International Monetary Economics

CHAPTER IX

SOME CURRENCY SYSTEMS AND MONETARY INTERNATIONALISM

THE analysis of the preceding chapter has led us to the very important conclusion that the process of restoring equilibrium in international payments does not depend in its structure on the monetary system adopted. Whatever the monetary organization, certain operations have proved to be indispensable if re-equilibrium is to be obtained; they depend—and this is the reason of their absolute necessity—on the very structure of the economic transactions which result in payments from country to country. The fundamental opposition is therefore, in this context, not between one monetary system and another, but between stability and instability in international monetary relations. In the present chapter we shall discuss the principal contemporary currency systems from the point of view of international monetary stability, that is, of monetary internationalism1; in the following chapter we shall then analyse some problems resulting from monetary nationalism and from the instability that it engenders. The scope of our present discussion will be limited to the gold standard, foreign-exchange standards and free paper currencies.

A. The International Gold Standard

(1)

Definitions of the gold standard are abundant and while they are not uniform there is little disagreement as to the nature of the system. A critical examination of the literature of the subject would bring out, however, important divergencies in the emphasis with which now this, now that, feature of the gold standard is stressed by the various writers on the subject. Some of the principal characteristics of the gold standard have already been referred to in the preceding chapters; here it is merely intended to recapitulate and systematize these various considerations.

According to the definition suggested when discussing the problem of monetary parities, a currency standard is a commodity the price of which in terms of the national currency is fixed by the monetary law of the country and maintained stable by the monetary authorities of that country.2 Substitute “gold” for “a commodity”, and the result is the definition of the gold standard. The essential feature of that system is the fact of fixing the price of gold in terms of the national monetary unit and the maintenance of that price by the central bank or the treasury, according to the way in which the monetary system is organized. It will be observed that gold-standard currency is not gold currency but paper money3 administered in such a way as to keep the price of gold stable. Convertibility, i.e. the obligation of the monetary authorities to buy and to sell unlimited quantities of gold against bank notes, which is an important feature of the gold standard, can be explained by the historic process which has led from gold currency to the gold standard. Besides this “genetic” explanation, however, it will be observed that convertibility is a necessary feature of the system, since this is precisely the instrument by means of which the price of gold is stabilized. More than that: the obligation for the central bank to buy and to sell gold at a certain price is a more fundamental feature of the system than the long-run fixity of that price. A currency remains linked to gold as long as gold can be bought and sold in unlimited quantities at a price controlled by the monetary authorities—even should that price be changed by the competent authorities at a subsequent date. Thus our definition of “monetary standard” ought perhaps to be qualified by saying that the price of the standard commodity is fixed by law or by the authority to which the function of determining that price has been delegated. Should this wider definition be adopted, then a country might be said to be on the gold standard even if the price of gold were not fixed by law but merely determined by the monetary authorities. In order to avoid confusion, I find it preferable to draw a distinction between a country being on the gold standard and its having a monetary system linked to gold; the former term should continue to designate the traditional notion of the gold standard as a system under which the price of gold is fixed by the monetary laws of the country, while the latter should be used to describe a situation where currency is linked to gold in a less determined way. By introducing this second notion one avoids the mistake of considering that by “going off the gold standard” a monetary system necessarily loses all its connection with gold,4 and then the price of gold becomes indeterminate. As long as the monetary authorities control that price it is not indeterminate even though it has ceased to be fixed by law. When all links between a monetary system and gold are severed we say that gold is de-monetized, not that the system is “off the gold standard”; if this happens, the price of gold ceases to be controlled by monetary authorities and begins to behave like that of any other commodity. Let us emphasize again that a gold-standard currency is a paper currency linked to gold through the legal determination of its price. This way of looking at it, while at variance with the traditional views about that system, corresponds much more closely to economic realities and has the advantage of stressing the non-automatic character of the gold standard.5 As long as the central bank fulfils its obligation of keeping the price of gold stable in terms of the monetary unit, it is free to pursue various courses of policy and be guided by various additional criteria. In fact the only real limitations to the freedom of action of central banks result from international relations, and it is only in these relations that one can find a justification of, and arguments for, the gold standard. In an isolated community (“closed economy”) there would be no point in establishing the gold standard except as a transitory system leading from an all-gold currency to a paper currency having no link with gold whatever.

(2)

What is then the international significance of the gold standard? Let us begin by quoting the following judicious statement by Professor Gregory:6

“When a whole series of countries possess currency systems which have a fixed relationship with gold . . . an international gold-standard system comes into existence, not in any formal fashion, but as a matter of fact. The only intelligible meaning to be assigned to the phrase ‘the international gold standard’ is the simultaneous presence, in a group of countries, of arrangements by which, in each of them, gold is convertible at a fixed rate into local currency and the local currency into gold, and by which gold movements from any one of these areas to any of the others are freely permitted by all of them.”

The fundamental feature of the “international gold standard” is that it represents a common standard upon which the various national monetary systems are organized. The consequence of having a common standard is the possibility of establishing precise parities between the various national monetary units.7 As we have seen the notion of parity has a precise meaning wherever the various national currencies are based on a common standard. This is indeed the only precise meaning that that notion can have, since the conception of purchasing power parities has been found wanting.8 In order to make possible this clear definition of parities it is indispensable that the standard commodity should be allowed to move freely from country to country subject neither to tariffs nor to any sort of quantitative restrictions. Only if the status of free trade is applied to it, can a standard commodity become a tertium comparationis between the various monetary units and relate them reciprocally in a clear and unequivocal way. Hence the freedom of international movements of gold is an essential feature of the international gold standard. It is even more important that gold should move freely between countries than that its price should be fixed by law and kept stable. England after 1931 remained nearer the gold standard than would have been the case had the former legal price of gold been maintained but an embargo on gold exports introduced. The gold standard not only makes it possible for the various national currencies to have precise parity relations with one another, but also limits considerably the fluctuations of exchange rates. As has been shown in Chapter VI, gold movements are substituted for further exchange fluctuations once these latter have reached the so-called gold points. Thus the international gold standard minimizes the disturbing effects which the plurality of national currencies can have upon international commercial and financial relations, and provides the best practicable approximation to a world currency. This result is achieved because the price of gold is fixed in all countries having adopted the gold standard and because (and as long as) it is a commodity enjoying free trade. Under these circumstances it is a commodity which can always be sold abroad at a fixed price and therefore whenever there is a deficit in the balance of payments this deficit can be temporarily compensated by additional sales of gold. This is another way of describing the compensatory effect of international gold movements and it helps, I believe, to place the function of these movements in its proper perspective. It is needless to repeat here what has been said in the preceding chapter about the need for gold-flows from country to country to be reversed in due time if the system is to go on working.9

The last observation touches upon a very vital aspect of the international gold standard. There is surely a great need, if the system is to go on, for a proper relation between the amount of gold that is available to meet momentary deficits in the balances of payments and the size of these deficits. This is a more important relation than that between “gold reserves” and the total note issue. It clearly follows from the preceding analysis that while convertibility is necessary in order to keep the price of gold stable, the principal occasion for converting notes into gold results from the necessity of shipping gold abroad. It is ever more realized to-day that internal monetary stability is not guaranteed by the fact that there is a certain relation between gold reserves and the note circulation; the maintenance of a minimum reserve may prevent a run-away inflation but serious monetary disturbances may occur without reserves falling below their minimum level. I am not making a case against minimum reserve requirements—I am merely stressing the fact that their significance should not be exaggerated. What must be made quite clear is that the function of gold reserves is not to “guarantee” the “value” of bank notes, but to make it easier to maintain international monetary stability.10 Once this is realized, more importance will be attached to the relation between the size of gold reserves of central banks and the size (measured in money units) of international transactions, and particularly, of deficits in international payments. Now, clearly, no gold reserves will be sufficient in the long run if deficits in balances of payments are lasting and gold losses cumulative. Unless the mechanisms of re-equilibrium can function, the gold standard must sooner or later break down, together with international monetary stability. Gold, in order to carry out its function, must circulate, not accumulate; whenever the latter happens instead of the former, one may safely infer that the mechanism of re-equilibrium does not function correctly.11

The notion of “a proper relation” between gold reserves and the size of deficits in international payments is not a precise one and no attempt should be made to express it by some numerical magnitude. All that can be said is that if values involved in international transactions should get out of proportion with the value of available monetary gold the system may break down. Now, the value of international trade depends not only upon quantities of goods traded but also upon prices of these goods, while the size of movements of short-term funds (to the instability of which attention has repeatedly been directed in the preceding chapters) depends, among other factors, upon the prevailing conditions of confidence and upon the monetary organization. In order to keep the values of international transactions in line with the value of gold reserves, it is therefore necessary: (a) to prevent cumulative tendencies of prices to rise, i.e. every form of price inflation; (b) to discourage both the accumulation and the international movements of large short-term balances. It will be observed that neither of these requirements has been fulfilled in the first post-war decade. We shall revert to the second of them in the section dealing with the gold-exchange standard, concentrating our attention here upon the question of prices. The larger the monetary expansion based on a given stock of monetary gold, the higher the prices and, ceteris paribus, the higher the value of international trade. Thus monetary expansion leading to a rise of prices and carried out on the basis of a constant gold reserve makes the system more vulnerable. While it does not matter much from the national point of view how large gold reserves are, it is important in the interest of international stability that they should not be too small. Hence the question of reserve requirements receives a new and different justification. By making it impossible for the monetary circulation to rise above a certain multiple of gold reserves held by the monetary authorities one limits the possibility for prices to rise and thus makes it easier for gold reserves to meet disequilibria in international payments.

