although I am aconvinced believer in the international gold standard—I regard the prospects of its restoration in the near futurenot without some concern. Nothing would be more fatalfrom a long run point of view than if the world attempteda formal return to the gold standard before people hadbecome willing to work it, and if, as would be quite pro-bable under these circumstances, this were soon followedby a renewed collapse. And although this would pro-bably be denied by the advocates of Monetary Nationalism,it seems to me as if we had reached a stage where theirviews have got such a hold on those in responsible posi-tions, where so much of the traditional rules of policyhave either been forgotten or been displaced by otherswhich are, unconsciously perhaps, part of the new phi-losophy, that much must be done in the realm of ideasbefore we can hope to achieve the basis of a stable inter-national system. These lectures were intended as a smallcontribution to this preparatory work which must pre-cede a successful reconstruction of such a system.

It was my good fortune to be asked to deliver theselectures at the Institut Universitaire de Hautes EtudesInternationales at Geneva. I wish here to express myprofound gratitude for the opportunity thus afforded andfor the sympathetic and stimulating discussion whichfollowed the lectures. My thanks are particularly due tothe directors of the Institute, Professors Rappard andMantoux, not only for arranging the lectures but also forundertaking their publication in the present series.

I am also indebted to a number of my friends and col-leagues at the London School of Economics, particularlyto Dr. F. Benham, Mr. F. Paish, Professor Robbins, andMr. G. H. Secord, who kave read the manuscript and

xiv PREFACE

offered much valuable advice as regards the subject matterand the form of exposition of these lectures. This wouldcertainly have been a much bigger and much better bookif I had seen my way to adopt and incorporate all theirsuggestions. But at the moment I do not feel preparedto undertake the larger investigation which my friendsrightly think the subject deserves. I alone must thereforebear the blame for the sketchy treatment of some impor-tant points and for any shortcomings which offend thereader.

I hope however it will be born in mind that these lec-tures were written to be read aloud and that this forbadeany too extensive discussion of the more intricate theore-tical points involved. Only at a few points, I have addeda further explanatory paragraph or restored sectionswhich would not fit into the time available for the lec-ture. That this will not suffice to provide satisfactoryanswers to the many questions I have raised I have nodoubt.

F.A. VON HAYEK.

London School of Economicsand Political Science.

May 1937.

LECTURE I

NATIONAL MONETARY SYSTEMS

WHEN I was honoured with the invitation to deliver atthe Institut five lectures " on some subject of distinctlyinternational interest ", I could have little doubt whatthat subject should be. In a field in which I am particu-larly interested I had been watching for years withincreasing apprehension the steady growth of a doctrinewhich, if it becomes dominant, is likely to deal a fatalblow to the hopes of a revival of international economicrelations. This doctrine, which in the title of these lec-tures I have described as Monetary Nationalism, is heldby some of the most brilliant and influential economistsof our time. It has been practised in recent years to anever increasing extent, and in my opinion it is largelyresponsible for the particular intensification of the lastdepression which was brought about by the successivebreakdown of the different currency systems. It willalmost certainly continue to gain influence for some timeto come, and it will probably indefinitely postpone therestoration of a truly international currency system.Even if it does not prevent the restoration of an interna-tional gold standard, it will almost inevitably bring aboutits renewed breakdown soon after it has been re-esta-blished.

When I say this I do not mean to suggest that a resto-ration of the gold standard of the type we have knownis necessarily desirable, nor that much of the criticism

2 NATIONAL MONETARY SYSTEMS

directed against it may not be justified. My complaint israther that most of this criticism is not concerned withthe true reasons why the gold standard, in the form inwhich we knew it, did not fulfill the functions for whichit was designed; and further that the only alternativeswhich are seriously considered and discussed, completelyabandon what seems to me the essentially sound prin-ciple—that of an international currency system—whichthat standard is supposed to embody.

But let me say at once that when I describe the doc-trines I am going to criticize as Monetary Nationalism Ido not mean to suggest that those who hold themare actuated by any sort of narrow nationalism. Thevery name of their leading exponent, Mr. J. M. Keynes,testifies that this is not the case. It is not the motiveswhich inspire those who advocate such plans, but theconsequences which I believe would follow from theirrealization, which I have in mind when I use this term.I have no doubt that the advocates of these doctrines sin-cerely believe that the system of independent nationalcurrencies will reduce rather than increase the causes ofinternational economic friction; and that not merely onecountry but all will in the long run be better off if thereis established that freedom in national monetary policieswhich is incompatible with a single international mone-tary system.

The difference then is not one about the ultimate endsto be achieved. Indeed, if it were, it would be useless totry to solve it by rational discussion. The fact is ratherthat there are genuine differences of opinion among eco-nomists about the consequences of the different types ofmonetary arrangements we shall have to consider, diffe-rences which prove that there must be inherent in theproblem serious intellectual difficulties which have notyet been fully overcome. This means that any discussion

NATIONAL MONETARY SYSTEMS 3

of the issues involved will have to grapple with consi-derable technical difficulties, and that it will have tograpple with wide problems of general theory if it is tocontribute anything to their solution. My aim through-out will be to throw some light on a very practical andtopical problem. But I am afraid my way will have tolead for a considerable distance through the arid regionsof abstract theory.

There is indeed another way in "which I might havedealt with my subject. And when I realized how muchpurely theoretical argument the other involved I wasstrongly tempted to take it. It would have been to avoidany discussion of the underlying ideas and simply to takeone of the many concrete proposals for independentnational currency systems now prevalent and to considerits various probable effects. I have no doubt that in thisform I could give my lectures a much more realisticappearance and could prove to the satisfaction of all whohave already an unfavourable opinion of Monetary Natio-nalism that its effects are pernicious. But I am afraid Iwould have had little chance of convincing anyone whohas already been attracted by the other side of the case.He might even admit all the disadvantages of the proposalwhich I could enumerate, and yet believe that its advan-tages outweigh the defects. Unless I can show thatthese supposed advantages are largely illusory, I shallnot have got very far. But this involves an examinationof the argument of the other side. So I have come ratherreluctantly to the conclusion that I cannot shirk the muchmore laborious task of trying to go to the»root of the theo-retical differences.

NATIONAL MONETARY SYSTEMS

But it is time for me to define more exactly what Imean by Monetary Nationalism and its opposite, an Inter-national Monetary System. By Monetary Nationalism Imean the doctrine that a country's share in the world'ssupply of money should not be left to be determined bythe same principles and the same mechanism as thosewhich determine the relative amounts of money in its dif-ferent regions or localities. A truly International Mone-tary System would be one where the whole world pos-sessed a homogeneous currency such as obtains withinseparate countries and where its flow between regionswas left to be determined by the results of the action ofall individuals. I shall have to define later what exactlyI mean by a homogeneous currency. But I should liketo make it clear at the outset that I do not believe that thegold standard as we knew it conformed to that ideal andthat I regard this as its main defect.

Now from this conception of Monetary Nationalismthere at once arises a question. The monetary relationsbetween small adjoining areas are alleged to differ fromthose between larger regions or countries; and this diffe-rence is supposed to justify or demand different monetaryarrangements. We are at once led to ask what is thenature of this alleged difference? This question is some-what connected but not identical with the question whatconstitutes a national monetary system, in what sense wecan speak of different monetary systems. But, as weshall see, it is very necessary to keep these questions apart.For if we do not we shall be confused between differenceswhich are inherent in the underlying situation and whichmay make different monetary arrangements desirable, anddifferences which are the consequence of the particular


monetary arrangements which are actually in existence.For reasons which 1 shall presently explain this dis-

tinction has not always been observed. This has led tomuch argument at cross purposes, and it is thereforenecessary to be rather pedantic about it.

I shall begin by considering a situation where there isas little difference as is conceivable between the money ofdifferent countries, a case indeed where there is so littledifference that it becomes doubtful whether we can speakof different " systems ". I shall assume two countriesonly and I shall assume that in each of the two countriesof which our world is assumed to consist, there is onlyone sort of widely used medium of exchange, namelycoins consisting of the same metal. It is irrelevant forour purpose whether the denomination of these coins inthe two countries is the same, so long as we assume, aswe shall, that the two sorts of coins are freely and withoutcost interchangeable at the mints. It is clear that themere difference in denomination, although it may meanan inconvenience, does not constitute a relevant diffe-rence in the currency systems of the two countries.*

1 Since these lines were written a newly published book hascome to my hand in which almost the whole argument in favourof Monetary Nationalism is based on the assumption that differentnational currencies are different commodities and that consequentlythere ought to be variable prices of them in terms of each other.(C. R. WHITTLESEY, International Monetary Issues, New York, 1937.)No attempt is made to explain why or under what conditions andin what sense the different national moneys ought to be regardedas different commodities, and one can hardly avoid the impressionthat the author has uncritically accepted the difference of deno-mination as proof of the existence of a difference in kind. Thecase illustrates beautifully the prevalent confusion between diffe-rences between the currency systems which can be made an argu-ment for national differentiations and those which are a


In starting from this case we follow a long establishedprecedent. A great part of the argument of the classicalwriters on money proceeded on this assumption of a" purely metallic currency ". I wholly agree with thesewriters that for certain purposes it is a very usefulassumption to make. I shall however not follow themin their practice of assuming that the conclusions arrivedfrom these assumptions can be applied immediately tothe monetary systems actually in existence. This beliefwas due to their conviction that the existing mixed cur-rency systems not only could and should be made tobehave in every respect in the same way as a purely metal-lic currency, but that—at any rate in England since theBank Act of 1844—the total quantity of money wasactually made to behave in this way. I shall argue later.that this erroneous belief is responsible for much confu-sion about the mechanism of the gold standard as it exis-ted; that it has prevented us from achieving a satisfactorytheory of the working of the modern mixed system, sincethe explanation of the r61e of the banking system wasonly imperfectly grafted upon, and never really integratedwith, the theory of the purely metallic currency; and thatin consequence (he gold standard or the existence of aninternational system was blamed for much which in fact

consequence of such differentiations. That it is "only a differencein nomenclature " (as Professor Gregory has well put it) whetherwe express a given, quantity of gold as Pounds, Dollars or Marks,and that this no more constitutes different commodities than thesame quantity of cloth becomes a different commodity when it isexpressed in meters instead of in yards, ought to be obvious.Whether different national currencies are in any sense differentcommodities depends on what we make them, and the real pro-blem is whether we should create differentiations between thenational currencies by using in each national territory a kind ofmoney which will be generally acceptable only within that terri-tory, or whether the same money should be used in the differentnational territories.


was really due to the mixed character of the system andnot to its " internationalism " at all.

For my present purpose, however, namely to findwhether and in what sense the monetary mechanism ofone country can or must be regarded as a unit or a sepa-rate system, even when there is a minimum of differencebetween the kind of money used there and elsewhere, thecase of the " purely metallic currency " serves extraordi-narily well. If there are differences in the working ofthe national monetary systems which are not merely aneffect of the differences in the monetary arrangements ofdifferent countries, but which make it desirable that thereshould be separate arrangements for different regions,they must manifest themselves even in this simplest case.

It is clear that in this case the argument for a nationalmonetary system cannot rest on any peculiarities of thenational money. It must rest, and indeed it does rest,on the assumption that there is a particularly close con-nection between the prices—and particularly the wages—within the country which causes them to move to a con-siderable degree up and down together compared withprices outside the country. This is frequently regardedas sufficient reason why, in order to avoid the necessitythat the " country as a whole " should have to raise orlower its prices, the quantity of money in the countryshould be so adjusted as to keep the " general price level "within the country stable. I do not want to consider thisargument yet. I shall later argue that it rests largely onan illusion, based on the accident that the statistical mea-sures of prices movements are usually constructed forcountries as such; and that in so far as there are genuinedifficulties connected with general downward adjust-ments of many prices, and particularly wages, the pro-posed remedy would be worse than the disease. But Ithink I ought to say here and now that I regard it as the


only argument on which the case for monetary nationa-lism can be rationally based. All the other argumentshave really nothing to do with the existence of an inter-national monetary system as such, but apply only to theparticular sorts of international systems with which weare familiar. But since these arguments are so inextri-cably mixed up in current discussion with those of a morefundamental character it becomes necessary, before wecan consider the main arguments on its merits, to con-sider them first.

The homogeneous international monetary system whichwe have just considered was characterised by the fact thateach unit of the circulating medium of each countrycould equally be used for payments in the other countryand for this purpose could be bodily transferred into thatother country and be bodily transformed into the cur-rency of that country. Among the systems which need tobe considered only an international gold standard withexclusive gold circulation in all countries would conformto this picture. This has never existed in its pure formand the type of gold standard which existed until fairlyrecently was even further removed from this picture thanwas generally realized. It was never fully appreciatedhow much the operation of the system which actuallyexisted diverged from the ideal pure gold standard. Forthe points of divergence were so familiar that they wereusually taken for granted. It was the design of the BankAct of 1844 to make the mixed system of gold and othermoney behave in such a way that the quantity of moneywould change exactly as if only gold were in circulation;and for a long time argument proceeded as if this inten-tion had actually been realized. And even when it was


gradually realized that deposits subject to cheque were noless money than bank notes, and that since they were leftout of the regulation, the purpose of the Act had reallybeen defeated, only a few modifications of the argumentwere thought necessary. Indeed in general this argu-ment is still presented as it was originally constructed,on the assumption of a purely metallic currency.

In fact however with the coming of modern banks acomplete change had occurred. There was no longer onehomogeneous sort of money in each country, the differentunits of which could be regarded as equivalent for allrelevant purposes. There had arisen a hierarchy of diffe-rent kinds of money within each country, a complex orga-nisation which possessed a definite structure, and whichis what we really mean when we speak of the circulatingmedium of a country as a " system ". It is probablymuch truer to say that it is the difference between thedifferent kinds of money which are used in any one coun-try, rather than the differences between the moneys usedin different countries, which constitutes the real diffe-rence between different monetary systems.

We can see this if we examine matters a little more clo-sely. The gradual growth of banking habits, that is thepractice of keeping liquid assets in the form of bankbalances subject to cheque, meant that increasing num-bers of people were satisfied to hold a form of the circu-lating medium which could be used directly only for pay-ments to people who banked with the same institution.For all payments beyond this circle they relied on theability of the bank to convert the deposits on demand intoanother sort of money which was acceptable in widercircles; and for this purpose the banks had to keep a" reserve " of this more widely acceptable or more liquidmedium.

But this distinction between bank deposits and " cash "


in the narrower sense of the term does not yet exhaustthe classification of different sorts of money, possessingdifferent degrees of liquidity, which are actually used ina modern community. Indeed, this development wouldhave made little difference if the banks themselves had notdeveloped in a way which led to their organisation intobanking " systems " on national lines. Whether thereexisted only a system of comparatively small local unitbanks, or whether there were numerous systems of branchbanks which covered different areas freely overlappingand without respect to national boundaries, there wouldbe no reason why all the monetary transactions withina country should be more closely knit together thanthose in different countries. For any excess paymentsoutside their circle the customers of any single bank, itis true, would be dependent on the reserve kept for thispurpose for them by their bank, and might therefore findthat their individual position might be affected by whatother members of this circle did. But at most the inha-bitants of some small town would in this way becomedependent on the same reserves and thereby on oneanother's action,1 never all the inhabitants of a big areaor a country.

It was only with the growth of centralized nationalbanking systems that all the inhabitants of a countrycame in this sense to be dependent on the same amountof more liquid assets held for them collectively as a natio-nal reserve. But the concept of centralisation in this con-nection must not be interpreted too narrowly as referringonly to systems crowned by a central bank of the familiartype, nor even as confined to branch banking systemswhere each district of a country is served by the branchesof the same few banks. The forms in which centralisa-

1 Compare on this and the following L. ROBBINS, EconomicPlanning and International Order, 1937, pp. 274 et seq.

NATIONAL MONETARY SYSTEMS l i

tion, in the sense of a system of national reserves whichis significant here, may develop, are more varied thanthis and they are only partly due to deliberate legislativeinterference. They are partly due to less obvious institu-lonal factors.

