Milton Gilbert*
This essay was written in the summer of 1967; its purpose was to explain the nature of the international monetary system and how it functioned up to mid-1967 as a background to the consideration of various possible improvements in the system. Hence, it deals with the system as it has been—not as it might become. While I have made drafting changes and clarifications, I have deliberately not extended the paper to cover the events of the past year so as to avoid discussion of matters about which there are differences in official views. My objective is to analyse the system and not to enter into the political problems of its future evolution.
More specifically, the essay aims to distinguish between difficulties arising from inadequate adjustment policies of individual countries and difficulties arising from a disequilibrium1 in the system as a whole, which concerns the relationship between gold and the dollar. The analysis is focused on the persistent deficit in the balance of payments of the United States and is designed to bring out its underlying and transient causes.
The present system is usually called the gold exchange standard. As it emerged from Bretton Woods and as it has functioned in the postwar period, however, it is more to the point to call it the gold-dollar system.
It may seem curious that economics textbooks do not provide an analytical model of the system. Indeed, there is not really a formal theory of the gold exchange standard, comparable to the theory of the gold standard. This may be, partly, because the system has not been static but has been developing under the changing conditions of the last two decades. However, it is also because the system does not lend itself easily to presentation by a simplified model, as does the gold standard, since central banks do not constitute a homogeneous universe and do not act according to a set pattern of economic considerations.
The absence of an accepted theory of the gold-dollar system has made for much confusion in public discussion. Economists and officials do not start with a generally agreed conception of how the system works or ought to work, such as they have, say, in dealing with problems of demand management. They do not have a model for the equilibrium of the system or a common view on the respective roles of gold and the dollar. In these circumstances, very diverse, and often contradictory, proposals have been offered to solve the problems of the system and the deficit of the United States. In the main, these proposals are either unconvincing as prescriptions for establishing equilibrium in the framework of the present system, or they involve changes so fundamental as to constitute a new system. It is hardly surprising that the political authorities have not been able to find their way out of the maze.
The system rests on a series of basic principles and behaviour characteristics which determine its mode of operation. These derive from law, from international agreements, from the policy aims of central banks and governments and, in some respects, from technical necessity. It is, of course, an evolving organism, which was different thirty years ago and which will, no doubt, be different thirty years hence. The concern here, to repeat, is with the system as it has existed during the past two decades.
Fixed exchange rates
1 Fundamental to the system is the aim of monetary authorities to adhere to fixed rates of exchange. Maintenance of the rate has a high priority with all countries and other objectives are often sacrificed to it. It is apparent that fixed rates have overwhelming support from the business and financial community also.
2 To say that we have a fixed-rate system raises the question of what the rates are fixed to. Under the IMF Articles, a country may declare its par value in terms either of gold or of the dollar of the gold weight and fineness in effect on 1 July, 1944. There is only a minor technical difference between these two standards.
However, the operative standard for most countries is the dollar as such, and central banks in practice intervene in the market when necessary by buying or selling dollars against their own currencies to keep the dollar exchange rate within agreed limits. The cross rates with other currencies are kept in line by market arbitrage. There are exceptions, of course, such as the countries of the sterling area, which peg their currencies to sterling and rely on the Bank of England to maintain the fixed rate between sterling and the dollar. But, generally, central banks operate directly on the market for dollars vis-à-vis their own currency, and it is the market rate on the dollar that is significant for their international competitive position— irrespective of the legal gold content of the dollar.
3 The exception in the system is the dollar itself, which both in law and in fact is fixed in terms of gold— at $35 an ounce. The United States is not obliged to intervene in the exchange market; it has only to be prepared to buy and sell gold at $35 and can leave it to the intervention of other central banks to maintain fixed rates to the dollar. The United States has intervened in the market at times in recent years, both spot and forward, but the purpose was to avoid losses of gold from temporary movements of funds rather than to keep rates in line.