Now, as we know, the size of gold stocks depends upon their physical volume and upon the price of gold. By altering that price one could make monetary inflation possible while staying on the gold standard, and this on a world-wide scale. Thus the gold standard is a safeguard (at least a moderately effective one) against inflation only if the price of gold is kept constant over long periods of time. The “adjustment” of the price of gold of the kind suggested in Chapter IV12 can only be advocated as an emergency measure adopted to meet important modifications in the production of gold.

“Gold-standard inflation” is conceivable in four cases, of which a change in the price of gold, carried out concurrently by the various countries, is one. This case presupposes a substantial measure of co-operation between central banks and governments. A second case, also referred to, is that of a general lowering of the ratio between gold reserves and the monetary circulation. This too presupposes central bank co-operation. This case is different from the first in that it weakens the resistance of the system to disequilibria in international payments.

The third case results not from co-operation but from its absence. It is the consequence of large accumulations of gold in some countries due, in turn, to the failure of mechanisms of re-equilibrium to operate. We find here one of the symptoms of disorganization of international economic relations; “sterilization of gold”, carried out in response to such accumulations by countries which do not wish to experience the inflation that would otherwise follow, is another such symptom. We shall discuss them in the next chapter dealing with problems resulting from economic nationalism.

Finally, the fourth case is the consequence of a large increase in the production of gold. This case is the converse of that in which gold is getting “scarce” and a price deflation impending; and, as in that other case, the remedy might be found in a world-wide change of the price of gold.

It follows that co-operation between central banks may be very useful, but that it may also be harmful as an instrumentality of world-wide inflation. Hence if one advocates it, one must supplement the plea by pointing out the objects assigned to co-operative and co-ordinated policy. The following principle recommended by the Macmillan Committee may well be quoted in the present context:

“It should be an object of policy to secure that the international gold standard should bring with it stability of prices as well as that it should guarantee stability of exchange.”13

This recommendation combines the international and internal objectives of the gold standard; the Macmillan Report rightly emphasizes that the two are quite compatible with one another.

We have seen that the problem of “adequate” gold reserves acquires a new significance when one considers the international function of gold as an instrument of meeting temporary deficits of balances of payments. The foregoing analysis needs however to be supplemented by some additional considerations.

(1) It has been demonstrated that the supply of monetary gold has only an indirect influence, and an indeterminate one, upon the formation of prices. It is therefore quite mistaken to consider gold stocks and gold production as determining factors of the “level” and variations of prices. Nevertheless gold movements must be of a size corresponding to disequilibria of international payments; hence the above-mentioned relation between the size of gold reserves on the one hand and prices as one of the factors affecting the value of international transactions, on the other. This consideration supplements our findings in Chapter IV. While it is quite possible that the value of gold stocks should fall in comparison with the volume of monetary circulation and with the various money-values, and while this is not in itself abnormal, or, in Professor Rist’s phrase, contrary to “the nature of things”, it follows inevitably that the system becomes less resistant to disequilibria in international payments and less shock-proof. It is this consideration, and this consideration only, that constitutes an argument for having as large gold reserves as possible, and not only in one or in some countries, but in all the countries having adopted the gold standard.

(2) Since the principal monetary function of gold is that of meeting deficits in international payments, it is quite in keeping with the conception of an international gold standard that convertibility should be limited in such a way as to discourage “internal drains” of gold without obstructing international movements of the metal. This was achieved after the war by means of suppressing the convertibility of notes into coin while retaining the obligation for the central bank to buy and to sell gold in bars (the so-called “gold bullion standard”).14 This is usually presented as being one of several measures intended to “economize” the use of gold. This question of gold economies will be dealt with presently; here I only wish to stress the fact that the transition from the “gold coin standard” to the “gold bullion standard” is in complete accordance with the views that are developed here about the internal and the international significance of the gold standard.

(3) Whether minimum gold reserve requirements should be maintained or not in a modern gold-standard system is a question which has been much argued but has hitherto remained unsettled. The argument in favour of their maintenance consists in the limiting effect which minimum requirements exercise upon the volume of monetary circulation. This result could be, however, obtained through appropriate central-bank policies, while there is much scope for credit expansion even under the system of minimum reserves. It is further argued that it is necessary to maintain the minimum reserve requirements in order not to shake the confidence of the public and avoid a crisis of confidence. This argument is quite forceful as regards the suppression of minimum reserve requirements carried out independently by some one country; one might doubt, on the other hand, whether an internationally co-ordinated action would bring about the same unfavourable results.

Arguments against minimum reserves go rather more deeply into the structure and operation of the monetary system. If fractional gold reserves are held against central-bank money, then the conversion of notes into gold results in a more-than-proportional reduction of the note circulation, since the legal ratio must be maintained; this makes it necessary in certain circumstances to increase the rate of discount beyond what would otherwise be its level. Thus the fact of making a net payment abroad causes a diminution of the national money supply that is larger than the payment made. Conversely, in the country receiving a net payment in gold the total amount of circulating medium can be increased by more than the amount received. Furthermore minimum reserve requirements are not the same in the various countries, so that international gold movements not only tend to create disproportionately great changes in national money supplies, but also affect the total amount of money in the world and its international distribution.

This type of argument is developed in a particularly striking way in Professor Hayek’s well-known book on Monetary Nationalism and International Stability where it is combined with an analogous argument against fractional reserves of central-bank notes held by commercial banks against deposits. The “ideal” system would be, according to Professor Hayek, one in which there would be in every country a completely homogeneous currency with a 100 per cent gold backing.15 He realizes however its impracticability and suggests as alternatives the two following remedies:

(1) To adopt a system of “international par clearance” in order to secure an absolute fixity of exchange rates,16 and (2) to make the gold reserves of all the central banks “large enough to relieve them of the necessity of bringing about a change in the total national circulation in proportion to the changes in their reserves”, in order “that any changes in the relative amounts of money in different countries should be brought about by the actual transfer of corresponding amounts from country to country without any ‘secondary’ contractions and expansions of the credit superstructure of the countries concerned. This would be the case only if individual central banks held gold reserves large enough to be used freely without resort to any special measures for their ‘protection’.”17

With these suggestions I fully concur. Let us add that the price of gold can be so fixed by international agreement as to make gold reserves sufficiently plentiful and that if minimum requirements were abolished there would be less reason than ever for premature “protective” policies. We are thus led to the conclusion that, while gold reserves should be as plentiful as possible, no minimum ratio should be fixed.

(3)

Much has been written about “rules of the game” of the gold standard though no list of such rules has ever been satisfactorily compiled. The reason for it is perhaps best expressed in the following quotation from the Macmillan Report.18

“It is difficult to define in precise terms what is implied by the ‘ rules of the game ’. The management of an international standard is an art and not a science, and no one would suggest that it is possible to draw up a formal code of action admitting of no exceptions and qualifications, adherence to which is obligatory on peril of wrecking the whole structure. Much must necessarily be left to time and circumstances.”

In spite of these reservations, the Report recognizes that “there are ‘rules of the game’ which, if not observed, will make the standard work with undesirable, rather than beneficial consequences”.19 Where now are we to look for these rules?

In the first place it is important not to confuse “rules” with objectives of the system. If “rules” are not adhered to the system gets into difficulties and collapses; on the other hand the system may be more or less well fitted to achieve various objectives.20 “Rules of the game” means: principles according to which the system is to be administered. Now the system is in itself very clear: (1) fixation of the price of gold and its maintenance through central banking policy and (2) freedom of international movements of gold. Whether the first result is obtained by converting notes into gold coin or by selling only gold bars against notes, is an important but a technical matter, and there are many more technicalities of central bank administration, and of the gold trade which are not negligible in the management of the system. But is this what we have in mind in inquiring after the “rules of the game”? I think not. We have seen that a condition which must be fulfilled if the gold standard is to go on working consists in the reversibility of gold-flows that take place between countries. If this condition is not fulfilled the system begins to work badly, the phenomenon of gold maldistribution and that of price disequilibrium appear, and finally the system breaks down under the strain, not without there having first been more or less important internal and international frictions. Thus the “rules of the game” are conditions which must be preserved and policies that must be adopted in order to maintain the reversibility of the international gold movements. Here we meet problems that have been analysed at length in Chapter VIII, for this reversibility of gold movements is nothing else than a symptom of good functioning of the mechanism of re-equilibrium of international payments. Thus we get back to the fundamental proposition of international monetary economics, i.e. that the maintenance of a system (of any system!) of monetary internationalism depends upon the smooth functioning of the mechanism of re-equilibrium of international payments. Any monetary system aiming at the maintenance of international stability must be administered in such a way and under such conditions as make the working of these mechanisms possible and easy. It follows from the preceding chapter that these conditions are not limited to the monetary system but include general economic policies as well. The latter may even be the most important of all. Thus the problem of the gold standard is closely linked with the general problem of maintaining stability in international monetary relations. What differentiates the gold standard from other systems (hypothetical or real) of monetary internationalism, working on the basis of stable exchanges, is the fact that the mechanism of adjustment is set into motion in response to movements of gold across national boundaries, rather than to some other indicator of disequilibrium. Once the mechanism begins to operate, its action is, as we have seen, essentially the same in all monetary systems. Any policy which helps the adjustment is conforming to the “rules of the game”, while any policy which obstructs the mechanism of re-equilibrium is contrary to these “rules”.