For even in the absence of a central bank and of branchbanking the fact that a country usually has one financialcentre where the stock exchange is located and throughwhich a great proportion of its foreign trade passes or isfinanced tends to have the effect that the banks in thatcentre become the holders of a large part of the reserveof all the other banks in the country. The proximity ofthe stock exchange puts them in a position to invest suchreserves profitably in what, at any rate for any singlebank, appears to be a highly liquid form. And the grea-ter volume of transactions in foreign exchange in such acentre makes it natural that the banks outside will relyon their town correspondents to provide them withwhatever foreign money they may need in the course oftheir business. It was in this way that long before thecreation of the Federal Reserve System in 1913 and inspite of the absence of branch banking there developedin the United States a system of national reserves underwhich in effect all the banks throughout their territoryrelied largely on the same ultimate reserves. And a some-what similar situation existed in Great Britain before thegrowth of joint stock banking.

But this tendency is considerably strengthened ifinstead of a system of small unit banks there are a fewlarge joint stock banks with many branches; still moreif the whole system is crowned by a single central bank,the holder of the ultimate cash reserve. This system,which to-day is universal, means in effect that additionaldistinctions of acceptability or liquidity have been artifi-cially created between three main types of money, and


that the task of keeping a sufficient part of the total assetsin liquid form for different purposes has been divided bet-ween different subjects. The ordinary individual willhold only a sort of money which can be used directly onlyfor payments to clients of the same bank; he relies uponthe assumption that his bank will hold for all its clientsa reserve which can be used for other payments. Thecommercial banks in turn will only hold reserves of suchmore liquid or more widely acceptable sort of money ascan be used for inter-bank payments within the country.But for the holding of reserves of the kind which can beused for payments abroad, or even those which arerequired if the public should want to convert a conside-rable part of its deposits into cash, the banks rely largelyon the central bank.

This complex structure, which is often described as theone-reserve system, but which I should prefer to call thesystem of national reserves, is now taken so much forgranted that we have almost forgotten to think about itsconsequences. Its effects on the mechanism of interna-tional flows of money will be one of the main subjects ofmy next lecture. To-day I only want to stress twoaspects which are often overlooked. In the first place Iwould emphasize that bank deposits could never haveassumed their present predominant r61e among the diffe-rent media of circulation, that the balances held on cur-rent account by banks could never have grown to tentimes and more of their cash reserves, unless some organ,be it a privileged central bank or be it a number of or allthe banks, had been put in a position, to create in caseof need a sufficient number of additional bank notes tosatisfy any desire on the part of the public to convert aconsiderable part of their balances into hand-to-handmoney. It is in this sense and in this sense only that the


existence of a national reserve system involves the ques-tion of the regulation of the note issue, alone.

The second point is that nearly all the practical pro-blems of banking policy, nearly all the questions withwhich a central banker is daily concerned, arise out ofthe co-existence of these different sorts of money withinthe national monetary system. Theoretical economistsfrequently argue as if the quantity of money in the coun-try were a perfectly homogeneous magnitude and entirelysubject to deliberate control by the central monetaryauthority. This assumption has been the source of muchmutual misunderstanding on both sides. And it has hadthe effect that the fundamental dilemma of all centralbanking policy has hardly ever been really faced : theonly effective means by which a central bank can con-trol an expansion of the generally used media of circula-tion is by making it clear in advance that it will not pro-vide the cash (in the narrower sense) which will berequired in consequence of such expansion, but at thesame time it is recognised as the paramount duty of a cen-tral bank to provide that cash once the expansion of bankdeposits has actually occurred and the public begins todemand that they should be converted into notes or gold.

I shall be returning to this problem later. But in thenext two lectures my main concern will be another set ofproblems. I shall argue that the existence of nationalreserve systems is the real source of most of the difficul-ties which are usually attributed to the existence of aninternational standard. I shall argue that these difficul-ties are really due to the fact that the mixed national cur-rencies are not sufficiently international, and that mostof the criticism directed against the gold standard quainternational standard is misdirected. I shall try toshow that the existence of national reserve systems altersthe mechanism of the international money flows from


what it would be with a homogeneous international cur-rency to a much greater degree than is commonly rea-lized.

But before I can proceed to this major task I mustshortly consider the third and most efficient cause whichmay differentiate the circulating media of differentcountries and constitute separate monetary systems. Upto this point I have only mentioned cases where the ratiobetween the monetary units used in the different coun-tries was given and constant. In the first case this wassecured by the fact that the money circulating in thedifferent countries was assumed to be homogeneous in allessential respects, while in the second and more realisticcase it was assumed that, although different kinds ofmoney were used in the different countries, there was yetin operation an effective if somewhat complicated mecha-nism which made it always possible to convert at aconstant rate money of the one country into money oftne other. To complete the list there must be added thecase where these ratios are variable : that is, where therate of exchange between the two currencies is subject tofluctuations.

With monetary systems of this kind we have of courseto deal with differences between the various sorts ofmoney which are much bigger than any we have yetencountered. The possession of a quantity of money cur-rent in one country no longer gives command over a defi-nite quantity of money which can be used in anothercountry. There is no longer a mechanism which securesthat an attempt to transfer money from country to countrywill lead to a decrease in the quantity of money in onecountry and a corresponding increase in the other. In


fact an actual transfer of money from country to countrybecomes useless because what is money in the one coun-try is not money in the other. We have here to deal withthings which possess different degrees of usefulness fordifferent purposes and the quantities of which are fixedindependently.

Now I think it should be sufficiently clear that any dif-ferences between merely interlocal and internationalmovements of money which only arise as a consequenceof the variability of exchange rates cannot themselves beregarded as a justification for the existence of separatemonetary systems. That would be to confuse effect andcause—to make the occasion of difference the justifica-tion of its perpetuation. But since the adoption of such asystem of " flexible parities " is strongly advocated as aremedy for the difficulties which arise out of other diffe-rences which we have already considered, it will be expe-dient if in the following lectures I consider side by sideall three types of conditions under which differences bet-ween the national monetary systems may arise. We shallbe concerned with the way in which in each case redis-tributions of the relative amounts of money in the differentcountries are effected. I shall begin with the only casewhich can truly be described as an international monetarystandard, that of a homogeneous international currency.Consideration of this case will help me to show whatfunctions changes in the relative quantities of money indifferent regions and countries may be conceived to serve;and how such changes are spontaneously brought about.I shall then proceed to the hybrid " mixed " system whichuntil recently was the system generally in vogue andwhich is meant when, in current discussion, the tradi-tional gold standard is referred to. As I said at thebeginning, I shall not deny that this system has seriousdefects. But while the Monetary Nationalists believe that


these defects are due to the fact that it is still an interna-tional system and propose to remove them by substitutingthe third or purely national type of monetary system forit, I shall on the contrary attempt to show that its defectslie in the impediments which it presents to the free inter-national flow of funds. This will then lead me first to anexamination of the peculiar theory of inflation and defla-tion on which Monetary Nationalism is based : then toan investigation of the consequences which we shouldhave to expect if its proposals were acted upon; andfinally to a consideration of the methods by which a moretruly international system could be achieved.

LECTURE II

THE FUNCTION AND MECHANISMOF INTERNATIONAL FLOWS OF MONEY

AT the end of my first lecture I pointed out that the threedifferent types of national monetary systems which wehave been considering differed mainly in the method bywhich they effected international redistributions ofmoney. In the case of a homogeneous international cur-rency such a redistribution is effected by actual transfersof the corresponding amounts of money from country tocountry. Under the " mixed " system represented by thetraditional gold standard—better called " gold nucleusstandard"—it is brought about partly by an actualtransfer of money from country to country, but largelyby a contraction of the credit superstructure in the onecountry and a corresponding expansion in the other. Butalthough the mechanism and, as we shall see, some ofthe effects, of these two methods are different, the finalresult, the change in the relative value of the total quan-tities of money in the different countries, is brought aboutby a corresponding change in the quantity of money, thenumber of money units, in each country. Under the thirdsystem, however, the system of independent currencies,things are different. Here the adjustment is broughtabout, not by a change in the number of money units ineach country, but by changes in their relative value. Nomoney actually passes from country to country, and what-ever redistribution of money between persons may be

18 INTERNATIONAL FLOWS OF MONEYinvolved by the redistribution between countries has tobe brought about by corresponding changes inside eachcountry.

Before, however, we can assess the merits of the diffe-rent systems it is necessary to consider generally the dif-ferent reasons why it may become necessary that the rela-tive values of the total quantities of money in differentcountries should alter. It is clear that changes in thedemand for or supply of the goods and services producedin an area may change the value of the share of theworld's income which the inhabitants of that area mayclaim. But changes in the relative stock of money,although of course closely connected with these changesof the shares in the world's income which different coun-tries can claim, are not identical with them. It is onlybecause people whose money receipts fall will in generaltend to reduce their money holdings also and vice-versa,that changes in the size of the money stream in the dif-ferent countries will as a rule be accompagnied bychanges in the same direction in the size of the moneyholdings. People who find their income increasing willgenerally at first take out part of the increased moneyincome in the form of a permanent increase in their cashbalances, while people whose incomes decrease will tendto postpone for a while a reduction of their expenditureto the full extent, prefering to reduce their cashbalances.x To this extent changes in the cash balancesserve, as it were, as cushions which soften the impactand delay the adaptation of the real incomes to thechanged money incomes, so that in the interval money isactually taken as a substitute for goods.

But, given existing habits, it is clear that changes inthe relative size of money incomes—and the same applies

1 For a full description of this mechanism cf. R. G. HAWTREY,Currency and Credit, chapter IV, 3rd ed., 1928, pp. 41-63.

INTERNATIONAL FLOWS OF MONEY 19

to the total volume of money transactions—of differentcountries make corresponding changes in the moneystocks of these countries inevitable; changes which,although they need not be in the same proportion, mustat any rate be in the same direction as the changes inincomes. If the share in the world's production whichthe output of a country represents, rises or falls, the shareof the total which the inhabitants of the country canclaim will fully adapt itself to the new situation only aftermoney balances have been adjusted.

Changes in the demand for money on the part of a par-ticular country may of course also occur independentlyof any change in the value of the resources its inhabi-tants can command. They may be due to the fact thatsome circumstances may have made its people want tohold a larger or smaller proportion of their resources inthe most liquid form, i.e. in money. If so, then for atime they will offer to the rest of the world more commo-dities, receiving money in exchange. This enables them,at any later date, to buy more commodities than theycan currently sell. In effect they decide to lend to therest of the world that amount of money's worth of com-modities in order to be able to call it back whenever theywant it.

The function which is performed by internationalmovements of money will be seen more clearly if we pro-ceed to consider such movement in the simplest case

1 Perhaps, intead of speaking of the world's output, I shouldhave spoken about the share in the command over the world'sresources, since of course it is not only the current consumableproduct but equally the command over resources which will yielda product only in the future which is distributed by this monetarymechanism.

20 INTERNATIONAL FLOWS OF MONEY

imaginable—a homogeneous international or " purelymetallic " currency. Let us suppose that somebody whoused to spend certain sums on products of country A nowspends them on products of country B. The immediateeffect of this is the same whether this person himself isdomiciled in A or in B. In either case there will arisean excess of payments from A to B—an adverse balanceof trade for A—, either because the total of such paymentshas risen or because the amount of payments in the oppo-site direction has fallen off. And if the initiator of thischange persists in his new spending habits, this flow ofmoney will continue for some time.

But now we must notice that because of this in A some-body's money receipts have decreased and in B some-body's money receipts have increased. We have longbeen familiar with the proposition that counteractingforces will in time bring the flow of money between thecountries to a stop. But it is only quite recently that theexact circumstances determining the route by which thiscomes about have been satisfactorily established1. Inboth countries the change in the money receipts of thepeople first affected will be passed on and disseminated.But how long the outflow of money from A to B will con-tinue depends on how long it takes before the successivechanges in money incomes set up in each country willbring about new and opposite changes in the balance ofpayments.

This result can be brought about in two ways in eachof the two countries. The reduction of money incomesin country A may lead to a decrease of purchases from B,

Cf. particularly F. W. PAISH, Banking Policy and the Balance ofInternational Payments (Economica, N. S., vol. Ill, no. 12, Nov.1996); K. F. MAIER, Goldwanderungen, Jena 1935, and P. B. WHALE,The Working of the Pre-War Gold Standard (Economic^ vol. IV,no. 13, February 1937).


or the consequent fall of the prices of some goods in Amay lead to an increase of exports to B. And the increaseof money incomes in country B may lead to an increase ofpurchases from A or to a rise in the prices of some com-modities in B and a consequent decrease of exports to A.But how long it will take before in this way the flow ofmoney from A to B will be offset will depend on thenumber of links in the chains which ultimately lead backto the other country, and on the extent to which at eachof these points the change of incomes leads first to achange in the cash balances held, before it is passed onin full strength. In the interval money will continue toflow from A to B; and the total which so moves willcorrespond exactly to the amounts by which, in thecourse of the process just described, cash balances havebeen depleted in the one country and increased in theother.

This part of the description is completely general. Buiwe cannot say how many incomes will have to bechanged, how many individual prices will have to bealtered upwards or downwards in each of the two coun-tries, in consequence of the initial changes. For thisdepends entirely on the concrete circumstances of eachparticular case. In some countries and under some con-ditions the route will be short because some of the firstpeople whose incomes decrease cut down their expenditureon imported goods, or because the increase of incomesis soon spent on imported goods.1 In other cases theroute may be long and external payments will be madeto balance only after extensive price changes haveoccured, which induce further people to change the direc-tion of their expenditure.

The important point in all this is that what incomes

1 Cf. on this particularly the article by F. W. P&ish just quoted.


and what prices will have to be altered in consequence ofthe initial change will depend on whether and to whatextent the value of a particular factor or service, directlyor indirectly, depends on the particular change indemand which has occurred, and not on whether it isinside or outside the same " currency area ". We cansee this more clearly if we picture the series of successivechanges of money incomes, which will follow on theinitial shift of demand, as single chains, neglecting forthe moment the successive ramifications which will occurat every link. Such a chain may either very soon lead tothe other country or first run through a great many linksat home. But whether any particular individual in thecountry will be affected will depend whether he is a linkin that particular chain, that is whether he has more orless immediately been serving the individuals whoseincome has first been affected, and not simply on whetherhe is in the same country or not. In fact this picture ofthe chain makes it clear that is is not impossible thatmost of the people who ultimately suffer a decrease ofincome in consequence of the initial transfer of demandfrom A to B may be in B and not in A. This is often over-looked because the whole process is presented as if thechain of effects came to an end as soon as payments bet-ween the two countries balance. In fact however each ofthe two chains—that started by the decrease of some-body's income in A, and that started by the increase ofanother persons income in B—may continue to run onfor a long time after they have passed into the othercountry, and may have even a greater number of linksin that country than in the one where they started. Theywill come to an end only when they meet, not only inthe same country but in the same individual, so finallyoffsetting each other. This means that the number ofreductions of individual incomes and prices (not their


aggregate amount) which becomes necessary in conse-quence of a transfer of money from A to B may actuallybe greater in B than in A.

This picture is of course highly unrealistic because itleaves out of account the infinite ramifications to whicheach of these chains of effects will develop. But evenso it should, I think, make it clear how superficial andmisleading the kind of argument is which runs in termsof the prices and the incomes of the country, as if theywould necessarily move in unison or even in the samedirection. It will be prices and incomes of particularindividuals and particular industries which will be affec-ted and the effects will not be essentially different fromthose which will follow any shifts of demand betweendifferent industries or localities.

This whole question is of course the same as that whichI discussed in my first lecture in connection with the pro-blem of what constitutes one monetary system, namelythe question of whether there exists a particularly closecoherence between prices and incomes, and particularlywages, in any one country which tends to make themmove as a whole relatively to the price structure outside.As I indicated then, I shall not be able to deal with itmore completely until later on. But there are twopoints which, I think, will have become clear now andwhich are important for the understanding of the contrastbetween the working of the homogeneous internationalcurrency we are considering, and the mixed system towhich I shall presently proceed.