Reserves
1 To maintain fixed rates the monetary authorities must hold reserves so that they are in a position to iron out fluctuations in supply and demand in the foreign-exchange market. Reserves consist of liquid international assets, readily available for intervening in the market, and are almost entirely confined to gold and to foreign-exchange assets in dollars and sterling.
However, while sterling is important to the international economy as a trading currency, it is active as a reserve currency only in settlements between the United Kingdom and sterling-area countries. Hence, it is a regional reserve currency and quite different from the dollar, which is the reserve currency of the system. To simplify matters I will discuss only dollars, thus treating the sterling area (and, likewise, any other monetary area) as a unit which holds reserves in gold and dollars.
It will be seen, therefore, that the system is a gold-dollar system for two reasons: first, currencies are fixed either to the dollar or to gold and, secondly, the reserves of the system are gold and dollars.
2 Each central bank is free to determine the composition of its reserves as between gold and dollars. Its policy in this respect is in its own hands, because it can sell or buy gold against dollars at the US Treasury. If there is any constraint on such exchanges, even psychological or political, the convertibility of the dollar at its fixed relationship to gold comes into question. In practice, there is a wide range among the countries in the ratio between holdings of gold and dollars, indicating that central banks give different weight to the benefits to themselves of the two categories of assets.
3 Dollars are held almost entirely in money-market instruments and time deposits, as these are liquid assets that earn interest. Gold, on the other hand, produces no revenue. It is important to realize that, if central banks could not earn interest on dollars, their reserves would be almost entirely in gold. Some central bankers have stressed that they are not primarily concerned with interest earnings in determining the composition of their reserves. This may be true of marginal changes in reserves; but the point is that, if the United States did not permit central banks to invest dollars at interest, they would never have acquired the dollars in the first place; they would have acquired gold.
Hence, the first requirement for a currency to become a reserve currency is that there must be an open money market in which foreign central banks can freely invest in short-term paper. In addition, the money market must be capable of absorbing large central-bank transactions, and the convertibility of the currency at a fixed rate must be rather secure. It is because New York and London are the only two open money markets of any size that the dollar and sterling are the only two significant reserve currencies. And it is because exchange rates between sterling and other currencies have not been secure that the dollar, supported by large gold reserves, supplanted sterling as the reserve currency of the system.
Other currencies have not become reserve currencies either because the central bank discourages placements of funds at interest by foreign central banks or because their convertibility at a fixed rate does not seem reasonably assured over the longer run. Continental European countries have not wanted to become reserve centres; they are reluctant to have their markets and reserves disturbed by large-scale operations of foreign central banks and some, also, see no point in their country bearing the interest burden attached to having their currency held as reserves.
It has been said that dollars are kept in reserves primarily as a matter of convenience, since dollars can be used directly in the exchange market whereas gold must first be converted into dollars for the purpose of market intervention. However, dollars held in money-market paper or on fixed-term deposit must equally be converted into cash to be available for market intervention, and there is no great difficulty in converting gold into cash.
While a variety of developments went to make the dollar the reserve currency of the system, the United States took no initiative in the matter; its action was only permissive.
4 With respect to its reserves, also, the United States is an exception. While other countries are free to hold their reserves in any combination of gold and dollars they wish, including 100 per cent in dollars, the United States must hold its reserves essentially in gold. This is because there is no other currency besides the dollar that can be used for general intervention in the exchange market; hence, any foreign currencies held by the United States cannot be used for general support of the dollar in the way that other countries use the dollar as a general support for their currencies. The United States can generally use foreign-exchange holdings only for bilateral settlements. To underline the importance of this point, France could hold all its reserves in dollars if it were so minded, but the United States could only hold a quite small fraction of its reserves in French francs.
