CHAPTER 6 The Heyday of Bretton Woods and the Reserve Debate
- Harold James
- Published Date:
- June 1996
At least in retrospect, a warm glow often overcomes policymakers as well as academic analysts looking back on the “classical Bretton Woods system” as it operated until 1971. “The golden age of capitalism,” “the most successful international monetary regime the world has ever seen,” “the best overall macro performance of any regime,” there are many such encomia,1 In 1977, the distinguished economist Simon Kuznets gave the following verdict on the whole history of economic growth: over the previous 25 years, in both the developed and the less-developed worlds, “material returns have grown, per capita, at a rate higher than that ever observed in the past.”2 Inflation rates were lower than in the following period of floating exchange rates, real long-term interest rates were lower, and growth higher. Above all, growth was more evenly distributed. Internationally, differences between regional performances were limited: the real annual growth rates of GDP in the 1960s stood at 5 percent in member countries of the Organization for Economic Cooperation and Development (OECD), 4 percent in Latin America, and 6 percent in Asia. Domestically, too, most countries moved toward greatly reduced inequalities of income distribution. Both these experiences present a marked contrast with the decades following.
In the era of boundless opportunities, of President Kennedy’s “New Frontier,” or of British Prime Minister Harold Macmillan’s “you’ve never had it so good,” no limits seemed to exist on the capacity of societies to organize themselves to produce growth. The proponents of “growthmanship” believed that the task of the international financial system lay in supporting growth. They feared that a shortage of reserves might constrain or curtail economic advance through the need to deal with balance of payments problems, and believed that counteraction was urgently required. The major function of reserves no longer lay in providing a backing for the issue of domestic currency, as it had under the gold and gold exchange standards, but rather in supplying a “buffer” that could allow sudden swings in the trade balance to be accommodated without requiring the imposition of either trade protection or monetary deflation. Such swings were bound to occur and could only be accommodated given an adequate supply of reserves. (The analogy is sometimes made with a taxi stand. There needs to be a line of taxis in order to accommodate the demands of a sudden flow of passengers without causing them delays; in normal times, taxis return and fill up the stand.) This chapter traces the evolution of the reserve discussion and the attempt to produce a workable reform that would command an international consensus. The latter requirement, however, would ensure that the new instrument, the SDR or special drawing right, in practice never came to play the central role in the international system that its advocates had envisaged.
Trade and Shocks
Perhaps the most striking feature of the classical Bretton Woods system in retrospect appears as the expansion of international trade at a rate unparalleled either before or since. This trade performance was correlated, at both the international and the national levels, with a very vigorous growth in output (see Figures 1-1 and 1-2).
One of the forces that propelled this development was the reduction in tariff levels for manufactured goods and the substantial elimination of quotas between industrial countries as the result of rounds of multilateral trade negotiations under the General Agreement on Tariffs and Trade (GATT). The most impressive reductions occurred at the beginning, at the Geneva Round of 1947, and the Annecy Round of 1949, when the United States negotiated bilaterally with 20 other governments.3 In 1951 the United Kingdom and Germany and in 1955 Japan joined the GATT, with the result that all major developed countries were now members. In the 1960s, the Kennedy Round extended the negotiating process. Instead of bilateral agreements, multilateral arrangements were made; and instead of productspecific concessions, liberalization was conducted across the board.
Trade growth and trade liberalization reinforced each other. Growth convinced policymakers of the concrete benefits to be obtained through liberalization, and encouraged them to take further steps, which in turn generated larger markets and further growth. Later, the falling off of growth rates in the 1970s helped to provoke a reversal of the progress made in trade talks, and the results then, conversely, further harmed the prospects for trade. Since liberalization gives rise to a virtuous cycle, and protectionism correspondingly produces a vicious cycle, politicians, policymakers, and economists felt more and more sensitive to the possibility of a shock that might reverse the beneficent process and send the world on the disastrous downward spiral. The longer the successes of the golden age lasted, the more nervous the participants became. Could the eventual shock come from the financial system?
How far was the impressive international economic performance the result of international financial arrangements, and how far was the successful outcome simply a consequence of the chance absence of destabilizing shocks? What is the relationship between the system and the incidence of shocks? Did the discipline imposed by the “classical Bretton Woods system” avoid little shocks by postponing them until they accumulated into the giant inflationary shock of the 1970s that destroyed the system?