(4)

Reference has been made already to the policy of introducing economies in the monetary use of gold, one consequence of which was the adoption of the so-called gold bullion standard in the course of the post-war monetary reconstruction. Much has been written about the “new gold standard” of the ’twenties;21 its principal distinctive features are gold economies and the spread of the gold exchange standard. The latter will be analysed in a later section of this chapter.

The monetary reconstruction of the ’twenties has been carried out in a situation which was both unfavourable for the introduction of a system of monetary internationalism and very unstable. Since I am not proposing to discuss in this book recent monetary history, I shall only mention some essential characteristics of that situation in order to place the problem of “economizing gold” in its proper perspective:

(1) Economic nationalism was steadily growing.

(2) There was a considerable maldistribution of gold in the world.

(3) Gold production was slowing down.

(4) Prices in gold-standard countries were very much higher than in 1914, yet the price of gold remained unchanged in terms of some of the principal currencies.

It is in these circumstances that the call for making economies in the monetary use of gold has become loud and widespread. The situation would have looked very different, of course, if one had proceeded from the outset to increase the price of gold and to create conditions favourable to the resumption of its international circulation. However, this would have required the adoption by the United States of two unpopular measures, the need for which would hardly have been understood. One of them was a radical change of commercial policy away from protectionism; the other, a devaluation of the dollar in terms of gold. Simultaneously the pound would also have been devalued instead of being forcibly brought back to the pre-war parity. The United States would have consented to lose a substantial part of its gold holdings. Only after all this had been done would the time have been ripe for reconstructing the gold standard. To be precise, the question of devaluing the dollar in terms of gold could have been left in abeyance until after the return to a “normal” international circulation of gold, and carried out only if the prophesied “scarcity” of that metal had really been observed.

In the actual course of events, however, nothing was done to make the functioning of a system of monetary internationalism possible. The gold standard has been reconstructed amidst growing economic nationalism, while the “maldistribution” of gold not only failed to be remedied but grew steadily worse. Instead economies in the use of gold were introduced and at the same time great stress was laid upon the maintenance through central-bank policy of a stable “price level” in the gold-standard countries. The two principal economies in the use of gold consisted in the adoption of a gold bullion standard and in the generalization of the gold exchange standard. The “new gold standard” was devised mainly at the Genoa Conference of 1922.22

Now I wish to argue that the new system was intrinsically less stable, more vulnerable than the pre-war gold standard and that this would have been true even if the gold exchange standard—a particularly great cause of instability—had not become widespread. If one considers the “level” of prices, the maldistribution of gold and the volume of short-term funds available for international movements, gold reserves were indeed in most countries relatively small as compared with the volume of international transactions. Now this would not have mattered much under fairly normal conditions, but international economic stability has never been strongly rebuilt in a world of expanding protectionism. Gold circulation of the pre-war type has not been re-established; I mean circulation of the metal which would set into motion the mechanism of re-equilibrium described in the preceding chapter. On the contrary, the idea of stabilizing the “price level” led to the adoption of open-market operations of a type exactly opposed to those which would have been required for the maintenance of a long-run equilibrium in international payments. As a result of the policies adopted the monetary circulation was prevented from falling and the bank rate from rising in the countries losing gold, while the monetary circulation was kept up and the bank rate down in countries receiving an inflow of the metal.23 Under these conditions mechanisms of adjustment of balances of payments could not, of course, work properly, and deep-rooted maladjustments inevitably developed.

The doom of the “new gold standard” was brought about by the obsession of an impending “scarcity” of gold24 and by the attempt to achieve price stability against the “rules of the game” instead of achieving it in accordance with and under the limitation of these rules.

(5)

It has been pointed out in the preceding chapter25 that international price “equilibrium” is a necessary feature of the long-run equilibrium in international payments based on fixed parities between the various currencies.26 As we have seen, there is no a priori criterion of the existence of an international price “equilibrium”; neither should that notion be considered as representing a precise numerical relation. All it means is that if price developments in the various countries follow divergent lines, this is likely to be reflected in the competitive position of various national industries in international markets. This will result in changes in the orientation of international trade and finally in a chronic disequilibrium of international payments. In order to avoid such a development, it is necessary that price movements that take place in the various countries should be “co-ordinated”. Now all this applies to the gold standard, just as it would apply to any other monetary system designed to maintain equilibrium in international payments on the basis of fixed parities.

The widespread use of the notion of “price level” and of index-numbers designed to measure numerically its fluctuations, and the adoption of the theory of purchasing-power parities, have considerably contributed to the presentation of the question of international price relationship in an incorrect perspective. The first misapprehension concerns the identity of prices. Clearly neither prices nor, indeed, their “levels” need be the same all over the world in order that the international price “equilibrium” should be realized. Not only do we know empirically that this was never the case during the years of successful operation of the pre-war gold standard, but it is obvious that it could not happen. The identity of prices throughout the world exists only for goods having a world market, that is, subject to international arbitrage, and even there, allowance must be made for the cost of transportation and so forth. The existence of import tariffs carries us further away from an identity of prices, while other restrictions to trade, such as import quotas, tend to destroy the world market for the commodities to which they are applied and the very notion of a “world price” of such commodities. Furthermore, there is a great variety of goods which never or very rarely enter international trade and for which no price-arbitrage is possible. Thus there is no reason to expect that prices quoted throughout the world for individual commodities or services are equalized through appropriate rates of exchange between the various currencies; and since this is true of individual prices, it must be equally true of statistically constructed “price levels”. The theory of purchasing-power parities has been discussed in Chapter VII; its great danger lies in the fact that it furthers the idea that “price levels” can and must be equalized by exchange rates and that whenever this does not happen some currency is either undervalued or overvalued in relation to others. This view does not stand up to a careful examination.

Since such is the situation, what are we to understand by “international co-ordination of prices”? We know that it does not mean “identity of prices”. Nor does it imply a complete parallelism in their movements. On the contrary, the functioning of mechanisms of re-equilibrium may require that short-term price movements should take place in opposite directions in the countries concerned. As has been pointed out before, these movements are not likely to have a considerable amplitude, nor to cause internal economic disturbances in countries in which they occur.

We are thus left with long-period tendencies of prices to rise or to fall or to be stable. The notion of co-ordination of prices in the various countries implies that these long-run tendencies should be the same in all the countries; the gold standard provides a mechanism of keeping price movements in the various countries in step with one another.27 This happens by means of gold movements taking place in response to the situation of the balance of payments and influencing monetary policy. An outflow of gold would be likely to ensue in a country which adopted a policy of monetary inflation, and if that country follows the “rules of the game” the outflow of gold must bring that inflationary policy to an end and even cause a restrictionist policy to be adopted in its stead. Similarly, to envisage but one among many other possible combinations, if all the countries but one should go in for a parallel monetary expansion, that one country will experience an inflow of gold in response to the development of its balance of international payments, and if it follows the “rules of the game” it will have to adopt in turn a policy of expansion. Thus, as long as its currency is based on an international standard, a country cannot adopt a monetary policy considerably at variance with that of other countries. What it amounts to is that under a system of monetary internationalism, such as the gold standard, money values (i.e. price structures) in the various countries are co-ordinated and a mechanism is provided to keep them co-ordinated. This brings us closer to the situation that would exist if there were a unified monetary system throughout the world. Since, however, monetary policy is a prerogative of national authorities, there must exist, if that co-ordination is to be achieved and maintained, an agreement determining the “rules of the game”, i.e. appropriate policies to be adopted in the various circumstances. Whenever there is no international co-ordination of national price structures, movements of short-term funds are likely to have disturbing rather than equilibrating effects upon the balance of international payments; this is particularly true whenever stability of exchange rates is sacrificed to the pursuit of “independent” national policies.28 International monetary instability inevitably follows. Thus we are once again facing the choice between monetary stability based upon an international co-ordination of national policies and of national price structures, on the one hand, and monetary instability combined with national “independence” in the field of economic policies, on the other.