The foreign-exchange assets that have appeared in the reserve statistics of the United States in recent years were always acquired for specific purposes. For example, there may be a temporary holding of DMarks which were acquired in the market in anticipation of repaying D-Mark Roosa bonds to the Bundesbank. Or, there may be small holdings of Swiss francs to be fed into the market when the dollar is under pressure so that the Swiss National Bank will not have to acquire the excess of dollars which it might then want to convert into gold.
The only currency that the United States has held in large amounts has been sterling. These holdings arose mainly because American assistance to the Bank of England was given in the form of swaps of dollars against sterling—rather than as simple advances. The sterling, of course, could not be used at the same time by the United States to meet its own deficit or to avoid gold losses and, therefore, was not ‘reserves’ in the ordinary sense of immediately marketable assets. To count such sterling assets in reserves is about as appropriate as it would be, say, for a business firm to include its accounts receivable in its cash.
5 Since the United States is the only country obliged to hold its reserves in gold, the function of gold as a ‘discipline’ against excessive money creation is primarily applicable to the United States. Other countries are subject to balance-of-payemnts discipline, but the discipline lies in the loss of any reserves—whether dollars or gold. Even so, a loss of gold makes a much greater impression on public opinion in many countries than a loss of foreign-exchange reserves or foreign borrowing by the central bank. If other countries entirely stopped acquiring gold, the discipline of gold on the United States would become rather theoretical. This is particularly so because increases in its liabilities to foreign official institutions seem to have exerted little discipline on the United States.
6 Why do central banks, apart from the United States, hold non-interestbearing gold at all and what determines the proportion between their holdings of gold and dollars? Several considerations are involved, to which the various countries attach different importance.
Given this variety of motives, it is apparent that the comparative benefits of holding gold relative to dollars cannot be calculated. In other words, central banks cannot know what reserve policy will make their country better off—and, perhaps, they cannot even define precisely what being ‘better off is. What many do, therefore, is work to some rule of thumb. A few years ago, for example, there were several central banks that aimed to have about a 50–50 ratio between gold and dollars, whereas others held mostly gold and still others mostly dollars. Reserve ratios generally are not set once for all, however, but are subject to change according to circumstances—particularly to changes in the degree of certainty regarding the gold convertibility of the dollar at its fixed price.
7 Besides reserves, IMF facilities are available in the system to assist countries that encounter balance-ofpayments difficulties. The amount any country may draw is originally fixed by its quota, which broadly reflects its size and economic strength. In establishing its quota, each country as a rule pays 25 per cent to the IMF in gold and 75 per cent in its own currency and agrees that its currency may be drawn upon in case of need to finance other countries’ drawings. The Fund may also finance drawings partly by selling gold in order to acquire the needed currencies.
The right of a country to draw on its gold subscription is practically automatic; so also is its right to draw on any credit balance it may have built up by having had its own currency drawn upon. These two amounts have come to be called a ‘reserve position in the Fund’. If a country draws on its quota above its reserve position in the Fund, it is taking credit and its right to this credit is conditional upon the IMF judging that its policies are likely to correct its external deficit.
From the standpoint of meeting a deficit, therefore, a Fund reserve position is equivalent to a country’s own reserves. But it differs from reserves in three respects:
There has been one instance of a country making a deposit with the IMF; the consideration involved was that the deposit was covered by a full gold guarantee, in contrast to the normal gold-value guarantee incorporated in the IMF Articles. Hence, it is not an arrangement with large possibilities of expansion, because the IMF could not assume the risk on the price of gold.
The credit tranches of a country’s quota are not comparable to reserves; they are conditional facilities and any credit obtained carries definite repayment obligations and interest charges. Some high officials have had the hope that drawings on the IMF would become fairly routine central-bank operations—like the use of bank credit by a business firm to supplement its working capital. Thus far, however, this idea has not been realized; drawings on the IMF have been indicative of a strained or crisis situation in which the IMF is called upon as a rescue organization. In fact, countries have at times emphasized drawing on the IMF so as to gain public support for necessary corrective policy actions.