In the first place it already appears very doubtfulwhether there is any sense in which the terms inflationand deflation can be appropriately applied to these inter-regional or international transfers of money. If, ofcourse, we define inflation and deflation as changes inthe quantity of money, or the price level, within a parti-


cular territory, then the term naturally applies. But it isby no means clear that the consequences which we canshow will follow if the quantity of money in a closedsystem changes will also apply to such redistributions ofmoney between areas. In particular there is no reasonwhy the changes in the quantity of money within an areashould bring about those merely temporary changes inrelative prices which, in the case of a real inflation, leadto misdirections of production—misdirections becauseeventually the inherent mechanism of these inflationstends to reverse these changes in relative prices. Nor doesthere seem to exist any reason why, to use a more modernyet already obsolete terminology, saving and investmentshould be made to be equal within any particular areawhich is part of a larger economic system.x But all thesequestions can be really answered only when I come todiscuss the two conflicting views about the main signi-ficance of inflation and deflation which underlie most ofthe current disputes about monetary policy.

The second point which I want particularly to stressy here is that with a homogeneous international currency *^

there is apparently no reason why an outflow of moneyfrom one area and an inflow into another should neces-sarily cause a rise in the rate of interest in the first area .and a fall in the second. So far I have not mentioned'the rate of interest, because there seems to be no generalground why we should expect that the causes which leadto the money flows between two countries should affectthe rate of interest one way or the other. Whether theywill have such an effect and in what direction will dependentirely on the concrete circumstances. If the initialchange which reduces the money income of some peoplein one country leads to an immediate reduction of theirexpenditure on consumers' goods, and if in addition they

1 a . J. M. KEWBS, A Treatise on Money, 1930, vol. I, chapter 4.

v


use for additional investments the surplus of their cashbalances which they no longer regard worth keeping, itis not impossible that the effect may actually be a fallin the rate of interest.* And, conversely, in the coun-try towards whose product an additional money streamis directed, this might very well lead to a rise in the rateof interest. It seems that we have been led to regardwhat happens to be the rule under the existing mixedsystems as due to causes much more fundamental thanthose which actually operate. But this leads me to themost important difference between the cases of a " purelymetallic " and that of a " mixed " currency. To the lattercase, therefore, I now turn.

3

If in the two countries concerned there are two sepa-rate banking systems, whether these banking systems arecomplete with a central bank or not, considerable trans-fers of money from the one country to the other will beeffected by the actual transmission of only a part of thetotal, the further adjustment being brought about by an

1 Although it is even conceivable that a fall in incomes mightbring about a temporary rise in investments, because the peoplewho are now poorer feel that they can no longer afford the luxuryof the larger cash balances they used to keep before, and proceedto invest part of them, this is neither a very probable effect norlikely to be quantitatively significant. Much more important,however, may be the effect of the fall of incomes on the demandfor investment. Particularly if the greater part of the existingcapital equipment is of a very durable character a fall in incomesmay for some time almost completely suspend the need for invest-ment and in this way reduce the rate of interest in the countryquite considerably. Another case where the same cause whichwould lead to a flow of money from one country to another wouldat the same time cause a fall in the rate of interest in the firstwould be if in one of several countries where population used toincrease at the same rate, this rate were considerably decreased.

26 INTERNATIONAL FLOWS OF MONEYexpansion or contraction of the credit structure accordingas circumstances demand. It is commonly believed thatnothing fundamentally is changed but something issaved by substituting the extinction of money in oneregion and the creation of new money in the other for theactual transfer of money from individual to individual.This is however a view which can be held only on themost mechanistic form of the quantity theory and whichcompletely disregards the fact that the incidence of thechange will be very different in the two cases. Consi-dering the methods available to the banking system tobring about an expansion or contraction, there is no rea-son to assume that they can take the money to be extin-guished exactly from those persons where it would inthe course of time be released if there were no bankingsystem, or that they will place the additional money inthe hands of those who would absorb the money if itcame to the country by direct transfer from abroad.There are on the contrary strong grounds for believingthat the burden of the change will fall entirely, or toan extent which is in no way justified by the underlyingchange in the real situation, on investment activity inboth countries.

To see why and how this will happen it is necessary toconsider in some detail the actual organisation of thebanking systems and the nature of their traditional poli-cies. We have seen that where bank deposits are usedextensively this means that all those who hold their mostliquid assets in this form, rely on their banks to providethem whenever needed with the kind of money which isacceptable outside the circle of the clients of the bank.The banks in turn, and largely because they have learntto rely on the assistance of other (note issuing) banks,particularly the central bank, have come themselves tokeep only very slender cash reserves, that is, reserves


which they can use to meet any adverse clearing balanceto other banks or to make payments abroad. These areindeed not meant to do more than to tide over any tem-porary and relatively small difference between paymentsand receipts. They are altogether insufficient to allowthe banks ever to reduce these reserves by the full amountof any considerable reduction of their deposits. The verysystem of proportional reserves, which so far as depositsare concerned is to-day universally adopted and even inthe case of bank notes applies practically everywhere out-side Great Britain, means that the cash required for theconversion of an appreciable part of the deposits has tobe raised by compelling people to repay loans.

We shall best see the significance of such a bankingstructure with respect to international money flows if weconsider again the effects which are caused by an initialtransfer of demand from country A to country B. Themain point here is that, with a national banking systemworking on the proportional reserve principle, unless theadverse balance of payments corrects itself very rapidly,the central bank will not be in a position to let the out-flow of money go on until it comes to its natural end.It caanot, without endangering its reserve position, freelyconvert all the bank deposits or banknotes which willbe released by individuals into money which can betransferred into the other countries. If it wants to pre-vent an exhaustion or dangerous depletion of its reservesit has to speed up the process by which payments fromA to B will be decreased or payments from B to A willbe increased. And the only way in which it can do thisquickly and effectively is generally and indiscriminatelyto bring pressure on those who have borrowed from itto repay their loans. In this way it will set up additionalchains of successive reductions of outlay, first on the partof those to whom it would have lent and then on the


part of all others to whom this money would graduallyhave passed. So that leaving aside for the moment theeffects which a rise in interest rates will have on inter-national movements of short term capital we can see thatthe forces which earlier or later will reduce paymentsabroad and, by reducing prices of home products, stimu-late purchases from abroad will be intensified. And ifsufficient pressure is exercised in this way, the periodduring which the outflow of money continues, and the-reby the total amount of money that will actually leavethe country before payments in and out will balanceagain, may be reduced to almost any extent.

The important point, however, is that in this case thepeople who will have to reduce their expenditure in orderto produce that result will not necessarily be the samepeople who would ultimately have to do so under ahomogeneous international currency system, and that theequilibrium so reached will of its nature be only tempo-rary. In particular, since bank loans, to any significantextent, are only made for investment purposes, it willmean that the full force of the reduction of the moneystream will have to fall on investment activity. This isshown clearly by the method by which this restrictionis brought about. We have seen before that under apurely metallic currency an outflow of money need notactually bring about a rise in interest rates. It may, butthis is not necessary and it is even conceivable that theopposite will happen. But with a banking structureorganised on national lines, that is, under a nationalreserve system, it is inevitable that it will bring a risein interest rates, irrespective of whether the underlyingreal change has affected either the profitability of invest-ment or the rate of savings in such a way as to justifysuch a change. In other words, to use an expressionwhich has given rise to much dispute in the recent past


but which should be readily understood in this connec-tion, the rise of the bank rate under such circumstancesmeans that it has to be deliberately raised above the equi-librium or " natural " rate of interest.* The reason forthis is not, or need not be, that the initiating change hasaffected the relation between the supply of investiblefunds and the demand for them, but that it tends to dis-turb the customary proportion between the different partsof the credit structure and that the only way to restorethese proportions is to cancel loans made for investmentpurposes.

To some extent, but only to some extent, the credit con-traction will, as I have just said, by lowering pricesinduce additional payments from abroad and in this formoffset the outflow of money. But to a considerable extentits effect will be that certain international transfers ofmoney which would have taken the place of a transfer ofgoods and would in this sense have been a final paymentfor a temporary excess of imports will be intercepted, sothat consequently actual transfers of goods will have totake place. The transfer of only a fraction of the amountof money which would have been transferred under apurely metallic system, and the substitution of a multiplecredit contraction for the rest, as it were, deprives theindividuals in the country concerned of the possibility ofdelaying the adaptation by temporarily paying for anexcess of imports in cash.

That the rise of the rate of interest in the country that1 This has been rightly pointed out, but has hardly been suffi-

ciently explained, in an interesting article by J. G. Gilbert on thePresent Position of the Theory of International Trade, The Reviewof Economic Studies, vol. Ill, no. 1, October 1935, particularlypp. 23-6 — To say that money rates of interest in a particular coun-try may be made to deviate from the equilibrium rate by monetaryfactors peculiar to that country is of course not to say that theequilibrium rate in that country is independent of internationalconditions.

30 INTERNATIONAL FLOWS OF MONEYis losing gold, and the corresponding reduction in thebank rate in the country which is receiving gold, needhave nothing to do with changes in the demand for orthe supply of capital appears also from the fact that, ifno further change intervenes, the new rates will have tobe kept in force only for a comparatively short period,and that after a while a return to the old rates will bepossible. The changes in the rates serve the temporarypurpose of speeding up a process which is already underway. But the forces which would have brought the flowof gold to an end earlier or later in any case do not there-fore cease to operate. The chain of successive reductionsof income in country A set up by the initiating changeswill continue to operate and ultimately reduce the pay-ments out of the country still further. But since pay-ments in and payments out have in the meantime alreadybeen made to balance by the action of the banks, thiswill actually reverse the flow and bring about a favou-rable balance of payments. The banks, wanting to reple-nish their reserves, may let this go on for a while, butonce they have restored their reserves, they will be ableto resume at least the greater part of their lending acti-vity which they had to curtail.

This picture is admittedly incomplete because I havebeen deliberately neglecting the part played by shortterm capital movements. I shall discuss these in myfourth lecture. At present my task merely is to show howthe existence of national banking systems, based on thecollective holding of national cash reserves, alters theeffects of international flows of money. It seems to meimpossible to doubt that there is indeed a very conside-rable difference between the case where a country, whoseinhabitants are induced to decrease their share in theworld'8 stock of money by ten per cent, does so byactually giving up this ten per cent in gold, and the case


where, in order to preserve the accustomed reserve pro-portions, it pays out only one per cent in gold and con-tracts the credit super-structure in proportion to the reduc-tion of reserves. It is as if all balances of internationalpayments had to be squeezed through a narrow bottleneck as special pressure has to be brought on people, whowould otherwise not have been affected by the change, togive up money which they would have invested pro-ductively.

Now the changes in productive activity which aremade necessary in this way are not of a permanentnature. This means not only that in the first instancemany plans will be upset, that equipment which has beencreated will cease to be useful and that people will bethrown out of employment. It also means that therevised plans which will be made are bound soon to beequally disappointed in the reverse direction and that thereadjustment of production which has been enforced willprove to be a misdirection. In other words, it is a dis-turbance which possesses all the characteristics of apurely monetary disturbance, namely that it is self-reversing in the sense that it induces changes which willhave to be reversed because they are not based on anycorresponding change in the underlying real facts.

It might perhaps be argued that the contraction of cre-dit in the one country and the expansion in the otherbrings about exactly the same effects that we should expectfrom a transfer of a corresponding amount of capital fromthe one country to the other, and that since the amountof money which would otherwise have to be transferredwould represent so much capital, there can be no harmin the changes in the credit structure. But the point isexactly that not every movement of money is in this sensea transfer of capital. If a group of people want to holdmore money because the value of their income rises,


while another group of people reduce their moneyholdings because the value of their income falls, there isno reason why in consequence the funds available forinvestment in the first group should increase and thoseavailable in the second group should decrease. It is, onthe other hand, quite possible that the demand for suchfunds in the first group will rise and in the second groupwill fall. In such a case, as we have seen, there wouldbe more reason to expect that the rate of interest will risein the country to which the money flows rather than inthe country from which the money comes.

The case is of course different when the initiating causeis not a shift in demand from one kind of consumers'goods to another kind of consumers' goods, but whenfunds which have been invested in one type of producers'goods in one country are transferred to investment inanother type of producers' goods in another country.Then indeed we have a true movement of capital and weshould be entitled to expect it to affect interest rates inthe usual manner. What I am insisting on is merelythat this need not be the general rule and that the factthat it is generally the case is not the effect of an inherentnecessity but due to purely institutional reasons.

There are one or two further points which I mustshortly mention before I can conclude this subject. Oneis the rather obvious point that the disturbing effects ofthe organisation of the world's monetary system on thenational reserve principle are of course considerablyincreased when the rate of multiple expansion or con-traction, which will be caused by a given increase ordecrease of gold, is different in different countries. Ifthis is the case, and it has of course always been the case


under the gold standard as we knew it, it means thatevery flow of gold from one country to another will meaneither an inflation or a deflation from the world point ofview, accordingly as the rate of secondary expansion isgreater or smaller in the country receiving gold than inthe country losing gold.

The second point is one on which I am particularlyanxious not to be misunderstood. The defects of themixed system which I have pointed out are not defectsof a particular kind of policy, or of special rules of cen-tral bank practice. They are defects inherent in the sys-tem of the collective holding of proportional cash reservesfor national areas, whatever the policy adopted by thecentral bank or the banking system. What I have saidprovides in particular no justification for the commoninfringements of the " rules of the game of the goldstandard ", except, perhaps for a certain reluctance tochange the discount rate too frequently or too rapidlywhen gold movements set in. But all the attempts tosubstitute other measures for changes in the discount rateas a means to " protect reserves " do not help, because it isthe necessity of " protecting " reserves rather than lettingthem go (i.e. using the conversion into gold as the propermethod of reducing internal circulation) not the methodsby which it has to be done, which is the evil. The onlyreal cure would be if the reserves kept were large enoughto allow them to vary by the full amount by which thetotal circulation of the country might possibly change; thatis, if the principle of Peel's Act of 1844 could be appliedto all forms of money, including in particular bankdeposits. I shall come back to this point in my last lec-ture. What I want to stress, however, is that in the yearsbefore the breakdown of the international gold standardthe attempts to make the supply of money of individualcountries independent of international gold movements


had already gone so far that not only had an outflow orinflow of gold often no effect on the internal circulationbut that sometimes the latter moved actually in the oppo-site direction. To " offset " gold movements, as wasapparently done by the Bank of England, * by replacingthe gold lost by the central bank by securities boughtfrom the market, is of course not to correct the defects ofthe mixed system, but to make the international standardaltogether ineffective.

One should probably say much more on this subject.But I am afraid I must conclude here. I hope that what

have said to-day has at least made one point clear whichI made yesterday; namely that many objections which are x^raised against the gold standard as we knew it, are notreally objections against the gold standard, or againstany international standard as such, but objections againstthe mixed system which has been in general vogue. Itshould be clear too that the main defect of this system S

V̂ was that it was not sufficiently international. Whether*'and how these defects can be remedied I can consider onlyat the end of this course. But before I can do this I shall

\ yet have to consider the more completely nationalist sys-Vterns which have been proposed.

1 Cf. Minutes of Evidence taken before the Committee on Financeand Industry, London, 1931, vol. I, Q. 363. Sir Ernest Harvey :" You will find if you look at a succession of Bank Returns thatthe amount of gold we have lost has been almost entirely replacedby an increase in the Bank's securities. "

LECTURE III

INDEPENDENT CURRENCIES

When the rates of exchange between currencies of dif-ferent countries are variable, the consequences whichwill follow from changes which under an internationalsystem would lead to flows of money from country tocountry, will depend on the monetary policies adopted bythe countries concerned. It is therefore necessary, beforewe can say anything about those effects, to consider theaims which will presumably guide the monetary policyof countries which have adopted an independent stan-dard. This raises immediately the question whether thereis any justification for applying any one of the principlesaccording to which we might think that the circulationin a closed system should be regulated, to a particularcountry or region which is part of the world economicsystem.

Now it should be evident that a policy of stabilization,whether it be of the general price level or the generallevel of money incomes is one thing if it be applied tothe whole of a closed system and quite another if thesame policy is applied to each of the separate regions intowhich the total system can be more or less arbitrarilydivided. In fact, however, this difficulty is generallyignored by the advocates of Monetary Nationalism, andit is simply assumed that the criteria of a good monetarypolicy which are applicable to a closed system are equallyvalid for a single country. We shall have to consider

36 INDEPENDENT CURRENCIES

later the theoretical problems here involved. But for themoment we can confine ourselves to an examination of theworking of the mechanism which brings about relativechanges in the value of the total money holdings of thedifferent nations when each nation follows indepen-dently the objective of stabilising its national price level,or income stream, or whatever it may be, irrespective ofits position in the international system.