It is possible under a system of monetary internationalism that a country which is an important monetary and financial centre should exercise a more or less considerable influence upon price developments in other countries provided that its monetary management enjoys confidence abroad. Such was, in the main, the position of England in the pre-war gold-standard world. Whenever all the countries do not tacitly accept the intermediate and long-run price policies which have been adopted by one of them, an international agreement must be reached on these matters lest international monetary stability may be endangered. This is why in the post-war world international monetary co-operation became so important a problem, while it was but little discussed before 1914. The “great depression” proved how an absence of agreement on basic lines of policy results in a disintegration of international monetary relations. If an economic depression is met by some countries with a counter policy of monetary (and economic) expansion and by other countries with a policy of “letting deflation take its course”, monetary stability can hardly be expected to last. It is interesting to note that in the years 1931–5 the exchange rates were fairly stable (and could have been quite stable) within the group of countries which adopted an expansionist policy and within the group of countries which followed a policy of deflation. While there was no absolute parallelism of price developments within either of these groups, the co-ordination was sufficient to maintain that exchange stability.29 The fact that one group of countries was on the gold standard and that the other was off the gold standard is here of secondary importance. It would have been quite possible to adopt an expansionist policy on the gold standard had the development of the financial position of Great Britain between 1925 and 1931 been healthier, had sterling not been forced off the gold standard, and had an agreement been reached between England and the United States on methods of fighting the depression. Even in 1933 it would have been possible to have a policy of monetary expansion in the gold-standard countries had the United States remained on the gold standard and had all the countries on that standard “played the game” according to its “rules”. In this case prices would have risen in the United States and gold would have flowed out of the country; in the other gold-standard countries gold reserves might have increased and the monetary policy would have accordingly been influenced; an agreement between the United States and France would, of course, have made things still easier.30 The French deflation of 1933–6 was mainly due to the desire to keep the price of gold at its former level, while its price in terms of the other leading currencies was increased.31 Paradoxically, countries which stayed on the gold standard after 1933 and kept the old parities were carrying out an “independent national monetary policy”, while the “sterling area” and the United States (after January 1934) became the new nucleus of monetary internationalism.

All this goes to show that the problem of monetary internationalism is wider than that of the gold standard and that, since under the gold standard various intermediate and long-run monetary policies can be adopted, it is essential to reach an international agreement about the aims of these policies.32 The gold standard provides by itself only a mode of organization and an instrument of action.

Too much passion has been put in recent years into the discussion of whether to have the gold standard or not. The real problems are: (1) what should be the objectives, under a régime of monetary internationalism, of long-run and intermediate-run price policy, and more fundamentally, (2) whether one should adopt a régime of monetary internationalism with the “discipline” which it calls for, or a régime of national “independence” of monetary policies with the instabilities to which it is likely to lead.

(6)

Let us conclude this rapid survey of the international gold standard by briefly discussing the problem of an international gold clearing, which represents perhaps the future mechanism of the gold standard. The theory of this proposal is very simple: instead of shipping gold from country to country, which involves transportation costs and risks, why not make gold payments by means of earmarking gold in favour of the receiving central bank without moving it? If in addition an institution were entrusted with the function of keeping gold accounts for the various central banks we should obtain in the field of international monetary relations an organization comparable to that which exists within a country, where the central bank acts if necessary as a clearinghouse for the various banking institutions. The British clearing system is particularly enlightening.33 The idea is not quite a new one; it became a practical proposition since the creation of the Bank for International Settlements, an institution which is extremely well fitted to play the rôle of a gold clearing-house. It will be noted that the “Young Plan”, submitted in June 1929, which is the basis of the Hague Agreements of January 1930, not only included proposals for the creation of the Bank for International Settlements but suggested at the same time that this new institution might vitally contribute to suppressing the costs and the risks inherent in shipping gold from country to country. Accordingly the Bank was authorized34 to “(a) buy and sell gold coin or bullion for its own account or for the account of central banks; (b) hold gold for its own account under earmark in central banks; (c) accept the custody of gold for account of central banks”. Thus the set-up for an international gold clearing came into existence; if the system were to be fully developed gold shipments would be replaced by entries in the books of the Bank for International Settlements.

The first consequence of adopting the “gold clearing standard”35 would be the reduction of the cost of moving gold from place to place, since “materially” gold would not move at all; it would remain where it happened to be, only it would be earmarked for the Bank of International Settlements which in turn would hold it for the account, now of one, now of another, central bank. This consequence is however not the principal one, and would not in itself justify the adoption of a system involving a considerable reform of actual practices. The most important consequence of introducing the “gold clearing standard” would be the possibility of suppressing the gold points and eliminating even those small fluctuations of foreign exchanges which exist under the gold standard.36 By this means one would come even closer to an international currency than under the traditional gold standards and, as Professor Hayek justly remarks,37 “one of the main causes of international movements of short-term funds” would thus be removed.38

This suggested reform of the gold standard is diametrically opposed to that which consists in making possible a wider range of exchange fluctuations than is the case in the traditional system. The most important proposal of the latter type is that, due to Mr. J. M. Keynes, of widening the gold points.39 Which plan is the more useful? The answer must depend upon the validity of the argument advanced in favour of allowing greater fluctuations of exchange rates to take place under the gold standard. It is argued that price structures have grown much more rigid since the beginning of the century and that it is impracticable, therefore, to throw the whole burden of adjusting balances of payments upon price movements; a part of the adjustment should be made through fluctuations of exchanges. The practical consequences of these wider fluctuations would be minimized by resorting to forward exchange operations.40

The last point is, however, inaccurate. Forward transactions involve the payment of a “premium”, the size of which depends on market anticipations. The fluctuations of the “premiums” contain in themselves an element of uncertainty for trade which must not be disregarded.41 It must also be observed, and this is a very important consideration, that the greater is the range of possible fluctuations of exchange rates the greater the incentive for speculation, i.e. for movements of short-term balances of a usually disequilibrating character.

Most important of all is, however, the question whether the basis of the whole argument is valid or not.

It is quite true that the price and cost structure has become increasingly rigid through the action of trade unions and of cartels. This makes it impossible, or at least very difficult, to make downward price adjustments of any considerable size. But under what circumstances are such adjustments called for? We have seen that there is a strong theoretical presumption against the contention that such adjustments are necessary for the normal operation of mechanisms of re-equilibrium; the matter will be further clarified when we have better factual information about the price movements that actually took place under the pre-war gold standard in response to movements of gold from one country to another. If this is granted there remain only adjustments to exceptional situations, such as result from an arbitrary fixation of a too high parity for a national currency, or from devaluations carried out abroad, or from exceptionally important downward price movements abroad or from a previous inflation at home. Thus the adjustments under consideration fall into one of two categories: (a) adjustments to former exchange manipulations; (b) adjustments to a situation resulting from the disruption of the international price “equilibrium”. Now there is no reason why in these two groups of cases the adjustment should not be made by the means of a change in parities; it will however be observed that the need for adjustment would not have arisen in the absence of former violations of the rules of monetary internationalism. Here we certainly face a situation which it is better and easier to prevent than to cure.

If we leave aside these situations resulting from economic and monetary nationalism and from a lack of international co-operation and consider the “normal” conditions of monetary internationalism, the need for large price adjustments does not seem to arise. Thus the argument in favour of more widely fluctuating exchanges42 loses its main supporting element. In that case one may hold that even the minor fluctuations of exchange rates which take place within the limits of the “traditional” gold points can be dispensed with and the “gold clearing standard” appears in its full importance. It must be noted that under absolutely fixed exchanges, international payments in gold will be probably larger than under the pre-war gold standard. This cannot have much importance however if the “rules of the game” are adhered to, i.e. if mechanisms of re-equilibrium are allowed to operate and if gold circulates (in the books of the Bank of International Settlements) instead of getting accumulated.

A limitation on the system of gold clearing is obviously imposed by the state of international confidence and of international relations in general. The fear of or preparation for war makes national authorities unwilling to keep abroad gold which in case of war may be most needed for making international cash payments. This is probably the principal reason why the gold clearing operations of the Bank of International Settlements, grown though they have, remain up to now very limited.

B. Exchange Standards

(1)

The foregoing analysis can be applied, mutatis mutandis, to any metallic standard. Let us now briefly examine the consequences that result from the adoption, as monetary standard, of the currency of another country instead of a metal.43 In such a case the monetary system will possess the following characteristics (in accordance with the definition of “monetary standard” used in this book):

(1) The monetary law of the country fixes the price of the foreign currency adopted as standard in terms of the national currency; in other words the parity between the national currency and that which is chosen to be its standard is defined by law;

(2) The monetary authorities (usually the central bank) of the country in question are entrusted with the task of maintaining that price (i.e. that parity). In order to do so it is indispensable to make the national currency convertible into the standard currency, while the central bank (or another competent authority) must buy at the legally determined price any amount of the standard currency that is offered to it. In order to be able to do so it must hold reserves of the standard currency. The analogy with relations that exist under the gold standard is striking; this was to be expected since these relations follow from the adopted definition of the monetary standard.