Since the IMF was established in 1944, there have been two general increases in countries’ quotas, by 50 per cent in 1959 and by 25 per cent in 1966, as well as special increases for particular countries. Also, the General Arrangements to Borrow was agreed to by the Group of Ten industrial countries in 1962, whereby they could lend additional resources to the Fund to help meet large drawings by members of the Group. The need for this arrangement arose because the Fund’s stock of convertible currencies could be inadequate to meet large drawings under conditions when a balance-of-payments deficit of the United States limited the Fund’s use of its dollar holdings or when the United States itself wanted to make a large drawing on the Fund. In either of these cases, therefore, the Fund could have a problem of liquidity; in fact, in the Bretton Woods’ arrangements it was probably not contemplated that the Fund’s dollar holdings might not be freely usable because of a large deterioration in the reserve position of the United States.
It should be noted that transactions under the GAB are covered by a full gold guarantee.
8 In addition to the IMF, short-term central-bank credit facilities have been arranged among a number of countries. These may be used on an ad hoc basis and are designed essentially to help meet reversible movements of private funds and to relieve the pressure on the reserves temporarily while the character of the demand for foreign exchange is being appraised. Such assistance is provided on the credit standing of the central bank, in which the size of the reserves is an essential consideration. They are not conditional in the sense of IMF credit facilities, since the borrowing central bank cannot make commitments about the adjustment policies of its government.
9 Besides reserves and official credit facilities, extensive use is made of foreign commercial-bank credit and other private liquid funds to meet strains on the exchange market. Central banks may do this on their own account, or they may arrange matters so that it is done by their own commercial banks. The scope for such operations has been much enlarged by the development of the Euro-dollar market and the market has in recent years been drawn upon by several countries for quite large amounts. Private credit facilities are certainly a flexible supplement to official resources and are likely to be of growing importance. It would be going too far, however, to consider them a substitute for monetary reserves, especially since a country with inadequate reserves is not likely to have a high credit rating with private banks.
The adjustment process: countries in deficit
1 A country with a balance-of-payments deficit can for a time hold its exchange rate by drawing on reserves and available borrowing facilities. As these are limited, however, and as drawing on them too much may make matters worse by leading to a flight from the currency, the authorities must sooner or later take action to get out of deficit. When this adjustment is not brought about, and the exchange parity depreciates, economic and financial policy are considered by the general public to have failed. Whether maintaining the rate is a reasonable objective in given circumstances, however, depends on whether the authorities can take sufficient policy action to eliminate the deficit.
2 The policy actions available to eliminate a deficit and some limitations on them are, briefly, as follows:
Hence, in adhering to the principle of no direct capital controls until a few years ago, the United States was almost alone among the convertible-currency countries. Because of its high per capita income and the huge volume of savings generated by its economy, the United States would have been the dominant capital market in the world in any case. But this position was reinforced by the controls and policies maintained in other countries.
In the last few years the United States has imposed direct capital controls to limit its gold losses. Most other industrial countries find this course perfectly natural and desirable. Indeed, some seem to believe that the deficit of the United States, apart from the effects of the Vietnam War, could be cured by stringent enough capital controls. This view, to my mind, does not take sufficient account of all the links there are between the capital and current accounts, or of the shifts that take place between the various categories of capital outflow and inflow when controls are applied.
3 The force of these instruments can be very substantial when they are used vigorously and there have been many instances in the postwar period of countries emerging successfully from a period of deficit by means of them—without undergoing deep recession or prolonged stagnation. However, cases can and do arise in which they are unable to restore external balance—usually because domestic inflation has brought internal prices and costs too far out of line. Hence, the aim of maintaining a fixed rate cannot be considered absolute.