The case which has figured most prominently in thesediscussions in recent years, and which is apparently sup-posed to represent the relative positions of England andthe United States, is that of two countries with unequalrates of technological progress, so that, in the one, costsof production will tend to fall more rapidly than in theother. Under a regime of fixed parities this would meanthat the fall in the prices of some products produced inboth countries could be faster than the fall in their costin the country where technological progress is slower,and that in consequence it would become necessary toreduce costs there by scaling down money wages, etc.The main advantage of a system of movable parities issupposed to be that in such a case the downward adjust-ment of wages could be avoided and equilibrium restoredby reducing the value of money in the one country rela-tive to the other country.

It is, however, particularly important in this connec-tion not to be misled by the fact that this argument isgenerally expressed in terms of averages, that is in termsof general levels of prices and wages. A change in thelevel of prices or of costs in one country relatively to thatof another means that, in consequence of changes in rela-tive costs, the competitive position of a particular indus-try or perhaps group of industries in the one country hasdeteriorated. In other words the lower prices in the onecountry will lead to a transfer of demand from the other

INDEPENDENT CURRENCIES 37

country to it. The case is therefore essentially similar tothat which we have been considering in the last lectureand it will be useful to discuss it in the same terms. Weshall therefore in the first instance again consider theeffects of a simple shift of demand if rates of exchangeare allowed to vary and if the monetary authorities ineach country aim either at stability of some national pricelevel, or—what amounts very much to the same thingfor our purpose—at a constant volume of the effectivemoney stream within the country. Only occasionally,where significant differences arise, I shall specially referto the case where the shift of demand has been inducedby unequal technological progress.

Now of course no monetary policy ca'n prevent theprices of the product immediately affected from fallingrelatively to the prices of other goods in the one country,and a corresponding1 rise taking place in the other.Nor can it prevent the effects of the change of the incomeof the people affected in the first instance from graduallyspreading. All it can do is to prevent this from leadingto a change in the total money stream in the country;that is it must see that there will be offsetting changes ofother prices so that the price level remains constant. Itis on this assumption that we conduct our investigations.For purposes of simplicity, too, I assume that at the outseta state of full employment prevails.

1 Where the shift of demand has been induced by a reductionof cost and a consequent fall of prices in the one country, thiswill only be a relative rise and will of course only partly coun-teract this fall in the price of the final product, but may bringabout an actual rise in the prices of the factors used in theirproduction.


It will be convenient to concentrate first on the countryfrom which demand has turned away and from whichunder an international monetary system there would inconsequence occur an outflow of money. But in thepresent case not only would a real outflow of money beimpossible, but it would also be contrary to the intentionsof the monetary authorities to sell additional quantitiesof foreign exchange against national money and to cancelthe national money so received. The monetary authori-ties might hold some reserves of foreign exchange to evenout what they regarded as merely temporary fluctuationsof exchange rates. But there would be no point in usingthem in the case of a change which they would have toregard as permanent. We can, therefore, overlook theexistence of such reserves and proceed as if only cur-rent receipts from abroad were available for outwardpayments.

On this assumption it is clear that the immediate effectof the adverse balance of payments will be that foreign ex-change rates will rise. But the full amount that importersused to spend on buying foreign exchange is not likelyto be spent on the reduced supply of foreign exchange;since with the higher price of imported goods some ofthe money which used to be spent on them will probablybe diverted to home substitutes.1 The foreign exchangeswill therefore probably rise less than in proportion to thefall in supply. But via the sale of foreign exchange at

1 The assumption that the demand for the commodities inquestion is elastic, that is that the total expenditure upon themwill be reduced when their prices rise and vice versa, will bemaintained throughout this discussion. To take at every stepthe opposite case into account would unduly lengthen the argu-ment without affecting the conclusion.


the higher rate those who continue to export successfullywill receive greater amounts of the national currency.For those whose sales abroad have not been unfavourablyaffected by the initial change in question this will meana net gain and the price of their products will correspon-dingly rise in terms of the national currency. And thosewhose exports have fallen in price will find that thisreduced price in terms of the foreign currency will nowcorrespond to a somewhat greater amount in the nationalcurrency than what they could obtain before the exchangedepreciation, although not as much as they receivedbefore the first change took place.

This impact effect of the rise of exchange rates on rela-tive prices in terms of the national currency will howeverbe temporary. The relative costs of the different quan-tities of the different commodities which are being pro-duced have not changed and it is not likely that they willgo on being produced in these quantities if their priceshave changed. Moreover all the changes in the directionof the money streams caused by the rise in exchange rateswill continue to work. More is being spent on homegoods, and this, together with the increased profitabilityof those export industries which have not been adverselyaffected by the initial change, will tend to bring about arise of all prices except those which are affected by thedecreased demand from the declining industry and fromthe people who draw their income from it.

It seems therefore that the argument in favour of depre-ciation in such cases is based on a too simplified pictureof the working of the price mechanism. In particular itseems to be based on the assumption (underlying muchof the classical analysis of these problems) that relativeprices within each country are uniquely determined by(constant) relative cost. If this were so, a proportionalreduction of all prices in a country relatively to those in

40 INDEPENDENT CURRENCIESthe rest of the world would indeed be sufficient to restoreequilibrium. In fact, however, there can be little doubtthat the changes in the relative quantities of goods to beproduced by the different industries which will becomenecessary in consequence of the initial change, can bebrought about only by changes in the relative prices andthe relative incomes of the different kinds of resourceswithin the country.

Without following the effects in all their complicateddetail it must be clear that the ultimate result of depre-ciation can only be that, instead of prices and incomesin the industry originally affected falling to the fullextent, a great many other prices and incomes will haveto rise to restore the proportions appropriate to cost con-ditions and the relative volume of output now required.Even disregarding the absolute height of prices, thefinal positions will not be the same as that which wouldhave beeen reached if exchanges had been kept fixed;because in the course of the different process of transitionall sorts of individual profits and losses will have beenmade which will affect that final position. But roughlyspeaking and disregarding certain minor differences, itcan be said that the same change in relative prices which,under fixed exchanges, would have been brought aboutby a reduction of prices in the industry immediatelyaffected is now being brought about largely by a corres-ponding rise of all other prices.

Two points, however, need special mention. One isthat the decrease of the comparative advantage of theexport industry originally affected cannot be changed inthis way; and that to this extent a contraction of theoutput of this industry will remain unavoidable. Theother is that at least in certain respects the process whichbrings about the rise in prices will be of a definitely infla-tionary character. This will show itself partly by some


industries becoming temporarily more profitable so thatthere will be an inducement to expand production there,although this increase will soon be checked and evenreversed by a rise in cost; and partly by some of the cashreleased by importers finding its way, via the repaymentof loans, to the banks, who will be able to increase theirloans to others and, in order to find lenders, will relax theterms on which they will be ready to lend. But this toowill prove a merely temporary effect, since as soon ascosts begin generally to rise it will become apparent thatthere are really no funds available to finance additionalinvestments. In this sense the effects of this redistribu-tion of money will be of that self-reversing characterwhich is typical of monetary disturbances. This leads,however, already to the difficult question of what consti-tutes an inflation or deflation within a national area.But before we can go on to this it is necessary to considerwhat happens in the converse case of the country whichhas been put in a more favourable condition by thechange.

Let us first assume that the monetary authorities hereas in the other country aim at a constant price levekanda constant income stream. The industry which directlybenefits from the initial shifts in demand will then findthat, because of the fall of foreign exchanges, the increaseof their receipts in terms of the national currency willnot be as large as would correspond to the increase oftheir sales in terms of foreign money, while the otherexport industries will see their receipts actually reduced.Similarly those home industries whose products competewith imports which are now cheaper in terms of thenational currency will have to lower their prices and will


find their incomes reduced. In short, if the quantity ofmoney in the country, or the price level, is kept constant,the increase of the aggregate value of the products of oneindustry due to a change in international demand willmean that there has to be a compensating reduction ofthe prices of the products of other industries. Or, in otherwords, part of the price reduction which under a regimeof stable exchanges would have been necessary in theindustry and in the country from which demand has tur-ned away, will under a regime of independent currenciesand national stabilisation have to take place in the coun-try towards which demand has turned, and in industrieswhich have not been directly affected by the shift indemand.

This at least should be the case if the principle of natio-nal stabilization were consistently applied. But it is ofcourse highly unlikely that it ever would be so applied.That in order to counteract the effects of a severe fall ofprices in one industry in a country other prices in thecountry should be allowed to rise, appears fairly plau-sible. But that in order to offset a rise of prices of theproducts of one industry which is due to an increase ininternational demand, prices in the other industriesshould be made to fall sounds far less convincing. I findit difficult to imagine the President of a Central Bankexplaining that he has to pursue a policy which meansthat the prices of many home industries have to be redu-ced, by pointing out that an increase of internationaldemand has led to an increase of prices in an importantexport industry, and it seems fairly certain what wouldhappen to him if he tried to do so.

Indeed, if we take a somewhat more realistic point ofview, there can be little doubt what will happen. While,in the country where in consequence of the changes ininternational demand some prices will tend to fall the


price level will be kept stable, it will certainly be allowedto rise in the country which has been benefited by thesame shift in demand. It is not difficult to see what thisimplies if all countries in the world act on this principle.It means that prices would be stabilized only in that areawhere they tend to fall lowest relatively to the rest of theworld, and that all further adjustments are brought aboutby proportionate increases of prices in all other countries.The possibilities of inflation which this offers if the worldis split up into a sufficient number of very small separatecurrency areas seem indeed very considerable. And why,if this principle is once adopted, should it remain con-fined to average prices in particular national areas?Would it not be equally justified to argue that no priceof any single commodity should ever be allowed to falland that the quantity of money in the world should beso regulated that the price of that commodity which tendsto fall lowest relatively to all others should be kept stable,and that the prices of all other commodities would beadjusted upwards in proportion ? We only need toremember what happened, for instance, a few years agoto the price of rubber to see how such a policy wouldsurpass the wishes of even the wildest inflationist. Per-haps this may be thought an extreme case. But, oncethe principle has been adopted, it is difficult to see howit could be confined to " reasonable " limits, or indeedto say what " reasonable " limits are.

But let us disregard the practical improbability that apolicy of stabilization will be followed in the countrieswhere, with stable exchanges, the price level would rise,as well as in the countries where in this case it wouldhave to fall. Let us assume that, in the countries which


benefit from the increase of the demand, the prices ofother goods are actually lowered to preserve stability ofthe national price level and that the opposite action willbe taken in the countries from which demand has turnedaway. What is the justification and significance of sucha policy of national stabilization?

Now it is difficult to find the theoretical case for natio-nal stabilization anywhere explicitly argued. It is usuallyjust taken for granted that any sort of policy whichappears desirable in a closed system must be equally bene-ficial if applied to a national area. It may therefore bedesirable before we go on to examine its analytical jus-tification, to trace the historical causes which havebrought this view to prominence. There can be littledoubt that its ascendancy is closely connected with thepeculiar difficulties of English monetary policy between1925 and 1931. In the comparatively short space of the sixyears during which Great Britain was on a gold standardin the post war period, it suffered from what is knownas overvaluation of the pound. Against all the teachingof " orthodox " economics—already a hundred yearsbefore Ricardo had expressly stated that he " shouldnever advise a government to restore a currency, whichwas depreciated 30 p. c , to par "*—in 1925 the Britishcurrency had' been brought back to its former gold value.In consequence, to restore equilibrium, it was necessaryto reduce all prices and costs in proportion as the valueof the pound had been raised. This process, particularlybecause of the notorious difficulty of reducing moneywages, proved to be very painful and prolonged. Itdeprived England of real participation in the boom whichled up to the crisis of 1929, and, in the end, its results

1 In a letter to John Wheatley, dated September 18, 1821, reprin-ted in Letters of David Ricardo to Hutches Trower and Others,edited by J. Bonar and J. Hollander, Oxford, 1899, p. 160.


proved insufficient to secure the maintenance of the res-tored parity. But all this was not due to an initial shiftin the conditions of demand or to any of the causes whichmay affect the condition of a particular country understable exchanges. It was an effect of the change in theexternal value of the pound. It was not a case wherewith given exchange rates the national price or cost struc-ture of a country as a whole had got out of equilibriumwith the rest of the world, but rather that the change inthe parities had suddenly upset the relations between allprices inside and outside the country.

Nevertheless this experience has created among manyBritish economists a curious prepossession with the rela-tions between national price- and cost- and particularlywage-levels, as if there were any reason to expect thatas a rule there would arise a necessity that the priceand cost structure of one country as a whole shouldchange relatively to that of other countries. And thistendency has received considerable support from thefashionable pseudo-quantitative economics of averageswith its argument running in terms of national " pricelevels ", " purchasing power parities '*, " terms of trade ",the " Multiplier ", and what not.

The purely accidental fact that these averages are gene-rally computed for prices in a national area is regarded asevidence that in some sense all prices of a country couldbe said to move together relatively to prices in othercountries. * This has strengthened the belief that there

1 The fact that the averages of (more or less arbitrarily selected)groups of prices move differently in different countries does ofcourse in no way prove that there is any tendency of the pricestructure of a country to move as a whole relatively to prices inother countries. It would however be a highly interesting subjectfor statistical investigation, if a suitable technique could be devised,to see whether, and to what extent, such a tendency exis-ted. Such an investigation would of course involve a comparison


is some peculiar difficulty about the case where " the "price level of a country had to be changed relatively to itsgiven cost level and that such adjustment had better beavoided by manipulations of the rate of exchange.

Now let me add immediately that of course I do notwant to deny that there may be cases where some changein conditions might make fairly extensive reductions ofmoney wages necessary in a particular area if exchangerates are to be maintained, and that under present con-ditions such wage reductions are at best a very painfuland long drawn out process. At any rate in the case ofcountries whose exports consist largely of one or a fewraw materials, a severe fall in the prices of these productsmight create such a situation. What I want to suggest,however, is that many of my English colleagues, becauseof the special experience of their country in recent times,have got the practical significance of this particular casealtogether out of perspective : that they are mistaken inbelieving that by altering parities they can overcomemany of the chief difficulties created by the rigidity ofwages and, in particular, that by their fascination withthe relation between " the " price level and " the " costlevel in a particular area they are apt to overlook themuch more important consequences of inflation anddeflation. *

not only of some mean value of the price changes in differentcountries, but of the whole frequency distribution of relative pricechanges in terms of some common standard. And it should besupplemented by similar investigations of the relative movementsof the price structure of different parts of the same country.

1 The propensity of economists in the Anglo-Saxon countries toargue exclusively in terms of national price and wage levels isprobably mainly due to the great influence which the writings ofProfessor Irving Fisher have exercised in these countries. Anothertypical instance of the dangers of this approach is the well-knowncontroversy about the reparations problem, where it was left toProfessor Ohlin to point out against his English opponents that


As I have already suggested at an earlier point, the diffe-rence of opinion here rests largely on a difference of viewon the meaning and consequence of inflation and defla-tion, or rather in the importance attached to two sortsof effects which spring from changes in the quantity ofmoney. The one view stresses what I have called beforethe self-reversing character of the effects of monetarychanges. It emphasizes the misdirection of productioncaused by the wrong expectations created by changes inrelative prices which are necessarily only temporary, ofwhich the most conspicous is of course the trade cycle.The other view emphasizes the effects which are due tothe rigidity of certain money prices, and particularlywages. Now the difficulties which arise when moneywages have to be lowered can not really be called mone-tary disturbances; the same difficulties would arise ifwages were fixed in terms of some commodity. It is onlya monetary problem in the sense that this difficulty mightto some extent be overcome by monetary means whenwages are fixed in terms of money. But the problem leftunanswered by the authors who stress this second aspectis whether the difficulty created by the rigidity of moneywages can be overcome by monetary adjustments withoutsetting up new disturbances of the first kind. And thereare in fact strong reasons to believe that the two aims ofavoiding so far as possible downward adjustments ofwages and preventing misdirections of production maynot always he reconcilable.