(3) Deficits in the balance of payments are met by means of payments in the standard currency; a diminution of the standard currency reserves held by the central bank follows. Surpluses in the balance of payments result similarly in an increase of these reserves.

The first two points do not call for any special comments, while with reference to the third point we must further examine the operation of Exchange Standards.44

It will be observed, in the first place, that monetary reserves will be held on deposit in the country whose currency is adopted as standard. Thus, instead of gold reserves, the central bank will hold balances of a foreign currency. Since the net payments resulting from the position of the balance of payments will be made or received in the standard currency, inflows and outflows of gold will be replaced respectively by increases and decreases of the balances of standard currency held by the central bank. Let us assume that a country A, is on the sterling standard and examine the process of international settlements in three stages.

(1) Settlements between country A and England. It is clear that if A’s payments to England exceed England’s payments to A the balances held in London by the central bank of A will be reduced, and that in the contrary case they will be increased.

(2) Settlements between country A and other countries which are on the sterling standard. Net payments will be made by means of transferring in London sterling balances from the account of one central bank to that of another. If all the countries on the sterling standard hold their monetary reserves at the Bank of England, that Institution will become a clearing-house for international settlements (limited to countries on the sterling standard). Thus an international sterling standard would be a sterling clearing standard.45

(3) Relations with countries not on the sterling standard. Among various theoretical possibilities let us choose the following one which might have the greatest practical importance: let us suppose that the currency chosen as standard by some countries is itself based on some standard; what will be the process of settlement with other countries which are on that latter standard? To make this very general enunciation clearer, let us assume that in our example the pound sterling is on the gold standard; what will then be the relations between sterling-standard countries and the other gold-standard countries? Since the sterling balances can be, under such conditions, converted into gold, a net payment to be made by sterling-standard countries to gold-standard countries will result in an outflow of gold from London, while there will be an inflow of gold to London in the opposite case. The net change that will occur in the size of the gold reserves of the Bank of England as a result of that association will correspond to the net aggregate payments or receipts of all the sterling-standard countries.

(2)

An international exchange standard assures the fixity of exchange rates, not only between each of the currencies based on that standard and the currency chosen as standard (which follows from our definitions), but also between the currencies of the various countries having adopted that standard (similarly to the position under the gold standard). There is, of course, in this system nothing comparable to “gold points”, and unless special measures are taken to the contrary,46 no fluctuations of exchange rates can take place between currencies based on the same exchange standard. In this respect the system operates more like the “gold clearing standard” than like the traditional gold standard.

It has been repeatedly pointed out in the preceding pages that mechanisms of re-equilibrium must work in essentially the same way under any monetary system. What varies from one system to another is the nature of the stimulus which sets the mechanism into motion. Under the gold standard it is the flows of metal in and out of the country which act as “starters” for the process of adjustment. In the present case it will be the changes that take place in the balances of the standard currency that are held by the respective central banks. As regards the measures to be adopted in response to an increase or to a decrease of these balances, they neither need be nor can be different from those adopted under the gold standard.

The last observation applies to relations between countries which have adopted the same exchange standard for their currency. But what about relations between any of these countries and the country whose currency is chosen as standard? To use our former example, what about the relations between the country A and England, the A currency being on the sterling standard?

For A the situation is clear and it is the same one as in the former case: changes in the size of its sterling balance will induce the monetary authorities to adopt measures intended to restore equilibrium in A’s international payments. As regards England the situation is more complex, since, under our assumptions, it belongs to two currency groups, of one of which it is the centre (sterling-standard countries) and of the other of which it is a member (gold-standard countries). Now in England’s relations with countries on the sterling standard, disequilibrium in international payments will result either in the growth of sterling balances belonging to the latter countries, or in the depletion of these balances. This will result automatically in an expansion or in a restriction of England’s monetary circulation provided that two conditions are fulfilled: (1) there must be no “offsetting” operations; (2) sterling balances held in London by central banks of the sterling-standard countries must be “idle”. A comparison will make the situation clearer. Let us compare balances held in London by a sterling-standard country to gold deposited with the Bank of England under earmark by a foreign central bank. Clearly such gold cannot be allowed to affect the monetary circulation of England; what may affect that circulation, however, ceteris paribus, is an addition to gold earmarked (on account of England’s deficit in international payments) or a diminution of the amount of that earmarked gold (on account of a net payment to be made by the country in question to England). The same applies in the case of exchange-standard countries holding their monetary reserves in the financial centre the currency of which they use as a standard for their own.

It follows that an exchange standard, if properly operated, is a monetary system not less costly than the gold standard.47 It must be realized that the maintenance of fixed exchanges is an objective which demands the maintenance of monetary reserves, i.e. of a stock of the substance chosen to be the monetary standard, and this involves costs, since monetary reserves are an asset which yields no income. If monetary reserves are invested in order to become an earning asset, not only do they cease to be liquid, which may endanger monetary stability, but the mechanism of re-equilibrium in international payments is prevented from working with its full efficiency. If, in our example, A’s sterling balances kept in London are invested, their increase or reduction will have no effect upon the monetary situation of England; thus the burden of adjustment weighing on the country A will be increased and in due time serious difficulties may arise. These observations will be recalled presently when discussing the “gold exchange standard”.

The analysis of international price relationships under the gold standard can a fortiori be applied to exchange standards. The country which administers the currency chosen as standard by other countries enjoys a position of undoubted leadership. If that country pursues a price policy of which the “members” of the “system” do not approve, and from which they wish to dissociate themselves, that “system” will ultimately break down. Its collapse may cause serious difficulties to the country whose currency serves as monetary standard to the others, since it may involve a rapid conversion of balances held in that “standard currency” into other currencies or into gold. In either case, movements of short-term funds will take place and the monetary stability of the country from which the function of monetary leadership has been taken away may become very precarious indeed.

Thus exchange standard systems require, even more than does the gold standard, some international agreement on national monetary (and, generally, economic) policies. If mismanaged, the system may collapse into complete chaos. Mr. Hawtrey is right when he observes that “an exchange standard works very well so long as conditions in the foreign countries in whose currencies the reserves are held are normal, but there can be no certainty that this will always be so”.48

Thus in addition to the two conditions stated before,49 one must point out that under an International Exchange Standard the country whose currency is chosen as standard cannot have a full sovereignty over its monetary policy, but must act in agreement with countries taking part in that “system”. The disintegration of the system is the penalty for violating this rule.50

(3)

Let us complete the foregoing discussion with the two following observations.

Exchange standards depend, just as does the gold standard, upon a precise determination of the price of the standard in terms of the national currency. If a de facto relation is maintained by the central bank (or another monetary authority), but subject to administrative alterations, we shall speak of a currency linked to another, not of the latter being the standard of the former. Thus the “sterling area” which developed after England went off the gold standard was a group of countries linking their currency to sterling and not a group of countries on the sterling standard. The transition from the former situation to the latter would have been obtained if the various countries belonging to the “sterling area” had adopted legal parities between their currencies and the pound sterling.

Another important distinction must be made between adopting an exchange standard or linking, more loosely, one currency to another, and making use of foreign-exchange operations as an instrument of monetary management. As Mr. Keynes rightly points out, one can adopt what he calls “exchange management” in a monetary system based on an exchange standard, but also in other monetary systems.51 Thus the gold standard (or another metallic standard) can be managed and often is managed with the help of foreign-exchange operations carried out by the central bank on the home market and on foreign markets. To use the terminology adopted throughout the present book, “exchange management” consists in short-term credit operations, i.e. in transfers of short-term funds from one currency into another. Under the gold standard this may reduce the size of gold movements required to make the necessary adjustments in international payments. “Exchange management” is a planned intervention carried out by the central bank on foreign-exchange markets; in order to be able to adopt this method of action a central bank must hold balances in foreign currencies; in doing so under the gold standard, one does not come any nearer to the exchange standard, since gold remains both the standard and the ultimate means of making international settlements.52

C. The Gold Exchange Standard

The gold exchange standard occupies a rather peculiar position between the gold standard and exchange standards. Historically it developed as a method of operating the gold standard; this evolution, however, rendered it in many respects similar to exchange standards. Under this system it is gold which remains the monetary standard, since it is the price of gold and not the price of a foreign gold-standard currency which is defined by the monetary law. There are, however, various important differences between the gold-exchange standard and the gold standard as defined and analysed in a previous section of this chapter. While, under the system we are now examining, the central bank is still bound to redeem its notes, it can decide whether it will redeem them in gold or in foreign exchange (of a gold-standard country). Thus under the gold exchange standard convertibility of notes into gold is often “indirect”; the central bank maintains the price of gold by buying the metal at the legally prescribed price and by selling either gold or foreign exchange convertible into gold. While one could say that the gold exchange standard consists in the exchange management of the gold standard,53 this method of management, very different from the traditional methods, introduces important modifications into the structure of the monetary system. The functioning of the gold exchange standard must be governed by the same principles as that of any exchange standard. These general “rules” have been analysed in the previous section.54 Their application to the particular case which is now under discussion can be formulated in the following way:

(1) As regards countries on the gold exchange standard:

(a) Their central banks must consider changes in their balances held in gold-standard countries (including any cash holdings of gold-standard currency which they may possess) as tantamount to changes in gold reserves. In taking into consideration the total of their reserves (gold and gold-standard exchange) their policy must be determined in exactly the same way as it would be under the traditional gold standard. In the contrary case the smooth working of the mechanism of re-equilibrium would be endangered.