A deficit position which requires a change in the exchange parity to bring about correction is called a ‘fundamental disequilibrium’ and it is provided in the IMF Articles that a country in such a situation may change its rate without sanctions. There is no legal definition of fundamental disequilibrium, but in practice countries do not apply to the IMF for a change in rate before the situation is perfectly obvious; they have always obtained approval. A country that resorts to extensive exchange restrictions in such a situation instead of adjusting the rate is not supposed to be eligible for IMF assistance, though, if the truth be told, some countries have got away with murder. It is far from pleasant for the IMF to insist that a country devalue as a condition to drawing IMF credit.
While the evidence of fundamental disequilibrium in some cases is unmistakable, the distinction between transitory and basic imbalance is difficult to make in others. There is no computer program by which the precise equilibrium rate of exchange can be determined, and, even if there were, no country would change its rate to correct a small disadvantage in the structure of exchange rates. There are several reasons which justify this attitude. First, to depreciate the rate by, say, 3 per cent or 5 per cent would be likely to do more harm than good, because of the distrust in the currency that it would engender. Secondly, the policy instruments available for maintaining external balance are sufficient to prevent prolonged reserve losses in such cases, without undue sacrifice of other objectives, for example, by rather small changes in the capital account. And thirdly, there is an adjustment process constantly at work which tends to correct small imbalances, particularly when it is helped along by appropriate demand policy and when it is not negated by continual wage inflation. This adjustment process takes place both within the given country and in the world econ omy on the outside; its reality is evident from the fact that reasonably well-managed countries are able to maintain fixed rates over long time spans.
Thus, the existence of fundamental disequilibrium is a matter of degree and to specify it in any given case is a matter of judgement. Such a judgement is particularly hazardous when external imbalance is accompanied by excess domestic demand and when there is likely to be some flight of capital contributing to the imbalance. For example, there were observers who considered that the lira had become overvalued in 1963, but this was proven to be a misjudgement as soon as the domestic inflation was brought under control. On the other hand, all competent analysts considered the French franc to be in fundamental disequilibrium in 1957—and they were right.
4 For the generality of countries in deficit, the availability of a change in rate, which improves the competitive position of exports relative to imports, means that a balance in external payments can always be restored. In fact, it always is restored. When a country delays action until it runs out of reserves and runs out of credit, it must in the end devalue. It may hide this fact from itself by tying its economy into knots with extreme exchange restrictions and multiple exchange rates and by turning its eyes from the black market which always springs into life in such circumstances. But, then, the currency has effectively been devalued de facto, if not de jure.
Consequently, there is nothing wrong with the adjustment process when it is viewed as including a change in exchange rate as the ultimate policy instrument. We have seen it work perfectly adequately in case after case. Where the external deficit was due merely to excess domestic demand, as in the Netherlands in 1956–7, Italy in 1963–4, or Germany in 1965–6, the deficit disappeared when effective monetary and fiscal measures were taken to restrict internal demand. And where such action would not do, because there was a fundamental disequilibrium, as in France in 1956–7, or Spain in 1957–9, the deficit disappeared when appropriate devaluation was undertaken in combination with restricting excess demand, which was the cause of the imbalance in the first place.
It would be far more satisfying, of course, if the monetary and economic behaviour of countries were always such that they avoided falling into fundamental disequilibrium. But if they do not, it is no reflection on the system. And if they choose to suffer the distortions and stagnation of an overvalued currency instead of adjusting to an equilibrium exchange rate, it is on their own responsibility as sovereign nations.
5 Here again, however, the United States is a significant exception because as a practical matter it cannot act directly on exchange rates. This follows to some degree from the fact that the dollar is fixed to gold, rather than to any particular currency. But it is a consequence even more of the weight of the United States in the world economy and the significance of the dollar in the international monetary system. Suppose the United States decided that its balance-of-payments deficit could not be corrected by acceptable adjustment policies, and that it had gone the limit in using its gold reserves and taking IMF and central-bank assistance. It could then either raise the dollar price at which the Treasury buys and sells gold or simply suspend gold sales by the Treasury without fixing a new price for the time being. Whether any changes would then be made in exchange rates vis-à-vis the dollar would depend upon the reaction of other countries. In the first case they could maintain their fixed parties to the dollar, with the result that the price of gold would be higher in all currencies. In the second case, also, they could intervene in the exchange market to maintain the peg to the dollar and let the price of gold in their own currencies be free to move with market forces.