This difference in emphasis is so important in connec-tion with the opinions about what are the appropriate

what mainly mattered was not so much effects on total price levelshut rather the effects on the position of particular industries.


principles of national monetary policy because, if onethinks principally in terms of the relation of prices togiven wages and particularly if one thinks in terms of na-tional wage " levels ", one is easily led to the conclusionthat the quantity of money should be adjusted for eachgroup of people among whom a given system of contractsexists. (To be consistent, of course, the argument shouldbe applied not only to countries but also to particularindustries, or at any rate to " non-competing groups " ofworkers in each country.) On the other hand, there isno reason why one should expect the self-reversing effectsof monetary changes to be connected with the change ofthe quantity of money in a particular area which is partof a wider monetary system. If a decrease or increase ofdemand in one area is offset by a corresponding changein demand in another area, there is no reason why thechanges in the quantity of money in the two areas shouldin any sense misguide productive activity. They are sim-ply manifestations of an underlying real change whichworks itself out through the medium of money.

To illustrate this difference let me take a statement ofone of the most ardent advocates of Monetary Nationalism,Mr. R. F. Harrod of Oxford. Mr. Harrod is not unfami-liar with what I have called the self-reversing effects ofmonetary changes. At any rate in an earlier publica-tion he argued that "if industry is stimulated to go for-ward at a pace which cannot be maintained, you arebound to have periodic crises and depressions " *. Yet

1 The International Gold Problem, Edited by the Royal InstituteOf International Affairs. London, 1931, p. 29. Cf. also, in thelight of this statement, the remarkable passage in the same author'sInternational Economics (London, 1933), p. 150, where it is argued,that " the only way to avoid a slump is to engineer a boom "although only two lines later a boom is still " defined as an in-crease in the rate of output which cannot be maintained in thelong period w.


for some reason he seems to think that these misdirec-tions of industry will occur even when the changes inthe quantity of money of a particular country take placein the course of the normal redistributions of money bet-ween countries. In his International Economics thereappears the following remarkable passage which seemsto express the theoretical basis, or as I think the fallacy,underlying Monetary Nationalism more clearly than anyother statement I have yet come across. Mr. Harrod isdiscussing the case of unequal economic progress in dif-ferent countries with a common standard and concludesthat " the less progressive countries would thus be afflictedwith the additional inconvenience of a deflatory monetarysystem. Inflation would occur just where it is most dan-gerous, namely in the rapidly advancing countries. Thisobjection appears in one form or another in all projectsfor a common world money ". r And the lesson whichMr. Harrod derives from these considerations is that " thecurrencies of the more progressive countries must bemade to appreciate in terms of the others ". *

It is interesting to inquire in what sense inflation anddeflation are here represented as additional inconve-niences, superimposed, as it were, on the difficultiescreated by unequal economic progress. One might thinkat first that what Mr. Harrod has in mind are the extradifficulties caused by the secondary expansions and con-tractions of credit which are made necessary by the natio-nal reserve systems which I have analyzed in an earlierlecture. But this interpretation is excluded by the expressassertion that this difficulty appears under all forms of acommon world money. It seems that the terms inflationand deflation are here used simply as equivalents to

* International Economics (Cambridge Economic HandbooksVIII), London, 1933, p. 170.

* Ibid., p. 174.


increases and decreases of money demand relatively togiven costs. In this sense the terms could equally beapplied to shifts in demand between different industriesand would really mean no more than a change in demandrelatively to supply. But the objection to this is not onlythat the terms inflation and deflation are here unneces-sarily applied to phenomena which can be described insimpler terms. It is rather whether in this case there isany reason to expect any of the special consequenceswhich we associate with monetary disturbances, that is,whether there really is any " additional inconvenience "caused by monetary factors proper. We might askwhether in this case there will be any of the peculiarself-reversing effects which are typical of purely mone-tary causes; in particular whether " inflation " as usedhere with reference to the increase of money in one coun-try at the expense of another, " stimulates industry to goforward at a pace which cannot be maintained "; andwhether deflation in the same sense implies a temporaryand avoidable contraction of production.

The answer to these questions is not difficult. Weknow that the really harmful effects of inflation and defla-tion spring, not so much from the fact that all priceschange in the same direction and in the same proportion,but from the fact that the relation between individualprices changes in a direction which cannot be main-tained; or in other words that it temporarily brings abouta distribution of spending power between individualswhich is not stable. We have seen that the internationalredistributions of money are part of a process which atthe same time brings about a redistribution of relativeamounts of money held by the different individuals ineach country, a redistribution within the nation/whichwould also have to come about if there were no interna-tional money. The difference, however, in the latter


case, the case of free currencies, is that here first the rela-tive value of the total amounts of money in each countryis changed and that the process of internal redistributiontakes places in a manner different from that which wouldoccur with an international monetary standard. We haveseen before that the variation of exchange rates will initself bring about a redistribution of spending power inthe country, but a redistribution which is in no waybased on a corresponding change in the underlying realposition. There will be a temporary stimulus to parti-cular industries to expand, although there are no groundswhich would make a lasting increase in output possible.In short, the successive changes in individual expendi-ture and the corresponding changes of particular priceswill not occur in an order which will direct industry fromthe old to the new equilibrium position. Or, in otherwords, the effects of keeping the quantity of money in aregion or country constant when under an internationalmonetary system it would decrease are essentially infla-tionary, while to keep it constant if under an inter-national system it would increase at the expense of othercountries would have effects similar to an absolutedeflation.

I do not want to suggest that the practical importanceof the deflationary or inflationary effects of a policy ofkeeping the quantity of money in a particular areaconstant is very great. The practical arguments whichto me seem to condemn such a policy I have already dis-cussed. The reason why I wanted at least to mentionthis more abstract consideration is that, if it is correct, itshows particularly clearly the weakness of the theoreticalbasis of Monetary Nationalism. The proposition that theeffects of keeping the quantity of money constant in aterritory where with an international currency it would


decrease are inflationary and vice-versa1 is of coursedirectly contrary to the position on which MonetaryNationalism is based. Far from admitting that changesin the relative money holdings of different nations whichgo parallel with changes in their share of the world'sincome are harmful, we believe that such redistributionsof money are the only way of effecting the change in realincome with a minimum of disturbance. And to speakin connection with such changes of national inflation ordeflation can only lead to a serious confusion ofthought.2

Before I leave this subject I should like to supplementthese theoretical reflections by a somewhat more prac-tical consideration. While the whole idea of a monetarypolicy directed to adjust everything to a " given " wagelevel appears to me misconceived on purely theoreticalgrounds, its consequences seem to me to be fantastic ifwe imagine it applied to the present world where this sup-posedly given wage level is at the same time the subject

1 Whithout giving disproportionate space to what is perhaps asomewhat esoteric theoretical point it is not possible to give herea complete proof of this proposition. A full discussion of thecomplicated effects would require almost a separate chapter. Buta sort of indirect proof may be here suggested. It would probablynot be denied that if without any other change the amount ofmoney in one currency area were decreased by a given amountand at the same time the amount of money in another currencyarea increased by a corresponding amount, this would have defla-tionary effects in the first area and inflationary effects in thesecond. And most economists (the more extreme monetary natio-nalists only excepted) would agree that no such effects wouldoccur if these changes were made simultaneous with correspondingchanges in the relative volume of transactions in the two countries.From this it appears to follow that if such a change in the relativevolume of transactions in the two countries occurs but the quan-tity of money in each country is kept constant, this must havethe effect of a relative inflation and deflation respectively.

* See on this point also L. ROBBINS, Economic Planning andInternational Order, 1937, pp. 281 et seq.


of political strife. It would mean that the whole mecha-nism of collective wage bargaining would in the futurebe used exclusively to raise wages, while any reduction—even if it were necessary only in one particular indus-try—would have to be brought about by monetary means.I doubt whether such a proposal could ever have beenseriously entertained except in a country and in a periodwhere labour has been for long on the defensive.1 It isdifficult to imagine how wage negotiations would becarried on if it became the recognised duty of the mone-tary authority to offset any unfavourable effect of a risein wages on the competitive position of national indus-tries on the world market. But of one thing we can pro-bably be pretty certain : that the working class would notbe slow to learn that an engineered rise of prices is noless a reduction of wages than a deliberate cut of moneywages, and that in consequence the belief that it is easierto reduce by the round-about method of depreciation thewages of all workers in a country than directly to reducethe money wages of those who are affected by a givenchange, will soon prove illusory.

1 It is interesting to note that those countries in Europe whereup to 1929 wages had been rising relatively most rapidly were onthe whole those most reluctant to experiment with exchangedepreciation. The recent experience of France seems also to sug-gest that a working class government may never be able to useexchange depreciation as an instrument to lower real wages.

LECTURE IV

INTERNATIONAL CAPITAL MOVEMENTS

For the purposes of this lecture, by international capitalmovements I shall mean the acquisition of claims on per-sons or of rights to property in one country by personsin another country, or the disposal of such claims or pro-perty rights in another country to people in that country.This definition is meant to exclude from capital move-ments the purchase and sale of commodities which passfrom one country to the other at the same time as theyare paid for and change their owners. But it alsoexcludes any net movement of gold (or other interna-tional money) in so far as these movements are paymentsfor commodities or services received (or " unilateral "payments) and therefore involve a transfer of ownershipin that money without creating a new claim from onecountry to the other. This is of course not the only pos-sible definition of capital movements, and strong argu-ments could be advanced in favour of a more comprehen-sive definition, which in effect would treat every transferof assets from country to country as a capital movement.The reason which leads me to adopt here the former defi-nition is that only on that definition it is possible to dis-tinguish between those items in international transactionswhich are, and those which are not, capital items.

The first kind of capital item of this sort and the onewhich will occupy us in this lecture more than any otheris the acquisition, or sale, of amounts of the national

INTERNATIONAL CAPITAL MOVEMENTS 55

money of one country by inhabitants of the other. x Theform which this kind of transaction to-day predominantlytakes is the holding of balances with the banks of onecountry on the part of banks and individuals in the othercountry. Such balances will to some extent be held evenif there is a safe and stable international standard, since,rather than actually send money, it will as a rule becheaper for the banks to provide out of such balancesthose of their customer's requirements which arise outof the normal day to day differences between paymentsand receipts abroad. And if it is possible to hold suchbalances either in the form of interest-bearing depositsor in the form of bills of exchange, there will be a stronginducement to hold such earning assets as substitutes forthe sterile holdings of international money. It was inthis way that what is called the gold exchange standardtended more and more to supplant the gold standardproper. In the years immediately preceding 1931 thisassumed very great significance.

If there exists a system of fluctuating exchanges, or asystem where people are not altogether certain about themaintenance of the existing parities, these balancesbecome even more important. There are two new ele-ments which enter in this case. In the first place it willthen no longer be sufficient if banks and others who owedebts in different currencies keep one single liquidityreserve against all their liabilities. It will become neces-

1 This is not to be interpreted as meaning that I subscribe tothe view that all money is in some sense a " claim M. The state-ment in the text applies strictly only to credit money and parti-cularly to bank deposits, which will be mainly considered in whatfollows. But it would not apply to the acquisition of gold byforeigners for export. The gold coins so acquired would therebycease to be " national " money in the sense in which this termis here used, that is, they would not be assets belonging to thecountry where they have been issued.

56 INTERNATIONAL CAPITAL MOVEMENTS

sary for them to keep separate liquid assets in each of thedifferent currencies in which they owe debts, and toadjust them to the special circumstances likely to affectliabilities in each currency. We get here new artificialdistinctions of liquidity created by the multiplicity of cur-rencies and involving all the consequential possibilitiesof disturbances following from changes in what is nowcalled " liquidity preference ". Secondly there will bethe chance of a gain or loss on these foreign balances dueto changes in the rates of exchange. Thus the anticipa-tion of any impending variation of exchange rates willtend to bring about temporary changes of a speculativenature in the volume of such balances. Whether thesetwo kinds of motives must really be regarded as different,or whether they are better treated as essentially the same,there can be no doubt that variability of exchange ratesintroduces a new and powerful reason for short termcapital movements, and a reason which is fundamentallydifferent from the reasons which exist under a well-secured international standard.

Foreign bank balances and other holdings of foreignmoney are of course only part, although probably themost important part, of the volume of short term foreigninvestment. It is here that the impact effect of anychange in international indebtedness arising out of cur-rent transactions will show itself; and it is here that therewill be the most ready response to changes in the rela-tive attractiveness of holding assets in the different coun-tries. Once we go beyond this field it becomes ratherdifficult to say what can properly be called movements ofshort term capital. In fact, with the exception of non-funded long-term loans almost any form of internationalinvestment may have to be regarded as short term invest-ment, including in particular all investments in marke-


table securities. * But for the monetary problems withwhich we are here concerned it is mainly the short termcredits which are of importance, because it is here that wehave to deal with large accumulated funds which are aptto change their location at comparatively slight provo-cation. Compared with these " floating " funds, the sup-ply of capital for long term investment, limited as it willbe to a certain part of new savings, will be relativelysmall.

Now the chief question which we shall have to consideris the question to what extent under different monetarysystems international capital movements are likely tocause monetary disturbances, and to what extent and bywhat means it may be possible to prevent such distur-bances. It will again prove useful if we approach thistask in three stages, beginning with a consideration ofthe mechanism and function of international capitalmovements under a homogeneous standard. Then we shallgo on to inquire how this mechanism and the effects aremodified if we have " mixed " currency systems orga-nized on the national reserve principle but with fixedexchange rates. And finally we shall have to see whatwill be the effects of the existence of variable exchangerates and the way in which fluctuations of the exchangeand capital movements mutually influence one another.

If exchange rates were regarded as invariably fixed weshould expect capital movements to be guided by no other

1 Even the intentions of the lender or investor would hardlyprovide a sufficient criterion for a distinction between what areshort and what are long term capital movements, since it mayvery well be clear in a particular case to the outside observer thatcircumstances will soon lead the investors to change their inten-tions.


considerations except expected net yield, including ofcourse adjustments which will have to be made for thedifferent degrees of risk inherent in the different sorts ofinvestments. This does not mean that there would notbe frequent changes in the flow of capital from countryto country. There might of course be a permanent ten-dency on the part of one country to absorb part of thecurrent savings of another at terms more favourable thanthose at which these savings could be invested in thecountry were they are made. Quite apart from theseflows of capital for more or less permanent investmenthowever, there would be periodic or occasional shortterm lending to make up for temporary differences bet-ween imports and exports of commodities and services.

Now there is of course no reason why exports andimports should move closely parallel from day to day oreven from month to month. If in all transactions pay-ment had to be made simultaneously with the delivery ofthe goods, this would mean, in external trade no lessthan in internal, a restriction of the possible range oftransactions similar in kind to what would occur if alltransaction had to take the form of barter. The possibi-lity of credit transactions, the exchange of present goodsagainst future goods, greatly widens the range of advan-tageous exchanges. In international trade it means inparticular that countries may import more than theyexport in some seasons because they will export morethan they import during other seasons. Whether this ismade possible by the exporter directly crediting theimporter with the price, or whether it takes place bysome credit institution in either country providing themoney, it will always mean that the indebtedness of theimporting country to the exporting increases tempora-rily, that is, that net short term lending takes place.

At this point it is necessary especially to be on guard


against a form of stating these relations which suggeststhat short term lending is made necessary by, or is inany sense a consequence of a passive balance of trade—thatthe loans are made so to speak with the purpose of cover-ing a deficit in the balance of trade. We shall get a morecorrect picture if we think of the great majority of theindividual transactions in both ways being credit transac-tions so that it is the excess lending in one directionduring any given period which has made possible a cor-responding excess of exports in the same direction. Ifwe look on the whole process in this way we can see howconsiderable a part of trade is only made possible by shortterm capital movements. We can see also how misleadingit may be to think of capital movements as exclusivelydirected by previous changes in the relative rates of inte-rest in the different money markets. What directs the useof the available credit and therefore decides in what direc-tion the balance of indebtedness will shift at a particularmoment is in the first instance the relation between pricesin different places. It is of course true that where eachcountry habitually finances its exports and borrows itsimports, any absolute increase of exports will tend tobring about an increase in the demand for loans andtherefore a rise in the rate of interest in the exportingcountry. But in such a case the rise in the rate of interestis rather the effect of this country lending more abroad,than a cause of a flow of capital to the country. Andalthough this rise in money rates may lead to a flow offunds in the reverse direction, that will be more a signthat the main mechanism for the distribution of fundsworks imperfectly than a part of this mechanism. Thereis no more reason to say that the international redistri-bution of short-term capital is brought about by changesin the rates of interest in the different localities than therewould be for saying that the seasonal transfers of funds


from say agriculture to coal mining are brought aboutby a fall of the rate of interest in agriculture and a rise incoal mining or vice versa.