(b) They should keep foreign balances used as reserves at the central banks of the respective gold-standard countries so as to be certain of their liquidity;55 furthermore an agreement should be concluded with governments of the countries where gold-exchange reserves are held, insuring the central banks who own these reserves, against any risk resulting from the monetary policy of the countries in question.56 It must be guaranteed that balances will always on demand be redeemed in gold at the price prevailing at the time when the balances were acquired, and that this gold will be free to leave the country at any time. If no such convention exists the country having adopted the gold exchange standard might see its monetary stability endangered should the country where its monetary reserves are held change the price of gold, suspend convertibility or impose an embargo on gold exports.

(2) As regards gold-standard countries in which monetary reserves of the gold-exchange-standard countries are held:

(a) The central banks may be unable to carry out a policy required by the rules of the gold standard, if countries on the gold exchange standard do not accept the principle of holding their foreign balances with the central bank. Hence it is most important to reach an agreement on this point.

(b) Since these countries are debtors of foreign central banks to the extent of reserves held in their banks, they must either keep a 100 per cent. gold reserve against these balances which can be withdrawn at any moment without notice, or else reach an agreement with the monetary authorities of the respective gold-exchange-standard countries regarding withdrawals of gold. Of course, if point (a) is not carried out and foreign monetary reserves are held with banks other than the central bank, no safeguard can be obtained against sudden withdrawals of gold. This introduces into the monetary conditions of the gold-standard countries in question an element of instability and of risk. The gold exchange standard increases the volume of international short-term indebtedness which may give rise to considerably increased movements of short-term funds from one country to another, unless counteracted by some explicit agreements. It will be observed that the agreement mentioned under point (1) (b) stands in contradiction, as to its aims, to the agreement now suggested. At the best a compromise could be reached, leaving both types of countries to face risks greater than those which exist under the traditional gold standard.

(c) It is important that balances held as monetary reserves by foreign countries should not enter the economic processes of the country where they are held. The reasons for this have been indicated in the foregoing analysis of exchange standards. Let it be emphasized that inflationary tendencies of a world-wide character may follow (and have followed in the twentieth century) the violation of this rule. If monetary reserves of certain countries are held in the form of ordinary balances with commercial banks of the gold-standard countries, a double expansion can take place on the basis of existing stocks of monetary gold: In the gold-standard country there exists the usual “inverted pyramid of credit”: gold, bank notes, demand deposits (“bank money”); now a part of the top stratum of that pyramid can become the basic stratum of another such pyramid in a country on the gold exchange standard. Every short-term credit made by a gold-standard country to one on the gold exchange standard may thus become the basis of a credit expansion in the latter without causing the least credit contraction in the former!

This is a further argument in favour of monetary reserves of foreign central banks being held exclusively with the central bank of the respective gold-standard country. It also further justifies the demand for a high gold reserve to be held against such balances. Finally it leads to a third requirement, to wit, that no interest whatever should be paid on the balances in question (since they cannot and should not be invested in any earning asset).

The fulfilment of the foregoing “rules” demands a large degree of international co-operation, much larger indeed than that which is required under the gold standard.57 One should never adopt the gold exchange standard without concluding beforehand a detailed international convention regarding the principles on which it will be operated.

The gold exchange standard as generalized after the war has added considerably to the risks of monetary instability, since not only has such a convention never been adopted (nor indeed discussed!), but the principles analysed above have been continuously and systematically violated.58 In this case, as in so many others, it is not the nature of the system which is the cause of trouble, but the way in which that system is operated. Hybrid though it is, the gold exchange standard need not be harmful if it is properly managed; under that system much harm can be done, however, if its management violates the elementary requirements of monetary internationalism. It is particularly dangerous to adopt empirically a system without analysing sufficiently its structure and determining the appropriate rules to be adopted for its administration—and yet this is what the Genoa Conference did with regard to the gold exchange standard. It was recommended, it is true, that a conference of central banks should be called; but no recommendation was made to the effect that the gold exchange standard should not be adopted anywhere before such a conference had worked out an appropriate monetary convention and before that convention had been ratified by the countries concerned; nor was it indicated that there was any need at all for determining what the “rules of the gold exchange standard” should be.

Granted that the gold exchange standard can do no harm if it is properly managed, can it do any good? In other words, why should it be adopted in preference to the gold standard? Our theoretical analysis does not provide us with any answer to that question; let: us look for a moment to monetary history. The gold exchange standard existed before the war and the best discussion of its merits will be found in the book on Indian Currency and Finance published in 1913, by Mr. Keynes, from which the following observations may be quoted:

“Gold is an international, but not a local currency. The currency problem of each country is to ensure that they shall run no risk of being unable to put their hands on international currency when they need it, and to waste as small a proportion of their resources of holdings of gold as is compatible with this . . . it is as an attempt to solve this problem that the Gold Exchange Standard ought to be judged.”59

This can be applied, no doubt, also to the post-war situation.60 The gold exchange standard was adopted in order (1) to economize on the monetary use of gold and (2) to make it easier for countries not disposing of sufficient gold reserves to return to some form of the gold standard. Let us see under what conditions these objectives can be attained and with what consequences.

In the first place it is just as easy (or just as difficult) for a country to secure gold as it is to secure foreign balances. This follows so clearly from the structure of international economic relations that it does not seem necessary to argue the point. I refer the reader to Chapters V and VI. From this point of view, therefore, it is not easier for a country to adopt the gold exchange standard than it would be to adopt the gold standard.

In the second place, it is certainly more costly to keep monetary reserves in the form of gold than it is to keep them in the form of interest-bearing assets. If a country acquires the monetary reserves which it needs in order to adopt the gold standard by contracting a foreign loan, the burden on its economy is less if it is possible to invest these reserves abroad and earn all or part of the interest that has to be paid on that loan. In this respect the gold exchange standard offers undoubtedly a financial advantage over the gold standard. There are, however, two reservations to be made; the one is that in investing monetary reserves one infringes the previously discussed “rules” of the system and renders easier both inflationary tendencies and international movements of short-term funds.61 The other reservation consists in questioning the soundness of adopting the gold standard (or gold exchange standard) on the basis of any foreign loan. If conditions of economic internationalism—conditions which are necessary for the maintenance of long-run equilibrium in international payments—are fulfilled it must be possible for a country to secure gold or foreign balances without contracting loans ad hoc. Should this, on the other hand, not be possible in a given situation, then, I think the presumption is against that country’s ability to maintain the gold (or gold exchange) standard introduced with foreign financial assistance. Recent monetary history seems to confirm rather than to contradict this view.

In the third place, it is difficult to see how, if the gold exchange standard is properly managed, the total amount of monetary gold needed in the world can be diminished, other things remaining the same, through the mere fact of its adoption. This effect can only occur if the gold exchange standard is made an instrument of inflation. If gold in country A (on the gold standard) serves as basis for the monetary circulation of that country and if one part of that circulation serves, as though it were gold, as basis for the monetary circulation of country B (on the gold exchange standard), then indeed one can “economize” on the monetary use of gold. All goes very well as long as the effects of inflation do not make themselves felt and as long as countries on the gold exchange standard do not avail themselves of the right to convert their foreign balances into gold. But since both these things are sooner or later likely to happen, the economy in the use of gold is associated with an increased instability of the monetary system. It is impossible to encourage gold economies by the means of the gold exchange standard and to avoid inflationary developments all at the same time.62 The introduction of the gold bullion standard has largely reduced the demand for gold for the purpose of internal circulation, while the lowering of minimum reserve requirements (or the abandonment of any precise lower limit to these requirements) can effect real “economies” without having these unfavourable consequences on the stability of the system.

In view of the discussion above, there seems to be no positive feature of the gold exchange standard which neither the gold standard nor an exchange standard would possess, while the possibilities of bad operation leading to monetary instability are likely to be greater under this system than under either of the other two.

D. Free Paper Currencies

Let us now deal briefly with some international problems arising out of the operation of monetary systems not based on any standard. “Free paper currencies” is a phrase in which the emphasis should be placed on the word “free” since, as has been pointed out before, both gold-standard and exchange-standard currencies are “paper currencies”. “Free” means in this context, “not based on any standard”. It implies, from the internal point of view, that the monetary authorities are not limited in their freedom of action by the obligation of stabilizing the price of some “standard”. But this is not a very important point, since from the exclusively internal point of view the obligation of stabilizing the price of the “standard” on some level is not an important limitation upon the freedom of action of the central bank (except in so far as it excludes the adoption of a policy of cumulative inflation). As has been pointed out above,63 the adoption of a standard for a currency introduces limitations mainly from the point of view of international relations. In the three preceding sections of this chapter we have seen how international monetary stability can be achieved under monetary systems based upon standards of various types. Since we have now to consider a system not based on any standard, the question immediately arises whether international monetary stability can be achieved at all between countries having free paper currencies.64

A discussion of the various objectives which can be aimed at by monetary management lies beyond the scope of the present inquiry. Two questions must, however, be dealt with, to wit: (a) whether durable stability of exchange rates between the various currencies can be achieved in the absence of a monetary standard, and (b) whether the long-run equilibrium of international payments can be obtained through fluctuating exchange rates.