I leave until later the question of what they might do under various conditions and here wish only to stress two points: the first is that the process available to the United States for removing a persistent deficit is different than for other countries; the second is that the equilibrium of the dollar involves the equilibrium of the whole system in a way that is different than for other currencies and is necessarily related to the price of gold. The difference between the dollar and other currencies in this regard may seem to be a difference of degree, but it is so large as to constitute a difference in kind.
6 This position of the dollar is what lies behind the official insistence on improvement of the adjustment process. There is not great concern about the adjustment process in general, because other countries cannot avoid adjustment. And even if they have to adjust by means of a change in exchange rate, it is largely a local affair which does not involve the system as a whole. The key target of the demand for better adjustment is the persistent deficit of the United States because it is likely to involve the stability of the system as a whole. There has been a strong feeling that somehow its deficit has reflected misbehaviour on the part of the United States—even when the United States was clearly not having excess demand, when the margin of unemployed resources was unnecessarily large, and when it could not be convincingly shown by what combination of policy measures the United States could meet the demand to eliminate its deficit. However, there have been very few in official circles bold enough to draw the apparently logical conclusion that the dollar was in fundamental disequilibrium; very few have felt that their exchange rate vis-à-vis the dollar ought to have been revalued. For its part, the United States took refuge in the idea that the trouble was with lack of adjustment by the surplus countries—and the charges back and forth left matters more or less at a standstill.
For the past several years, also, criticism of the adjustment process has been directed at the United Kingdom. However, the United Kingdom was having substantial excess demand, domestic inflation, and overfull employment. And at the same time it was asking for very large assistance from abroad to finance its external deficit. Hence, the grounds for complaint were quite different than in the case of the United States before the start of the Vietnam inflation.
7 A final point with regard to deficits. Given the nature of the policy instruments available for correcting a significant deficit position, it will be apparent that the process of adjustment is necessarily a relatively short-term affair. When it does not take place fairly quickly, it simply means that the authorities have not taken the appropriate measures—either deflation and capital controls, if the imbalance is not fundamental, or devaluation, if there is fundamental disequilibrium. And when the exchange rate is significantly overvalued, there is no way to adjust other than by changing the exchange rate.
Governments are often reluctant to accept this proposition because of the stigma usually connected with a change in the exchange rate; so they think up all sorts of pseudo-measures for the long-run correction of the deficit. In recent times, however, there is not a single successful case of long-run adjustment of a sizeable balance-of-payments deficit—apart from the special case of reconstruction of war damage to the productive potential of the economy. And even those cases did not take very long. In former times, when stagnation of the economy led to declining wages and prices, such adjustments often occurred. In our day of downward rigidity of wages and prices and of the high priority given to full employment, however, such an adjustment can take place only through wages in the deficit country rising less than in the outside world—and the margin of correction that has been possible by this process has proven relatively small.
The United States, in particular, has had a long-term programme to restore balance for seven years and yet the goal is as elusive as ever. Failure to face up to this reflects political attitudes—not economic analysis.
The adjustment process: countries in surplus
1 It is often said, from the standpoint of the system as a whole, that both surplus and deficit countries must share the responsibility for achieving balance in international payments. However, the primary responsibility, and the active role in the adjustment process, falls in fact on deficit countries because it is their exchange rates that are in jeopardy. When a country is in surplus, the central bank can feel free to concentrate on domestic objectives of full employment and growth. But when the country is in deficit, it cannot. Thus, there is a natural bias toward being in surplus, since the surest way to avoid any risks to the exchange rate is to stay on the right side of the line. When a country is in moderate surplus, therefore, it will not take deliberate action to reduce the surplus and, even when the surplus is fairly large, deliberate corrective action is rather limited. The cooperative actions taken by surplus countries have been confined largely to facilitating the financing problem—such as prepayment of long-term debt, provision of special facilities to the banks to acquire foreign-exchange assets, and accepting special exchange-guaranteed assets instead of gold. Several countries also have used special techniques to limit the inflow of funds from abroad.