Changes in short term international indebtedness musttherefore be considered as proceeding largely concur-rently with normal fluctuations in international trade;and only certain remaining balances will be settled by aflow of funds, largely of an inter-bank character, inducedby differences in interest rates to be earned. It is of coursenot to be denied that, apart from changes in internationalindebtedness which are more directly connected withinternational trade, there may also be somewhat suddenand considerable flows of funds which may be causedeither by the sudden appearance of very profitable oppor-tunities for investment, or by some panic which causesan insistent demand for cash. In this last case indeedit is true that the flow of short term funds may transmitmonetary disturbances to parts of the world whichhave nothing to do with the original cause of thedisturbance, as say a war-scare in South-America mightconceivably lead to a general rise in interest rates in Lon-don. But, apart from such special cases, it is difficult tosee how under a homogeneous international standard,capital movements, and particularly short term capitalmovements, should be a source of instability or lead toany changes in productive activity which are not justi-fied by corresponding changes in the real conditions.

3

This conclusion has, however, to be somewhat modi-fied if, instead of a homogeneous international currency,we consider a world consisting of separate national mone-tary and banking systems, even if we still leave the pos-sibility of variations in exchange rates out of account.


It is of course a well-known fact that one of the mainpurposes of changes in the discount rate of central banksis to influence the international movements of short termcapital.a A central bank which is faced with an outflowof gold will raise its discount rate in the hope that byattracting short term credits it will offset the gold out-flow. To the extent that it succeeds it will postpone thenecessity of more drastic credit contraction at home,and—if the cause of the adverse balance of trade is tran-sitory—it may perhaps altogether avoid it. But it is byno means evident that it will attract the funds just fromwhere the gold would tend to flow, and it may well bethat it only passes on the necessity of credit contraction toanother country. And if for some reason all or the majo-rity of central banks should at a particular moment feelthat they ought to becorile more liquid and for this pur-pose raise their discount rates, the sole effect will be akind of general tug-of-war in which all central banks,trying to prevent an outflow of funds and if possible to at-tract funds, only succeed in bringing about a violent con-traction of credit at home. But although the fact that cen-tral banks react to all major gold movements withchanges in the rate of discount may mean that changes inthe volume and direction of short term credits will be morefrequent and violent if we have a number of banking sys-tems organized on national lines, it is again not the factthat the system is international, but rather that is createsimpediments ot the free international flow of funds whichmust be regarded as responsible for these disturbances.

Again we must be careful not to ascribe this difficulty

1 If this effect was disregarded in the discussion of changes inthe discount rates in the two preceeding lectures, this was doneto make the effects discussed there stand out more clearly; hutthis must not be taken to mean that this effect on capital move-ments is not, at any rate in the short run, perhaps the most im-portant effect of these changes.


to the existence of central banks in particular, althoughin a sense the growth of the sort of credit structure towhich they are due was only made possibe by the exis-tence of some such institutions. The ultimate source ofthe difficulty is the differentiation between moneys ofdifferent degrees of acceptability or liquidity, the exis-tence of a structure consisting of superimposed layers ofreserves of different degrees of liquidity, which makes themovement of short term money rates, and in consequencethe movement of short term funds, much more dependenton the liquidity position of the different financial institu-tions than on changes in the demand for capital for realinvestment. It is because with " mixed " national mone-tary systems the movements of short term funds are fre-quently due, not to changes in the demand for capital forinvestment, but to changes in the demand for cash asliquidity reserves, that short term international capitalmovements have such or bad reputation as causes a mone-tary disturbances. And this reputation is not altogetherundeserved.

But now the question arises whether this defect can beremoved not by making the medium of circulation in thedifferent countries more homogeneous, but rather, as theMonetary Nationalists wish, by severing even the remain-ing tie between the national currencies, the fixed paritiesbetween them. This question is of particular importancesince the idea that the national monetary authoritiesshould never be forced by an outflow of capital to takeany action which might unfavourably affect economicactivity at home is probably the main source of thedemand for variable exchanges. To this question there-fore we must now turn.


4

The chief questions which we shall have to considerhere are three : will the volume of short term capitalmovement be larger or smaller when there exists uncer-tainty about the future of exchange rates ? Are the natio-nal monetary authorities in a position either to preventcapital movements which they regard as undesirable, orto offset their effects? And, finally, what further mea-sures, if any, are necessary if the aims of such a policyare to be consistently followed?

We have already partly furnished the answer to thefirst question. Although the contrary has actually beenasserted, I am altogether unable to see why under aregime of variable exchanges the volume of short termcapital movements as well as the frequency of changesin their direction should be anything but greater.x Everysuspicion that exchange rates were likely to change in thenear future would create an additional powerful motivefor shifting funds from the country whose currency was

1 The only argument against this view which I find at all intelli-gible is that, under the gold standard, movements to one of thegold points will create a certain expectation that the movementwill soon be reversed and thus provides a special inducement tospeculative shifts of funds. But while this is perfectly true, itonly shows that the defects of the traditional gold standard weredue to the fact that it was not a homogeneous international cur-rency. If the same arrangements applied to international asto infranational payments the problem would disappear. Thiswould be the case either if within the country as much as betweencountries the costs of transfers of money were not borne by someinstitution like the central banks and consequently (as in the UnitedStates before the establishment of the Federal Reserve System)rates of exchange between the different towns were allowed tofluctuate, and if at the same time gold were freely obtainable nearthe frontier as well as in the capital, or on the other hand, if thesystem of par clearance were applied to international as well asnational payments. On the last point compare below, lecture V,p. 84.


likely to fall or to the country whose currency waslikely to rise. I should have thought that the experienceof the whole post-war period and particularly of the lastfew years had so amply confirmed what one might haveexpected a priori that there could he no reasonable doubtabout this. * There is only one point which perhaps stilldeserves to be stressed a little further. Where the possiblefluctuations of exchange rates are confined to narrowlimits above and below a fixed point, as between the twogold points, the effect of short term capital movementswill be on the whole to reduce the amplitude of the actualfluctuations, since every movement away from the fixedpoint will as a rule create the expectation that it willsoon be reversed. That is, short term capital movementswill on the whole tend to relieve the strain set up by theoriginal cause of a temporarily adverse balance of pay-ments. If exchanges, however, are variable, the capitalmovements will tend to work in the same direction as theoriginal cause and thereby to intensify it. This meansthat if this original cause is already a short term capitalmovement, the variability of exchanges will tend to mul-tiply its magnitude and may turn what originally mighthave been a minor inconvenience into a major distur-bance.

Much more difficult is the answer to the second ques-tion : can the authorities control these movements; sincewhat the monetary authorities can achieve in a particular

1 Since it is being more and more forgotten that the periodbefore 1931 was, on pre-war standards, already one of markedinstability — and uncertainty about the future — of exchangerates, it is perhaps worth stressing that in particular the accumu-lation of foreign balance* in London during that period was almostentirely a consequence of the fact that Sterling was regarded asrelatively the most safe of the European currencies. Cf. on thisT. E. GREGORY, The Gold Standard and it$ Future, Third edition,1984, pp. 48 et seq.


direction will largely depend on what other consequencesof their action they are willing to put up with. In theparticular case the question is mainly whether theywould be willing to let exchange rates fluctuate to anydegree or whether they would not feel that althoughmoderate fluctuations of exchange rates were not worththe cost of preventing them, yet they must not be allowedto exceed certain limits, since the unsettling effects fromlarge fluctuations would be worse than the measures bywhich they could be prevented. In practice we mustprobably assume that even if the authorities are preparedto allow a slow and gradual depreciation of exchanges,they would feel bound to take strong action to counteractit as soon as it threatened to lead to a flight of capital ora strong rise of prices of imported goods.

The theory that by keeping exchange rates flexible acountry could prevent dear money abroad from affectinghome conditions is of course not a new one. It was forinstance argued by the opponents of the introduction ofthe gold standard in Austria in 1892 that the paper stan-dard insulated and protected Austria from disturbancesoriginating on the world markets. But I doubt whetherit has ever been carried quite as far as by some of ourcontemporary Monetary Nationalists, for instance Mr. Har-rod, who declared that he could not accept exchangestabilisation " if thereby a country is committed to aninterior monetary policy which involves raising the bankrate of interest ".* The modern idea apparently is thatnever under any circumstances must an outflow of capi-tal be allowed to raise interest rates at home, and the advo-cates of this view seem to be satisfied that if the centralbanks are not commited to maintain a particular parity

1 Cf. Report of the Proceedings of the Meeting of Economists heldat the Antwerp Chamber of Commerce on July the 11th, 12th, and13th 1936, published by the Antwerp Chamber of Commerce, p. 107.


they will have no difficulty either in preventing an out-flow of capital altogether or in offsetting its effect bysubstituting additional bank credit for the funds whichhave left the country.

It is not easy to see on what this confidence is founded.So long as the outward flow of capital is not effectivelyprevented by other means, a persistent effort to keep inte-rest rates low can only have the effect of prolonging thistendency indefinitely and of bringing about a continuousand progressive fall of the exchanges. Whether the out-ward flow of capital starts with a withdrawal of balancesheld in the country by foreigners, or with an attempt onthe parts of nationals of the country to acquire assetsabroad, it will deprive banking institutions at home offunds which they were able to lend, and at the same timelower the exchanges. If the central bank succeeds inkeeping interest rates low in the first instance by substi-tuting new credits for the capital which has left the coun-try, it will not only perpetuate the conditions underwhich the export of capital has been attractive; the effectof capital exports on the rates of exchange will, as wehave seen, tend to become self-inflammatory and a" flight of capital " will set in. At the same time the riseof prices at home will increase the demand for loansbecause it means an increase in the " real " rate of profit.And the adverse balance of trade which must necessarilycontinue while part of the receipts from exports is usedto repay loans or to make new loans abroad, means thatthe supply of real capital and therefore the " natural "or " equilibrium " rate of interest in the country willrise. It is clear that under such conditions the centralbank could not, merely by keeping its discount rate low,prevent a rise of interest rates without at the same timebringing about a major inflation.


If this is correct it would be only consistent if the advo-cates of Monetary Nationalism should demand that mone-tary policy proper should be supplemented by a strict con-trol of the export of capital. If the main purpose of mone-tary management is to prevent exports of capital fromdisturbing conditions of the money market at home, thisclearly is a necessary complement of central bankingpolicy. But those who favour such a course seem hardlyto be conscious of what it involves. It would certainlynot be sufficient in the long run merely to prohibit themore conspicous forms of sending money abroad. It isof course true that if there are no impediments to theexport of capital the most convenient and therefore per-haps the quantitatively most important form which theexport of capital will take is the actual transfer of moneyfrom country to country. And it is conceivable that thismight be pretty effectively prevented by mere prohitionand control. To make even this really effective would ofcourse involve not only a prohibition of foreign lendingand of the import of securities of any description, butcould hardly stop short of a full-fledged system of foreignexchange control. But exchange control designed toprevent effectively the outflow of capital would reallyhave to involve a complete control of foreign trade, sinceof course any variation in the terms of credit on exportsor imports means an international capital movement.

To anyone who doubts the importance of this factor,I strongly recommend the very interesting memorandumon International Short Term Indebtedness which hasrecently been published by Mr. F. G. Conolly of the staffof the Bank for International Settlements in the recentjoint publication of the Carnegie Endowment and the


International Chamber of Commerce.* I will quote onlyone paragraph. " It has been the experience of everycountry whose currency has come under pressure ",writes Mr. Conolly, " that importers tend not only torefuse to utilise the normal period of credit but to covertheir requirements for months in advance; they prefer toutilise the home currency while it retains its internationalvalue rather than run the risk of being forced to payextra for the foreign currency necessary for their pur-chases. Exporters, on the other hand, tend to allowforeign currencies, the proceeds of exports already made,to lie abroad and to finance their current operations asfar as possible by borrowing at home. Thus a doublestrain falls on the exchange market; the normal supplyof foreign currencies from export dries up while thedemands from importers greatly increase. For a countrywith a large foreign trade the strain on the exchangemarket due to the effects of this" change over in tradefinancing may be very considerable. " 2. What Mr. Co-nolly here describes amounts, of course, to an export ofcapital which could only be prevented by controlling theterms of every individual transaction of the country'sforeign trade, an export of capital which may be equallyformidable whether the country carries on its foreigntrade " actively " or " passively ",8 that is whether itnormally provides the capital to finance the trade herselfor borrows it. Indeed to anyone who has had any

1 The Improvement of Commercial Relations Between Nations.The Problem of Monetary Stabilization. Separate Memoranda fromthe Economists consulted by the Joint Committee of the CarnegieEndowment and the International Chamber of Commerce andpractical Conclusions of the Expert Committee appointed by theJoint Committee. Paris, 1936, pp. 352 et seq.

• Ibid., p. 360• Cf. N. G. PIERSON, The Problem of Value in a Socialist Commu-

nity. Collectivist Economic Planning, London, 1935.


experience of foreign exchange control there should beno doubt possible that an export of capital can only beprevented by controlling not only the volume of exportsand imports so that they will always balance, but alsothe terms of credit of all these transactions.

At first indeed, and so long as discrepancies betweennational rates of interest are not too big and people havenot yet fully learnt to adapt themselves to fluctuatingexchanges, much less thoroughgoing measures may bequite effective. I can already hear some of my Englishfriends point out to me the marvellous discipline of theCity of London which on a slight hint from the Bank thatcapital exports would be undesirable will refrain fromacting against the general interest. But we need onlyvisualize how big the discrepancies between nationalinterest rates would become if capital movements werefor a time effectively stopped in order to realize how illu-sionary must be the hopes that anything but the strictestcontrol will be able to prevent them.

But let us disregard for the moment the technical diffi-culties inherent in any effective control of internationalcapital movements. Let us assume that the monetaryauthorities are willing to go any distance in creating newimpediments to international trade and that they actuallysucceed in preventing any unwanted change in interna-tional indebtedness. Will this successfully insulate acountry against the shocks which may result fromchanges in the rates of interest abroad? Or will these notstill transmit themselves via the effect such a change ofinterest rates will have on the relative prices of the inter-nationally traded securities and commodities? It is pro-bably obvious that so long as there is a fairly free inter-national movement of securities no great divergence inthe movement of rates of interest in the different coun-tries can persist for any length of time. But Monetary


Nationalists would probably not hesitate at any rateto attempt to inhibit these movements. It is not so gene-rally recognised however that commodity movementswill have a similar effect and perhaps this needs a fewmore words of explanation.

It will probably not be denied that a considerable risein the rate of interest will lead to a fall in the prices ofsome commodities relatively to those of others, particu-larly of those which are largely used for the productionof capital goods and of those of which large stocks areheld, compared with those which are destined for moreor less immediate consumption. And surely, in theabsence of immediate adjustments in tariffs or quotas,such a fall will transmit itself to the prices of similarcommodities in the country in which interest rates at firstare not allowed to rise. But if the prices of the goodswhich are largely used for investment fall relatively to theprices of other goods this means an increased profitabilityof investment compared with current production, conse-quently an increased demand for loans at the existingrates of interest, and, unless the central bank is willingto allow an indefinite expansion of credit, it will be com-pelled by the rise of interest rates abroad to raise its ownrate of interest, even if any outflow of capital has beeneffectively prevented. Although the supply of capital maynot change, the kind of goods which under the changedcircumstances it will be most profitable to import andexport will still alter the demand for capital with the sameeffects.1

The truth of the whole matter is that for a countrywhich is sharing in the advantages of the internationaldivision of labour it is not possible to escape from the

1 Cf. on this L. ROBBINS, The Great Depression, London, 1934,p. 175.


effects of disturbances in these international trade rela-tions by means short of severing all the trade ties whichconnect it with the rest of the world. It is of course truethat the less the points of contact with the rest of theworld the less will be the extent to which disturbancesoriginating outside the country will affect its internalconditions. But it is an illusion that it would be pos-sible, while remaining a member of the internationalcommercial community, to prevent disturbances from theoutside world from reaching the country by following anational monetary policy such as would be indicated ifthe country were a closed community. It is for this rea-son that the ideology of Monetary Nationalism has proved,and if it remains influential will prove to an even greaterextent in the future, to be one of the main forces des-troying what remnants of an international economic sys-tem we still have.