The second question has already been dealt with in Chapter VIII and I refer the reader to that analysis.65 The following conclusion was there reached: exchange fluctuations may restore equilibrium in international payments and may provide a method for maintaining long-run equilibrium in international payments, provided that the various countries adopt policies which are in substance very similar to the gold-standard policies, particularly as regards changes in short-term interest rates. The main difference consists then in the fact that a wider range of fluctuations of exchange rates replaces international movements of gold. It has also been pointed out that in order that this mechanism should operate smoothly it is necessary that a long-run stability of exchange rates is anticipated. If this condition is not fulfilled, movements of short-term funds of a disturbing character are likely to set in and to bring about monetary instability. Thus in a régime of fluctuating exchanges there exists an abstract possibility of monetary internationalism; it depends however on three conditions:

(1) On the maintenance of a long-run stability of rates of exchange; their short-run fluctuations should amount to mere oscillations around a position of “long-run equilibrium”;

(2) on a co-ordination of price policies in the various countries so as to preserve the international price “equilibrium”;66

(3) on the proper functioning of the short-run elements of the “mechanism of re-equilibrium”.

If these conditions are satisfied, then exchange fluctuations can be allowed to take place within the framework of monetary internationalism. When we pass, however, from the “abstract model” to actual economic experience, there arise certain reservations and doubts. The principal among them concerns the usefulness of having this type of exchange fluctuations rather than their fixity. If the three above-mentioned conditions are satisfied, exchange fluctuations cannot be made to cover “autonomous” national policies; if such policies take place, exchange fluctuations will cease to represent mere oscillations and will become irreversible, one-way modifications. But exchange fluctuations of such a type are likely to provoke unsettling movements of short-term funds and disturb long-term capital movements as well as international trade. Experiences made in the course of the last twenty years provide many striking illustrations of this point. While one can admit the abstract possibility of having simultaneously both fluctuating exchanges and a system of monetary internationalism, the past record of actual monetary developments seems to indicate that this abstract possibility has a low probability. It seems to be much more likely that a choice between fluctuating exchanges and monetary internationalism is inevitable except where exchange fluctuations are limited by the operation of an international standard.67

No argument that I know of, favouring fluctuating exchanges68 is free from one of the following assumptions:

(1) that the “abstract model” corresponds with a large degree of probability to empirical conditions;

(2) that monetary internationalism is of secondary importance compared to the maintenance of “internal stability” by means of “autonomous” national policies;

(3) that exceptionally great disturbances are to be faced to which adjustment can hardly be reached without more or less important modifications of monetary parities.

As regards these assumptions, I have already dealt with the first, the second will be discussed in the next chapter since it is that of economic nationalism, and the third corresponds to circumstances which are by definition exceptional and call for substantial changes in various economic relations; they do not correspond to “normal” conditions. As has been pointed out before, it is particularly important to apply, inasmuch as it is possible, policies capable of preventing the emergence of such conditions of acute disequilibrium.69

Let us return to the “abstract model”. In order to carry it out in actual practice a stern discipline in the field of internal economic policies and a large degree of international agreement and co-operation are indispensable. If they can be obtained, then it is possible to realize with a system of free paper currencies not only a long-run stability of exchange rates but maybe even their complete fixity. Since, in my opinion, the former can hardly offer any advantages over the latter and since even the former requires a determination of long-run “equilibrium rates of exchange”70 around which the fluctuating exchanges would oscillate, let us envisage now the problem of whether and by what means parities can be determined and maintained in the absence of a common monetary standard.

In the case of a common international standard, the price of that standard is defined by law in each particular country71 and thus a tertium comparationis gets established which allows mutual parities between each pair of national currencies to be determined by a simple exercise in arithmetic. The element of arbitrariness, of “fiat”, consists in determining the price of the standard in terms of the individual national currencies.72 It will be noted, incidentally, that under a common standard (the gold standard for example), no international agreement is needed for determining parities; each country fixes the price of gold in terms of its national monetary unit.73

Now in the case of free paper currencies the process of determining parities is much more complicated and the necessary amount of co-operation very much greater than when currencies are based on a common standard. Since there is no means of defining currencies in terms of a tertium comparationis one must define parities directly, which involves a more complicated procedure and a much greater need for agreement. The difficulty is increased by the fact that there is no possible recourse to a criterion such as that provided by the concept of “purchasing power parities.”74 Thus the process must involve an all-round agreement on an arbitrary set of relations; their arbitrariness could be only reduced by the application of a process of “trial and error”. They would thus be the outcome of a more or less prolonged period of “experimentation”, during which short-term funds, moving from country to country in search of speculative gains, might delay or prevent any determination of parities. Considering how much easier it is to solve that problem if one adopts an international standard for the various currencies, one may doubt the wisdom of adopting a system of free paper currencies. This attitude is further accentuated by the consideration of the difficulties inherent in the maintenance of these fixed parities. We have seen before that it is movements of the standard from country to country which limit or suppress the fluctuations of exchange rates. There is only one alternative solution, which consists in replacing movements of the standard by movements of short-term balances.75 Thus the various central banks (or treasuries, according to the organization of the internal monetary administration) would have to keep balances of various foreign currencies as their reserves and calculate all the time the value of these reserves expressed in terms of the local currency. Variations of that total would stimulate appropriate monetary policies just as under the gold standard. Under that system not only is the handling of the mechanism more clumsy than it is under the gold standard, but it requires more mutual confidence and therefore a higher degree of co-operation between the various countries than is the case under the gold standard. The system is less easy to administer and more vulnerable, and in trying to find a remedy for these two drawbacks one is inevitably led to the adoption of an international standard. Under the conditions prevailing in the world, the fixity of exchanges is most easily achieved and most safely preserved when the various currencies are based on a common standard. It is likely, furthermore, that under a system of free paper currencies it is the claims of national “autonomy” that will triumph over those of international stability.76 Thus we may conclude that monetary internationalism is likely to be neither smoothly worked nor durably maintained on the basis of national systems of free paper currencies; and that the adoption of an international monetary standard is most important for the maintenance of the long-run equilibrium in international payments.77, 78