2 Besides the general aim of protecting the exchange rate, several more specific factors militate against an active adjustment policy by surplus countries:
The mercantile flavour of official attitudes toward the balance of payments comes from recognition that a surplus approximates equilibrium in a growing world better than does a constant level of reserves. And, of course, economists have stressed the need for global reserves to rise. The Netherlands central bank formerly set a quantitive target for the growth of its foreign reserves and tried to arrange domestic liquidity creation so that the target would be met. Other central banks are influenced by the same idea, if in a less precise fashion.
3 While a reasonable, or even moderately large, surplus may be in the vicinity of equilibrium, cases arise of persistent extreme surplus. This is a situation in which, after total demand has been pushed to a fullemployment level, the surplus does not fall to reasonable proportions. In the conventions of international cooperation, there is no obligation on surplus countries to pursue inflationary demand policies in order to bring down the surplus.
The only provision in the Fund Articles for such cases is the ‘scarce currency’ clause. It has never been called upon and, indeed, was expected only to be applicable to the United States. A country declared to have a ‘scarce currency’ would not have to do anything itself, but other countries would have a right to discriminate against it in their trade and payments regulations.
To be parallel with devaluation in cases of fundamental disequilibrium, countries in exreme surplus ought to revalue their currencies. But the high priority on fixed rates holds for currency appreciation as well as devaluation, and revaluation is a rare occurrence. The only recent case was the revaluation in 1961 of the German mark and the Dutch florin by 5 per cent.
However, revaluation must be recognized as the ultimate policy weapon available to countries that want to stop the inflationary consequences of an extreme surplus. The surplus is an inflationary force because, in pegging the rate, the central bank has to buy the excess of dollars offered in the market against domestic currency and there are practical limits to the extent to which the authorities can neutralize this increase of domestic currency by other policy actions. The revaluation of the Deutsche Mark was undertaken precisely on these grounds.
Two other cases show that the authorities do have the power to act against an extreme surplus when they feel that it constitutes an intolerable danger to internal monetary stability. Canada adopted a floating rate in 1950 to combat a huge inflow of investment funds from the United States; the Canadian dollar appreciated between 5 and 8 per cent and a balance was achieved in the market with a less inflationary inflow of funds from abroad. Switzerland, for some years after the war, also allowed its rate to float for most transactions other than trade and a part of tourism, when faced with an unmanageable inflow of funds from abroad. The Swiss franc on the free market appreciated as much as 30 per cent to the dollar, against the par of 4.37. It should be noted that both Canada and Switzerland could have converted their dollar inflows into gold in those days without any reproach from Washington. But it was the large surplus they did not want even in gold.
Although rare, these instances of revaluation are significant from two standpoints. First, they show that the United States cannot foist any amount of dollars on the rest of the world; beyond a certain point other countries would sever their fixed ties to the dollar. The fact that they have not done so indicates that they have not considered their currencies to be undervalued. Secondly, the system provides countries in extreme surplus with a remedy, if they care to use it. If they do not, it is on their own responsibility. The saving grace from the standpoint of other countries is that an extreme surplus tends to be eroded by internal inflation.
4 By comparison with any previous time, there has been a high degree of international monetary cooperation in these postwar years. It must be recognized, however, that every country gives priority to its own basic interests and that the demands on co-operation cannot violate those interests. As the world is made up of sovereign states, ultimate responsibility both for the convertibility of each currency and for the control of inflation within each country is a national responsibility.[…]
Since the eruption of the market price of gold in 1960, there has been a steady deterioration in the operation of the system and a change in its character. I summarize the main aspects of this deterioration from the review of developments already recounted.