There are two more points which I should like speciallyto emphasize before I conclude for to-day. One is thatup to this point I have, following the practice of theMonetary Nationalists, considered mainly the disturbingeffects on a country of changes in the demand for capitaloriginating abroad. But there is of course an other sideto this picture. What from the point of view of the coun-try to which the effects are transmitted from abroad is adisturbance is from the point of view of the country wherethe original change takes place a stabilising effect. Tohave to give up capital because somewhere else a suddenmore urgent demand has arisen is certainly unsettling.But to be able to obtain capital at short notice if a suddenunforeseen need arises at home will certainly tend to sta-bilise conditions at home. It is more than unlikely thatfluctuations on the national capital market would besmaller if the world were cut up into watertight compart-ments. The probability is rather that in this case flue-


tuations within each national territory would be muchmore violent and disturbing than they are now.

Closely conected with this is the second point, onwhich I can touch only even more shortly. I have alreadymentioned the probability that the restrictions on capitalmovements involved in a policy of Monetary Nationalismwould tend to increase the differences between nationalinterest rates. This would of course be due to the factthat while instability of exchange rates would tend toincrease the volume and frequency of irregular flows ofshort term funds, it would to an even greater degreedecrease the volume of international long term invest-ment. Although by some this is regarded as a goodthing, I doubt whether they fully appreciate what itwould mean. The purely economic effects, the restric-tion of international division of labour which it implies,and the reduction in the total volume of investment towhich it would almost certainly lead, are bad enough.But even more serious seem to me the political effects ofthe intensification of the differences in the standard oflife between different countries to which it would lead.It docs not need much imagination to visualize the newsources of international friction which such a situationwould create.* But this leads me beyond the properscope of these lectures and I must confine myself todrawing your attention to it without attempting to elabo-rate it any further.

* Cf. in this L. ROBBINS, Economic Planning and InternationalOrder, pp. C8 et seq.

LECTURE V

THE PROBLEMSOF A REALLY INTERNATIONAL STANDARD

I have now concluded the negative part of my argu-ment, the case against independent national currencies.While I cannot hope in the space of these few lecturescompletely to have refuted the theoretical basis of Mone-tary Nationalism, I hope at least to have shown threethings : that there is no rational basis for the separateregulation of the quantity of money in a national areawhich remains a part of a wider economic system; thatthe belief that by maintaining an independent nationalcurrency we can insulate a country against financialshocks originating abroad is largely illusory; and that asystem of fluctuating exchanges would on the contraryintroduce new and very serious disturbances of interna-tional stability. I do not want now further to add to thisexcept that I might perhaps remind you that my argu-ment throughout assumed that such a system would berun as intelligently as is humanly possible. I haverefrained from supporting my case by pointing to theabuses to which such a system would almost certainlylend itself, to the practical impossibility of different coun-tries agreeing on what degree of depreciation is justified,to the consequent danger of competitive depreciation, andthe general return to mercantilist policies of restriction

74 A REALLY INTERNATIONAL STANDARD

which now, as in earlier centuries, are the inevitablereaction to debasement in other countries.x

We must recognise, therefore, that independent regu-lation of the various national currencies cannot beregarded as in any sense a substitute for a rationally regu-lated world monetary system. Such a system may to-dayseem an unattainable ideal. But this does not mean thatthe question of what we can do to get as near the ideal asmay be practicable does not present a number of impor-tant problems. Of course some " international " systemswould be far from ideal. I hope I ha/e made it clear inparticular that I do not regard the sort of internationalsystem which we have had in the past as by any meanscompletely satisfactory. The Monetary Nationalists con-demn it because it is international; I, on the other hand,ascribe its shortcomings to the fact that it is not inter-national enough. But the question how we can make itmore satisfactory, that is more genuinely international,I have not yet touched upon. It is a question which raisesexceedingly difficult problems; I can survey them onlyrapidly in this final lecture.

The first, but by no means the most important or mostinteresting question which I must consider is the questionwhether the international standard need be gold. Onpurely economic grounds it must be said that there arehardly any arguments which can be advanced for, andmany serious objections which can be raised against, theuse of gold as the international money. In a securelyestablished world State with a government immune

1 I feel I must remind the reader here that limitations of timemade it impossible for me to dwell in these lectures on the tre-mendously important practical effects of a policy of Monetary Natio-nalism on commercial policy as long as I should have wished. Al-though this is well trodden ground, it cannot be too often reite-rated that without stability of exchange rates it is vain to hopefor any reduction of trade barriers.

A REALLY INTERNATIONAL STANDARD 75

against the temptations of inflation it might be absurd tospend enormous effort in extracting gold out of the earthif cheap tokens would render the same service as goldwith equal or greater efficiency. Yet in a world consist-ing of sovereign national States there seem to me to existcompelling political reasons why gold (or the preciousmetals) alone and no kind of artificial international cur-rency, issued by some international authority, could beused successfully as the international money. It is essen-tial for the working of an international standard that eachcountry's holdings of the international money shouldrepresent for it a reserve of exchange medium which inall eventualities will remain universally acceptable ininternational transactions. And so long as there are sepa-rate sovereign States there will always loom large amongthese eventualities the danger of war, or of the break-down of the international monetary arrangements forsome other reason. And since people will always feelthat against these emergencies they will have to holdsome reserve of the one thing which by age-long customcivilized as well as uncivilized people are ready to accept—that is, since gold alone will serve one of the purposesfor which stocks of money are held—and since to someextent gold will always be held for this purpose, therecan be little doubt that it is the only sort of internationalstandard which in the present world has any chance ofsurviving. But, to repeat, while an international stan-dard is desirable on purely economic grounds, the choiceof gold with all its undeniable defects is made necessaryentirely by political considerations.

What should be done if the well-known defects of goldshould make themselves too strongly felt, if violentchanges in the condition of its production or the appea-rance of a large new demand for it should threatensudden changes in its value, is of course a problem of


major importance. But it is neither the most interestingnor the most important problem and I do not propose todiscuss it here. The difficulties which I want to considerare rather those which were inherent in the internationalgold standard, even before 1914, and to a still greaterdegree during its short post-war existence. They are theproblems which arise out of the fact that the so-calledgold currencies are connected with gold only through thecomparatively small national reserves which form thebasis of a multiple superstructure of credit money whichitself consists of many different layers of different degreesof liquidity or acceptability. It is, as we have seen, thisfact which makes the effects of changes in the interna-tional (low of money different from merely interlocalshifts, to which is due the existence of separate nationalmonetary systems which to some extent have a life oftheir own. The homogeneity of the circulating mediumof different countries has been destroyed by the growthof separate banking systems organized on national lines.Can anything be done to restore it?

Il is important here first to distinguish between theneed for some " lender of last resort " and the organiza-tion of banking on the " national reserve " principle.That an extensive use of bank deposits as money wouldnot be possible, that deposit banking of the modern typecould not exist, unless somebody were in a position toprovide the cash if the public should suddenly want toconvert a considerable part of its holdings of bank depo-sits into more liquid forms of money, is probably beyonddoubt. It is far less obvious why all the banking institu-tions in a particular area or country should be made torely on a single national reserve. This is certainly not a


system which anybody would have deliberately devised onrational grounds and it grew up as an accidental by-pro-duct of a policy concerned with different problems.a Therational choice would seem to lie between either a systemof " free banking ", which not only gives all banks theright of note issue and at the same time makes it neces-sary for them to rely on their own reserves, but alsoleaves them free to choose their field of operation andtheir correspondents without regard to national boun-daries, 2 and on the other hand, an international centralbank. I need not add that both of these ideals seemutterly impracticable in the world as we know it. But Iam not certain whether the compromise we have chosen,that of national central banks which have no direct powerover the bulk of the national circulation but which holdas the sole ultimate reserve a comparatively small amountof gold is not one of the most unstable arrangementsimaginable.

Let us recall for a moment the essential features of theso-called gold standard systems as they have existed inmodern times. The most widely used medium ofexchange, bank deposits, is not fixed in quantity. Addi-tional deposits may at any time spontaneously spring up(be " created " by the banks) or part of the total maysimilarly disappear. But while they are predominantlyused in actual payments, they are by no means the onlyforms in which balances can be held to meet such pay-ments. In this function deposits on current account areonly one item—a very liquid one, although by no meansthe most liquid of all—in a long range of assets of varying

1 Gf. W. BAGEHOT, Lombard Street, and V. C. SMITH, The Rationaleof Central Banking, London, 1936.

1 Cf. L. V. MISES, Geldwertstabilisierung und Konjunkturpolitik,Jena, 1928.


degree of liquidity.x Overdraft facilities, saving depositsand many types of very marketable securities on the onehand, and bank notes and coin on the other, will at diffe-rent times and to different degrees compete with bankdeposits in this function. And the amounts which willbe held on current account to meet expected demandsneed not therefore fluctuate with the expected magnitudeof these payments; they may also change with any changein the views about the ease with which it will be pos-sible to convert these other assets into bank deposits. Thesupply of bank deposits on the other hand will dependon similar considerations. How much the banks will bewilling to owe in this form in excess of the ready cashthey hold will depend on their view as to how easy it willbe tc convert other assets into cash. It is when generalconfidence is high, so that comparatively small amountsof bank deposits will be. needed for a given volume of pay-ments, that the banks will be more ready to increase theamount of bank deposits. On the other hand, anyincrease of uncertainty about the future will lead to anincreased demand for all the more liquid forms of assets,that is, in particular, for bank deposits and cash, and toa decrease in the supply of bank deposits.

Where there is a central bank the responsibility for theprovision of cash for the conversion of deposits is dividedbetween the banks and the central bank, or one shouldprobably better say shifted from the banks to the centralbank, since it is now the recognized duty of the centralbanks to supply in an emergency—at a price—all thecash that may be needed to repay deposits. Yet whilethe ultimate responsibility to provide the cash whenneeded is thus placed on the central bank, until thisdemand actually arises, the latter has little power to pre-

1 Cf. particularly J. R. HICKS, A Suggestion for the Simplifi-cation of the Theory of Money (Economica, N. S. vol. II/5, February1936, and F. LAVINGTON, The English Capital Market, 1921, p. 30.


vent the expansion leading to an increased demand forcash.

But with an international standard a national centralbank is itself not a free agent. Up to this point the cashabout which I have been speaking is the money createdby the central bank which within the country is gene-rally acceptable and is the only means of payment outsidethe circle of the customers of a particular bank. The cen-tral bank, however, has not only to provide the requiredamounts of the medium generally accepted within thecountry; it has also to provide the even more liquid, inter-nationally acceptable, money. This means that in a situa-tion where there is a general tendency towards greaterliquidity there will be at the same time a greater demandfor central bank money and for the international money.But the only way in which the central bank can restrictthe demand for and increase the supply of the interna-tional money is to curtail the supply of central bankmoney. In consequence, in this stage as in the precedingone, any increase in the demand for the more liquid typeof money will lead to a much greater decrease in the sup-ply of the somewhat less liquid kinds of money.

This differentiation between the different kinds ofmoney into those which can be used only among thecustomers of a particular bank and those which can beused only within a particular country and finally thosewhich can be used internationally—these artificial dis-tinctions of liquidity (as I have previously called them)—have the effect, therefore, that any change in the relativedemand for the different kinds of money will lead to acumulative change in the total quantity of the circulatingmedium. Any demand on the banks for conversion ofpart of their deposits into cash will have the effect of com-pelling them to reduce their deposits by more than theamount paid out and to obtain more cash from the central


bank, which in turn will be forced to take counter-mea-sures and so to transmit the tendency towards contractionto the other banks. And the same applies, of course,mutatis mutandis to a decrease in the demand for themore liquid type of assets, which will bring about a con-siderable increase in the supply of money.

All this is of course only the familiar phenomenonwhich Mr. R. G. Hawtrey has so well described as the" inherent instability of credit ". But there are two pointsabout it which deserve special emphasis in this connec-tion. One is that, in consequence of the particular orga-nisation of our credit structure, changes in liquidity pre-ference as between different kinds of money are probablya much more potent cause of disturbances than thechanges in the preference for holding money in generaland holding goods in general which have played such agreat role in recent refinements of theory. The other isthat this source of disturbance is likely to be much moreserious when there is only a single bank for a wholeregion or when all the banks of a country have to relyon a single central bank; since the effect of any changein liquidity preference will generally be confined to thegroup of people who directly or indirectly rely on thesame reserve of more liquid assets.

It seems to follow from all this that the problem withwhich we are concerned is not so much a problem of cur-rency reform in the narrower sense as a problem ofbanking reform in general. The seat of the trouble iswhat has been very appropriately been called the per-verse elasticity of bank deposits x a s a medium of circula-tion, and the cause of this is that deposits, like otherforms of " credit money ", are claims for another, more

• L. CURRIE, The Supply and Control of Money in the UnitedStates, Cambridge, 1934, pp. 130 et seq.


generally acceptable sort of money, that a proportionalreserve of that other money must be held against them,and that their supply is therefore inversely affected bythe demand for the more liquid type of money.

By far the most interesting suggestion on BankingReform which has been advanced in recent years, notbecause in its present form its seems to be practicable oreven theoretically right, but because it goes to the heartof the problem, is the so-called Chicago or 100 per centplan.* This proposal amounts in effect to an extensionof the principles of Peel's Act of 1844 to bank deposits.The most praticable suggestion yet made for its executionis to give the banks a sufficient quantity of paper moneyto increase the reserves held against demand deposits to100 per cent and henceforth to require them to maintainpermanently such a 100 per cent reserve.

In this form the plan is conceived as an instrument ofMonetary Nationalism. But there is no reason why itshould not equally be used to create a homogeneous inter-national currency.2 A possible, although perhaps some-what fantastic, solution would seem to be to reduce pro-portionately the gold equivalents of all the differentnational monetary units to such an extent that all themoney in all countries could be covered 100 per cent bygold, and from that date onwards to allow variations inthe national circulations only in proportion to changes

1 On the significance of the " Chicago Plan " compare particu-larly the interesting and stimulating article by H. C. SIMONS, Ruleversus Authority in Monetary Policy (Journal of Political Economy,vol. 44, no. 1, February 1936, and F. LUTZ, Das Grundproblem derGeldverfassung, 1936, where references to the further literature onthe proposal will be found.

* Gf. H. C. SIMONS, loc. cit.t p. 5, note 3.


in the quantity of gold in the country.1 Such a planwould clearly require as an essential complement an inter-national control of the production of gold, since theincrease in the value of gold would otherwise bring aboutan enormous increase in the supply of gold. But thiswould only provide a safety valve probably necessary inany case to prevent the system from becoming all too rigid.

The undeniable attractivenes of this proposal liesexactly in the feature which makes it appear somewhatimpracticable, in the fact that in effect it amounts, as isfully realized by at least one of its sponsors, to an aboli-tion of deposit banking as we know it.2 It does provide,instead of the variety of media of circulation which to-dayrange according to their degree of acceptability from bankdeposits to gold, one single kind of money. And it woulddo away effectively with that most pernicious feature ofour present system : namely that a movement towardsmore liquid types of money causes an actual decrease inthe total supply of money and vice versa. The mostserious question which it raises, however, is whether byabolishing deposit banking as we know it we would effec-tively prevent the principle on which it rests from mani-festing itself in other forms. It has been well remarkedby the most critical among the originators of the schemethat banking is a pervasive phenomenon * and the ques-tion is whether, when we prevent it from appearing in itstraditional form, we will not just drive it into other andless easily controllable forms. Historical precedent rathersuggests that we must be wary in this respect. The Actof 1844 was designed to control what then seemed to bethe only important substitute for gold as a widely used

1 A perhaps somewhat less impracticable alternative might beinternational bimetallism at a suitable ratio.

• Cf. H. C. SIMONS, loc. cit., p. 16.• Ibid., p. 17.


medium of exchange and yet failed completely in itsintention because of the rapid growth of bank deposits.Is it not possible that if similar restrictions to those placedon bank notes were now placed on the expansion of bankdeposits, new forms of money substitutes would rapidlyspring up or existing ones would assume increasingimportance? And can we even to-day draw a sharp linebetween what is money and what is not? Are there notalready all sorts of " near-moneys "1 like saving deposits,overdraft facilities, bills of exchange, etc., which satisfyat any rate the demand for liquid reserves nearly as wellas money?