• 1There is much difference between the views that are widely held in this matter in Great Britain and the United States, and views that are commonly accepted on the Continent. This explains the emphasis laid on that issue. Vide infra, pp. 68 and 69.
• 2The reader may find it useful in this context to contrast the United States’ unified monetary system, now embodying a Federal Reserve System, with what would have existed had each of the 48 States of the Union sovereign powers over its particular monetary system.
• 3The word “international” is used, of course, with utter disregard of its etymological significance; but this usage is universally recognized.
• 4F. A. von Hayek, Monetary Nationalism and International Stability, London, 1937.
• 5loc. cit., p. 4.
• 6Except when a World State exists.
• 7i.e., “real” factors of production (goods), monetary factors (capital) and, thirdly, labour.
• 8“In the last analysis . . . it is the political factor which gives rise to the economic problem of international monetary transfer. If historical accident had not created independent sovereign states . . . no such ‘problem’ would have arisen”. Lionel Robbins, Economic Planning and International Order, London, 1937, pp. 279–80.
• 9To change the weight of a coin (“debasement”) is analogous to the modern forms of devaluation. While, as we shall see presently, the expression “changes in the gold contents of a currency unit” is open to criticism, it shows at least clearly the analogy in question. But what Sovereign Prince of the days gone by would have accepted an international undertaking never to debase the currency? And even if he did accept it, this would be an analogous matter to the type of international monetary convention mentioned in an earlier part of this chapter. The Latin Union was to some extent an example of a limited convention of this kind. But it all goes to show that not even a general limitation of monetary circulation to precious metals would automatically insure a system which would work internationally in the same manner as a monetary system works within a country in regulating the supply of money in the different regions.
• 10loc. cit., pp. 270–80.
• 11Cf. my “Monetary Internationalism and its Crisis”, a chapter in The World Crisis, London, 1938, pp. 348 et seq.
• 12I expect to discuss it at considerable length in my book on the Theory of Banking which is in the course of preparation.
• 13For example England, the United States, etc. Vide infra, chap. ix.
• 14Statements to the contrary which can be found in economic literature rest obviously on a misunderstanding. Thus, for example, Professor Cassel writes (The Theory of Social Economy, New Edition, 1932, vol. ii, p. 458): “On a gold standard there is no sharp division between the quantity of money and the stock of gold.” If Professor Cassel means “an all-gold currency” then this statement is correct; but it is much more likely that the gold standard is really meant: and then this statement is essentially wrong. What is worse is that most of Cassel’s theory on the relations between gold and prices and on the role of gold in the monetary system rests on that assumption expressed in the sentence quoted. One cannot emphasize too often the fact that a gold-standard currency is a paper currency attached to gold (and sometimes convertible into gold); it is not a gold currency. Vide infra, chaps. iii and iv.
• 15It is important to note that the so-called “going-off the gold standard” is not tantamount to the suppression of gold as monetary metal. In the latter case we should have to speak of “demonetization of gold”. Vide infra, p. 177.
• 16Relations between gold and prices are particularly undetermined when the system existing is a de facto system consisting in Central Banks (or Treasuries) buying gold at prices fixed by them, while the regular gold standard is suspended. The price of gold is then subject to variations, but at the price prevailing at every particular moment, monetary authorities are buyers of gold, which as we shall see is one of the features of the gold standard. On the other hand, when the gold standard is suspended there are no longer any provisions about the gold cover of notes, and the value of gold stocks may become indeterminate. All this makes relations between gold stocks and prices very vague indeed, without making them non-existent.
• 17Cf. with the Macmillan Report (Committee on Finance and Industry Report), London, 1931, H.M.S.O., Cmd. 3897, para. 45.
• 18Thus, for example, The Economist wrote the following in its Editorial of 15 May 1937 (p. 404): “The world, in fact, has learned a new lesson. The value of a country’s gold stock does not merely depend upon the physical quantity of gold lying in the vaults of the Central Bank or the Treasury. It depends also upon the price of each ounce of gold expressed in the national currency.” (The italics are mine. I fear that for many it has really been a “new lesson”!)
• 19vide Appendix.
• 20I find myself in substantial agreement with conclusions reached by A. D. Gayer in his book on Monetary Policy and Economic Stabilization, p. 63, though his criticisms are considerably less severe than the ones I have developed in this chapter. Charles O. Hardy says, similarly, in the book Is There Enough Gold?: “Other factors, such as the use of bank deposits as money, and changes in the absorption of gold by India, are brought into Cassel’s analysis to account for the major deviations of the price trend from the ‘relative supply of gold’, (p. 22.) If, as Cassel believes, the increase in the amount of gold stock required to maintain prices during the 60 years before the war corresponded accurately or even approximately to the growth of production, it was because for the period as a whole these other important factors cancelled each other out.” (p. 24.) As a matter of fact these factors did not cancel each other out as there were considerable price movements in the course of these 60 years. Cancellation can only have existed in so far as prices were more stable than they would otherwise have been.
• 21In his study On Quantitative Thinking in Economics, Oxford, 1935, p. 107, Cassel says in this connection the following: “During the period within these limits Great Britain was on the gold standard and, in addition, was essentially a free-trade country. British prices, therefore, may be taken fairly well to represent gold standard prices in the world market, and British index-numbers to reflect the variations of the general price level in terms of gold.” Still it does not follow that prices in Great Britain depend upon the size of world stocks of gold! In the same context Cassel says, rightly, that “the analysis must be confined to a period in which the general system of money has been essentially unaltered”. But in the period 1850–1910 this has not been true even for Great Britain, where the great development of banking took place precisely in the second half of the nineteenth century; the fact of staying on the gold standard does not exclude the possibility of great changes taking place in the internal organization of the monetary system. If the statement that in that period of sixty years “the general system of money has been essentially unaltered” is not true for Great Britain it is much less true for the world at large; thus the condition which Professor Cassel recognizes himself as a necessary prerequisite of his analysis has not been fulfilled in the period to which he applies that analysis.
• 22This, not the “price level”, is the subject that deserves our interest, though it is clear that, thus formulated, the changes cannot be expressed by a single numerical time-series. How to represent them adequately is a problem for statisticians to solve!
• 23A change of the price of gold may, of course, affect quantities produced, besides affecting their value and the value of existing stocks.
• 24Gustav Cassel, The Downfall of the Gold Standard, Oxford, 1936, p. 7.
• 25Neither has Cassel done anything to demonstrate this sweeping statement.
• 26Ibid., pp. 3–4.
• 27Which is merely a statistical construction to be accepted for what it is worth. Vide Appendix.
• 28This is, for example, the position taken by Professor Rist. Vide infra.
• 29And, of course, the policy of commercial banks, which is more or less closely connected with that of the central bank.
• 30In The Theory of Social Economy (1932 edition, vol. ii, p. 494), he says that “there is no obvious reason why an increase of the gold supplies should cause a permanent alteration in the average proportion between means of payment and bank reserves. . . . We shall therefore find it quite natural that the gold supply completely controls the secular variations of the curve of the general price level.” Now the fact is that this proportion did change, and did change very considerably if we include, in the total of existing means of payment, demand deposits transferable by cheque.
• 31The Theory of Social Economy, 1932 edition, vol. ii, p. 495.
• 32In Cassel’s opinion “it is sufficient for upholding the quantity theory, to assume that the total demand for gold is proportional to the general price level” (The Theory of Social Economy, 1932, vol. ii, p. 459); this view is shared by some economists but I think that it is inaccurate and may lead to regrettable misunderstandings. There is no mention of gold as such in the “quantity theory”; it relates the total volume of monetary circulation to the “general price level”, not the “demand for gold”. Gold can be brought into the picture only indirectly, because of institutional arrangements, and never enters into the “quantity theory” as one of the variables.
• 33The Theory of Social Economy, vol. ii, p. 496.
• 34Ibid., pp. 496–7.
• 35Vide supra, chap. ii.
• 36The Theory of Social Economy, vol. ii, p. 497.
• 37See Wicksell’s Interest and Prices (first published in German in 1898; English translation published in 1936).
• 38Vide infra, chap. vii.
• 39See the following publications by Professor Rist:
• 40Rist 1938, pp. 309–10.
• 41Ibid., p. 308.
• 42A limitation which is hardly valid for England and the United States.
• 43Rist 1938, p. 338 (italics are mine).
• 44Vide supra, p. 30.
• 45See, particularly, Rist 1935.
• 46Rist 1935, p. 235.
• 47Ibid.
• 48Rist 1935, p. 236.
• 49Ibid.
• 50Ibid., p. 237.
• 51Rist 1935, p. 237 (the italics are in the text).
• 52I do not repeat again all the objections to price averages and index-numbers of the “general price level”. Vide Appendix for detailed comments on that matter.
• 53For the distinction between gold prices and prices that exist under the gold standard, vide supra, chap. ii.
• 54I must leave aside the question of whether such “eliminations” are at all significant. The critical and sceptical observations which the reader will find in the Appendix apply to this problem as well. I cannot discuss further here the question whether the application of a “linear trend” is at all justified.
• 55“. . . des adversaires de la théorie quantitative qui sont en réalité les adversaires de toute théorie accordant à l’accroissement des métaux précieux une action sur les prix”, Rist 1935, p. 233.
• 56Vide supra, p. 44.
• 57Rist 1935, pp. 237 and 238.
• 58Rist 1938, p. 282. (Italics are mine.) The italicized phrase reads in French as follows: “le monde est revenu graduellement à un niveau de prix conforme à la nature des choses”.
• 59Rist 1935, p. 234. (Italics are in the text.)
• 60This distinction is rather confusing because there is no strict delimitation possible between “world gold prices” and the other “gold prices”. I suggest that the definitions given in chap. ii above are preferable, since they are clear cut and do not give rise to either “illusions” or misunderstandings!
• 61Rist 1938, p. 282.
• 62This sweeping view appears to me unacceptable. Vide infra, p. 71. Cf. also my study of Monetary Internationalism and its Crisis, published in The World Crisis, London, 1938, pp. 350–1.
• 63Rist 1935, p. 238.
• 64Carl Snyder, Overproduction and Business Cycles, in the Proceedings of the Academy of Political Science, vol. xiv, No. 3, New York, June 1931.
• 65Rist 1935, p. 238. Snyder’s “Index” is one of the statistical constructions the basis of which is discussed in the Appendix and found open to serious objections. It should not be used at all; it is better not to have any index of production than to be misled by one that is void of any economic significance!
• 66Vide supra, p. 51.
• 67Cf. with the analysis given in chap. ii above.
• 68Rist 1935, p. 241.
• 69Cf. with sect. F of chap. viii, pp. 170 4; see also pp. 201–2.
• 70Vide supra, pp. 136 and 140.
• 71The more usual but less instructive formulation would be as follows: “the national monetary unit of each particular country is defined by law in terms of that standard.”
• 72Vide supra, chap. vii.
• 73The usual, incorrect formula: “gold contents of the monetary unit”, represents an anachronism carried over from the era of “hard money”.
• 74Vide supra, chap. vii.
• 75This is not to be confused with the mechanism of “exchange standards”; it is true that there also gold movements are replaced by movements of balances, but of balances of the currency used as standard and exclusively of that currency.
• 76Cf. with chap. x, sect. B.
• 77As defined above, pp. 144–7.
• 78Cf. with F. A. von Hayek, Monetary Nationalism and International Stability, chap. iii.