1 The standing of the dollar as the reserve currency of the system has become compromised as there is less readiness to hold dollars freely. To minimize conversion of dollars to gold under these circumstances, the United States has resorted to giving guarantees on various of its external liabilities.
2 The United States has used moral suasion to prevent dollars being converted to gold. It is no secret that such conversions are considered to be at least uncooperative, and in some cases unfriendly. Some foreign central banks have refrained from demanding gold for dollars so as not to rock the boat. Hence, central banks no longer have full freedom over the composition of their reserves; nor is it quite right to say that the dollar is still freely convertible de facto.
3 After the rise of the market price of gold in 1960, the principal central banks formed the gold pool in order to keep control over the price. At the start, the assumption of the pool was that there would normally be an excess of market supply over demand—which may include buying by central banks that are not members of the pool. By 1967 the pool had to supply not only the deficit of gold for private demand but the market demand of nonmember central banks as well. The residual supplier, of course, was the United States, since the other pool members could offset their gold losses to the pool by purchases from the United States.
4 The threat overhanging gold has restricted its use in official settlements; except in desperate circumstances or for political ends, central banks try to meet temporary difficulties by other means.
5 While gold losses act as a discipline on the United States, they have become an uncertain guide for judging its balance-of-payments performance. At one moment the authorities expressed a firm intention to balance the external accounts. When they realized that this unilateral undertaking was impossible, they said that the surplus countries must carry a fair share of the burden of adjustment—without specifying what a fair share for the United States would be. The latest posture seems to be a resigned attitude towards the balance-of-payments deficit, with its persistence and size being attributed to the war in Vietnam. One can only conclude that the authorities of the United States have not formulated a set of standards for judging whether its responsibilities for the reserve currency of the system are being fulfilled. In fact, of course, with the shortage of new gold anything like as severe as it is at present, it cannot be done.
6 With the growing tightness of the system, it has become a matter of high priority to prevent any excitement on the exchange markets and to resort to extreme means to gain market confidence. One aspect of this is a fear of changes in exchange rates and a belief that almost any change in rates constitutes a threat to the stability of the system. This is in the face of the necessity for rates to be much more finely adjusted in conditions of a gold shortage than would be needed with an adequate flow of gold.
7 Owing to limitations on the growth of reserves through gold and dollars, the system no longer has a built-in mechanism for the increase in reserves. As a consequence, the growth of reserves has depended to a large extent on special credits negotiated to finance deficits. Such arrangements are often influenced by political considerations, to the general detriment of strictly monetary and financial standards in the system. And, as the repayment of such credits would require a substantial contraction of global reserves, it is not easy to visualize their orderly liquidation.
The changes in the system that have occurred since 1960 are often presented as an evolution and strengthening of the system. While there have been innovations of permanent value, the essence of the matter has been a series of shoring-up operations to accommodate to a basic disequilibrium of the system. Far from the system being strengthened, it has been disintegrating. This can hardly be considered an evolution of the gold-dollar system, since it consists of replacing both gold and dollars with quite different instruments for the growth of reserves. The past six years have been transitional, and it is evident that the pattern of gold and credit financing followed in those years cannot be repeated in the next six years. If ways are found in the years ahead to suppress the official demand for gold, it will mean that a basically new system has come into being.
1 My use of the terms ‘equilibrium’ and ‘disequilibrium’ to characterize concrete situations in the real world seems to unbalance the editorial equanimity of Fritz Machlup, who has long and resolutely maintained that these terms should be used only with reference to theoretical models with all variables fully specified. For want of more suitable terms, I shall continue to use the proscribed ones in their real-world meanings.
* From Essays in International Finance, no. 70, Princeton, Princeton University Press, 1968, pp. 1–20, 46–7, abridged.