I am afraid all this must be admitted, and it conside-rably detracts from the alluring simplicity of the 100 percent banking scheme. It appears that for this reason ithas now also been abandoned by at least one of its ori-ginal sponsors.2 The problem is evidently a much widerone and I agree with Mr. H. C. Simons that it " cannot bedealt with merely by legislation directed at what we callbanks ".s Yet in one respect at least the 100 per centproposal seems to me to point in the right direction.Even if, as is probably the case, it is impossible to drawa sharp line between what is to be treated as money andwhat is not, and if consequently any attempt to fix rigidlythe quantity of what is more or less arbitrarily segre-gated as " money " would create serious difficulties, ityet remains true that, within the field of instrumentswhich are undoubtedly generally used as money, thereare unnecessary and purely institutional distinctions ofliquidity which are the sources of serious disturbancesand which should as far as possible be eliminated. If thiscannot be done for the time being by a general return to

1 Op. ciL, p. 17.• Ibid., p. 17.• Ibid.


the common use of the same international medium in thegreat majority of transactions, it should at least be pos-sible to approach this goal by reducing the distinctionsof liquidity between the different kinds of money actuallyused, and offsetting as far as possible the effects of changesin the demand for liquid assets on the total quantity ofthe circulating medium.

This brings me to the more practical question of whatcan be done to diminish the instability of the credit struc-ture if the general framework of the present monetarysystem is to be maintained. The aim, as we have justseen, must be to increase the certainty that one form ofmoney will always be readily exchangeable against otherforms of money at a known rate, and that such changesshould not lead to changes in the total quantity of money.In so far as the relations between different national cur-rencies are concerned this leads, of course, to a demandfor reforms in exactly the opposite direction from thoseadvocated by Monetary Nationalists. Instead of flexibleparities or a widening of the " gold points ", absolutefixity of the exchange rates should be secured by a systemof international par clearance. If all the central banksundertook to buy and sell foreign exchange freely at thesame fixed rates, and in this way prevented even fluctua-tions within t h e " gold points ", the remaining diffe-rences in denomination of the national currencies wouldreally be no more significant than the fact that the samequantity of cloth can be stated in yards and in meters.With an international gold settlement fund on the linesof that operated by the Federal Reserve System, whichwould make it possible to dispense with the greater partof the actual gold movements which used to take place in


the past, invariable rates of exchange could be main-tained without placing any excessive burden on the cen-tral banks. * The main aim here would of course berather to remove one of the main causes of internationalmovements of short term funds than to prevent suchmovements or to offset their effects by means which willonly increase the inducement to such movements. 2

But invariability of the exchange rates is only one pre-condition of a successful policy directed to minimizemonetary disturbances. It eliminates one of the institu-tional differentiations of liquidity which are likely to giverise to sudden changes in favour of holding one sort ofmoney instead of another. But there remains the furtherdistinction between the different sorts of money whichconstitute the national monetary systems; and, so long asthe general framework of our present banking systems isretained, the dangers to stability which arise here canhardly be combatted otherwise than by a deliberatepolicy of the national central banks.

The most important change which seems to be neces-

1 The founders of the Bank for International Settlements defi-nitely contemplated that the Bank might establish such a fundand article 24 of it Statutes specifically states that the bank mayenter into special agreements with central banks to facilitate thesettlement of international transactions between them. — " Forthis purpose it may arrange with central banks to have goldearmarked for their account and transferable on their order, toopen accounts through which central banks can transfer theirassets from one currency to another and to take such other mea-sures as the Board may think advisable within the limits of thepowers granted by these Statutes ".

a In a book which has appeared since these lectures were deli-vered (G. R. WHITTLESEY, International Monetary Issues, New York,1937) the author, after pointing out that a widening of the goldpoints would have the effects of increasing the volume of shortterm capital movements of this sort (p. 116) concludes that " theonly way of overcoming this factor would be to eliminate the goldpoints " (p. 117). But the only way of eliminating the gold pointsof which he can think is to abolish the gold standard!


sary here is that the gold reserves of all the central banksshould be made large enough to relieve them of the neces-sity of bringing about a change in the total national cir-culation in proportion to the changes in their reserves;that is, that any change in the relative amounts of moneyin different countries should be brought about by theactual transfer of corresponding amounts from countryto country without any " secondary " contractions andexpansions of the credit super-structure of the countriesconcerned. This would be the case only if individual cen-tral banks held gold reserves large enough to be usedfreely without resort to any special measures for their" protection ".

Now the present abundance of gold offers an excep-tional opportunity for such a reform. But to achieve thedesired result not only the absolute supply of gold butalso its distribution is of importance. In this respect itmust appear unfortunate that those countries which com-mand already abundant gold reserves and would there-fore be in a position to work the gold standard on theselines, should use that position to keep the price artificiallyhigh. The policy on the part of those countries which arealready in a strong gold position, if it aims at the restora-tion of an international gold standard, should have been,while maintaining constant rates of exchange with allcountries in a similar position, to reduce the price ofgold in order to direct the stream of gold to those coun-tries which are not yet in a position to resume gold pay-ments. Only when the price of gold had fallen suffi-ciently to enable those countries to acquire sufficientreserves should a general and simultaneous return to afree gold standard be attempted.

It may seem at first that even if one could start withan appropriate distribution of gold between countrieswhich at first would put each country in a position where


it could allow its stock of gold to vary by the absoluteamounts by which its circulation would have to increaseor decrease, some countries would soon again find theirgold stocks so depleted that they would be compelled totake traditional measures for their protection. And itcannot be denied that so long as the stock of gold of anycountry is anything less than 100 per cent of its total cir-culation, it is at least conceivable that it may be reducedto a point were in order to protect the remainder themonetary authorities might have to have recourse to apolicy of credit contraction. But a short reflection willshow that this is extraordinarily unlikely to happen if acountry starts out with a fairly large stock of gold and ifits monetary authorities adhere to the main principle notonly with regard to decreases but equally with regard toincreases in the total circulation.

If we assume the different countries to start with a goldreserve amounting to only a third of the total monetarycirculation 1 this would probably provide a margin amplysufficient for any reduction of the country's share in theworld's stock of money which is likely to become neces-sary. That a country's share in the world's income, andtherefore its relative demand for money, should fall offby more than this would at any rate be an exceptionalcase requiring exceptional treatment.2 If history seems

1 At the present value of gold the world's stock of monetarygold (at the end of 1936) amounts to 73.5 per cent, of all sightliabilities of the central banks plus the circulation of Governmentpaper money. The percentage would of course be considerablylower if, as would be necessary for this purpose, the comparisonwere made with the total of sight deposits with commercial banksplus bank notes etc. in the hands of the public. But there can beno doubt that even if the price of gold should be somewhat lowered(say by one seventh, i. e. from 140 to 120 shillings or from $ 35to $ 30 per ounce) there would still be ample gold available toprovide sufficient reserves.

2 If in spite of this in an individual case the gold reserves ota country should be nearly exhausted, the necessary remedy would


to suggest that such considerable losses of gold are not atall infrequent, this is due to the operation of a differentcause which should be absent if the principle suggestedwere really applied. If under the traditional gold stan-dard any one country expanded credit out of step withthe rest of the world this did usually bring about an out-flow of gold only after a considerable time lag. This initself would mean that, before equilibrium would be res-tored by the direct operations of the gold flows, anamount of gold approximately equal to the credit createdin excess would have to flow out of the country. If,however, as has often been the case, the country shouldbe tardy in decreasing its circulation by the amount ofgold it has lost, that is, if it should try to " offset " thelosses of gold by new creations of credit, there would beno limits to the amount of gold which may leave thecountry except the size of the reserves. Or in otherwords, if the principle of changing the total circulation bythe full amount of gold imported or exported were strictlyapplied, gold movements would be much smaller thanhas been the case in the past, and the size of the goldmovements experienced in the past create therefore nopresumption that they would be equally large in thefuture.

These considerations will already have made it clear thatthe principle of central banking policy here proposed by

be to acquire the necessary amount of gold through an externalloan and to give this amount to the central bank in repayment ofpart of the state debt which presumably will constitute at leastpart of its non gold assets (or in payments of any other assetswhich the bank would have to sell to the Government). The mainpoint here is that the acquisition of this gold must be paid forout of taxation and not by the creation of additional credit by thecentral bank.


no means implies that the central banks should be relievedfrom all necessity of shaping their credit policy accordingto the state of their reserves. Quite the contrary. It onlymeans that they should not be compelled to adhere to the.mechanical rule of changing their notes and deposits inproportion to the change in their reserves. Instead ofthis they would have to undertake the much more diffi-cult task of influencing the total volume of money intheir countries in such a way that this total would changeby the same absolute amounts as their reserves. And sincethe central bank has no direct power over the greater partof the circulating medium of the country it would haveto try to control its volume indirectly. This means thatit would have to use its power to change the volume ofits notes and deposits so as to make the superstructure ofcredit built on those move in conformity with its reserves.Rut as the amount of ordinary bank deposits and otherforms of common means of exchange based on a givenvolume of central bank money will be different at diffe-rent times, this means that the central bank, in order tomake the total amount of money move with its reserves,would frequently have to change the amount of centralbank money independently of changes in its reserves andoccasionally even in a direction opposite to that in whichits reserves may change.

It should perhaps always have been evident that, witha banking system which has grown up to rely on theassistance of a central bank for the supply of cash whenneeded, no sort of control of the circulating medium canbe achieved unless the central bank has power and usesthis power to control the volume of bank deposits inordinary times. And the policy to make this controleffective will have to be very different from the policy ofa bank which is concerned merely with its own liquidity.It will have to act persistently against the trend of the


movement of credit in the country, to contract the creditbasis when the superstructure tends to expand and toexpand the former when the latter tends to contract.

It is to-day almost a commonplace that, with a deve-lopped banking structure, the policy of the central bankcan in no way be automatic. It would indeed requirethe greatest art and discernment for a central bank tosucceed in making the credit money provided by the pri-vate banks behave as a purely metallic circulation wouldbehave under similar circumstances. But while it mayappear very doubtful whether this ideal will ever be fullyachieved, there can be no doubt that we are still so farfrom it that very considerable changes from traditionalpolicy would be required before we shall be able to saythat even what is possible has been achieved.

In any case it should be obvious that the existence ofa central bank which does nothing to counteract theexpansions of banking credit made possible by its exis-tence only adds another link in the chain through whichthe cumulative expansions and contractions of credit ope-rate. So long as central banks are regarded, and regardthemselves, only as " lenders of last resort " which haveto provide the cash which becomes necessary in conse-quence of a previous credit expansion with which, untilthis point arrives, they are not concerned, so long ascentral banks wait until " the market is in the Bank "before they feel bound to check expansion, we cannothope that wide fluctuations in the volume of credit willbe avoided. Certainly Mr. Hawtrey was right with hisnow celebrated statement that " so long as the credit isregulated with reference to reserve proportions, the tradecycle is bound to recur ".1 But I am afraid only one andthat not the more important of the essential corrolaries

1 Monetary Reconstruction, London, 1923, p. 144.


of this proposition is usually derived from this statement.What is usually emphasized is the fact that concern withreserve proportions will ultimately compel central banksto stop a process of credit expansion and actually to bringabout a process of credit contraction. What seems to memuch more important is that sole regard to their ownreserve proportions will not lead central banks to coun-teract the increase of bank deposits, even if it means anincrease of the credit circulation of the country relativelyto the gold reserve, and although it is an increase largelymade possible by the certain expectation - on the part ofthe other banks that the central bank will in the endsupply the cash needed.

On the question how far central banks are in practicelikely to succeed in this difficult task different opinionsare clearly possible. The optimist will be convinced thatthey will be able to do much more than merely offset thedangers which their existence creates. The pessimist willbe sceptical whether on balance they will not do moreharm than good. The difficulty of the task, the impossi-bility of prescribing any fixed rule, and the extent towhich the action of the central banks will always beexposed to the pressure of public opinion and politicalinfluence certainly justify grave doubts. And though thealternative solution is to-day probably outside of therealm of practical politics, it is sufficiently important todeserve at least a passing consideration before we leavethis subject.

As I have pointed out before, the " national reserveprinciple " is not insolubly bound up with the centra-lization of the note issue. While we must probably takeit for granted that the issue of notes will remain reservedto one or a few privileged institutions, these institutionsneed not necessarily be the keepers of the national reserve.There is no reason why the Banks of Issue should not be


entirely confined to the functions of the issue departmentof the Bank of England, that is to the conversion of goldinto notes and notes into gold, while the duty of holdingappropriate reserves is left to individual banks. Therecould still be in the background—for the case of a run onthe banks—the power of a temporary " suspension " ofthe limitations of the note issue and of the issue of anemergency currency at a penalizing rate of interest.

The advantage of such a plan would be that one tier inthe pyramid of credit would be eliminated and the cumu-lative effects of changes in liquidity preference accor-dingly reduced. The disadvantage would be that theremaining competing institutions would inevitable haveto act on the proportional reserve principle and thatnobody would be in a position, by a deliberate policy, tooffset the tendency to cumulative changes. This mightnot be so serious if there were numerous small bankswhose spheres of operation freely overlapped over thewhole world. But it can hardly be recommended wherewe have to deal with the existing banking systems whichconsist of a few large institutions covering the same fieldof a single nation. It is probably one of the ideals whichmight be practical in a liberal world federation but whichis impracticable where national frontiers also mean boun-daries to the normal activities of banking institutions.The practical problem remains that of the appropriatepolicy of national central banks.

6

It is unfortunately impossible to say here more aboutthe principles which a rational central banking policywould have to follow without going into some of themost controversial problems of the theory of the tradecycle which clearly fall outside the scope of these lee-


tures. I must therefore confine myself to pointing outthat what I have said so far is altogether independent ofthe particular views on this subject for which I have beenaccused, I think unjustly, of being a deflationist.Whether we think that the ideal would be a more or lessconstant volume of the monetary circulation, or whetherwe think that this volume should gradually increase ata fairly constant rate as productivity increases, the pro-blem of how to prevent the credit structure in any coun-try from running away in either direction remains thesame.

Here my aim has merely been to show that whateverour views about the desirable behaviour of the total quan-tity of money, they can never legitimately be applied tothe situation of a single country which is part of an inter-national economic system, and that any attempt to do sois likely in the long run and for the world as a wholeto be an additional source of instability. This means ofcourse that a really rational monetary policy could becarried out only by an international monetary authority,or at any rate by the closest cooperation of the nationalauthorities and with the common aim of making the cir-culation of each country behave as nearly as possible asif it were part of an intelligently regulated internationalsystem.

fiut I think it also means that so long as an effectiveinternational monetary authority remains an Utopiandream, any mechanical principle (such as the gold stan-dard) which at least secures some conformity of mone-tary changes in the national area to what would happenunder a truly international monetary system is far pre-ferable to numerous independent and independentlyregulated national currencies. If it does not provide areally rational regulation of the quantity of money, it atany rate tends to make it behave on roughly foreseeable


lines, which is of the greatest importance. And sincethere is no means, short of complete autarchy, of pro-tecting a country against the folly or perversity of themonetary policy of other countries, the only hope ofavoiding serious disturbances is to submit to some com-mon rules, even if they are by no means ideal, in orderto induce other countries to follow a similarly reasonablepolicy. That there is much scope for an improvement ofthe rules of the game which were supposed to exist inthe past, nobody will deny. The most important step inthis direction is that the rationale of an internationalstandard and the true sources of the instability of ourpresent system should be properly appreciated. It wasfor this reason that I felt that my most urgent task wasto restate the broader theoretical considerations whichbear on the practical problem before us. I hope that byconfining myself largely to these theoretical problems Ihave not too much disappointed the expectations to whichthe title of these lectures may have given rise. But, as Isaid at the beginning of these lectures, I do believe thatin the long run human affairs are guided by intellectualforces. It is this belief which for me gives abstract con-siderations of this sort their importance, however slightmay be their bearing on what is practicable in the imme-diate future.

hayek, f. (1937) 'monetary nationalism and international stability

Documents