The counterpart of trade liberalization was the adoption in Europe between 1958 and 1961 of current account convertibility, and the acceptance of Article VIII of the Fund Articles of Agreement on convertibility in place of the protection offered by Article XIV. It is only at this moment that it became possible to speak of a genuine international monetary system: indeed, the phrase only entered the general vocabulary in the context of discussions in the early 1960s regarding the creation of the General Arrangements to Borrow.4 There was no longer simply a world of nation-states, and an institution supervising rules on conduct, but rather a system in which nation-states, and the rules affecting their behavior, interacted with markets, and in particular with growing currency and capital markets. (This makes the international financial system conceptually entirely different from the international political system: in the latter, states alone interact with each other, while in the former there is a large body of other agents, private as well as official, continually judging and second-guessing the decisions made by the states.)
As trade expanded and the “system” developed, its management required additional resources or reserves. Inadequate reserves would be the most frequent reason given for a refusal to open trade, or for a willingness to contemplate higher levels of protection. A monetary shock might reverse the successes of trade liberalization. Finding a solution to the reserve problem thus took on the character of a quest for stability in the system as a whole. Unfortunately, this question touched the most sensitive political nerves. In what had become more and more a dollar system (or perhaps a dollar-sterling system), the expansion of dollar reserves looked to many non-Americans like an extension of U.S. power. Even more frustratingly for Europeans and others, it was a costless extension: by making the rest of the world hold dollars, the United States was exchanging its paper for the goods and services of others.
In order to operate smoothly, any international financial system requires three elements: liquidity, confidence, and adjustment.5Liquidity, in the form of a maintenance of adequate reserve levels, was essential to the operation of convertible currencies with par values: the reserves allowed the financing of short-term deficits in the balance of payments, in cases when the causes of those deficits were assumed to be temporary. Confidence allowed the augmentation of reserves through borrowing. If there were longer-term difficulties, eventually the supply of reserves and the fall-back supply of credit would be limited and the country concerned would require adjustment of its external position.
The pre-1914 gold standard had generated a high degree of confidence and had operated on surprisingly low levels of reserves (liquidity), but it had required the political sacrifice of painful adjustment to deflation by its members in the 1870s and 1880s. After the First World War, the gold exchange standard had provided a greater degree of liquidity (relative to world trade) than the prewar system, but had suffered and in the end collapsed because of confidence problems.
After 1945, and particularly in the 1960s, the existence of confidence allowed the delaying of adjustment, especially in the case of two major reserve centers, the United Kingdom and the United States; as a result, when adjustment was eventually required, when liquidity was exhausted and when confidence ebbed, the extent of the adjustment that would have been needed to return the system to smooth operating was too large to be acceptable to national policymakers. If adjustment under the gold standard had been too painful, under Bretton Woods it was too delayed because of the high level of expectations about growth. The success of the Bretton Woods system actually contributed, through increasing expectations, to the growing difficulties of managing the system. The postponement of the adjustment mechanism was further encouraged by the great difficulty of making the exchange rate an instrument of international economic policy. And when adjustment came to be required, in 1971, it was too large to be handled in any way except by the improvised measures of crisis management.
The system worked as long as it did for two reasons: the first economic, the second political.
In the first place, the expansion of international credit during the 1960s provided a means of establishing larger reserves and of postponing adjustment. The composition of reserves changed over the course of the decade (Figure 6-1), and foreign exchange played a larger role, as the relative share of other reserves fell. The U.S. dollar in particular became the world’s major reserve. The supply of gold was limited by production conditions and the rise of production costs relative to the fixed price of gold ($35 an ounce); IMF quotas were increased in 1959 and 1966, but failed to keep pace with the rise in world trade (Figure 6-2). Reserves outside the United States were stocked up, as the counterpart of the outflow of dollars from the United States (largely for investment) in practice filled the “reserve gap.”
Figure 6-1.Total World Reserves as a Share of World Imports
Source: International Monetary Fund.
Figure 6-2.IMF Quotas as a Share of World Imports
Source: International Monetary Fund.
Second, confidence in the operation of the system was increased because of the interlinking of political and economic calculations. The major industrial economies were linked not simply through economic intercourse, but also through security provisions and alliance systems. The existence of the military and security calculation enhanced the working of the economic relationship, which otherwise would have been much more strained, between the United States and the United Kingdom, and the United States and Germany. France, the country that most vigorously challenged the principles on which the international monetary system operated, was also the state with the most hesitation about the security system and the North Atlantic Treaty Organization (NATO). When the United States began to show signs of political overstretch, economic confidence began to wobble also.
The development that had most facilitated the exchange liberalization at the end of the 1950s was the international liquidity generated as a result of large U.S. balance of payments deficits. These provided at least a temporary solution to the liquidity issue. A current account surplus of $3,556 million in 1957 turned into a small deficit of $5 million in 1958 and a deficit of $2,138 million in 1959. In 1958 and 1959 capital flows produced large balance of payments deficits: measured on a liquidity basis the deficits amounted to $3,365 million in 1958 and $3,870 million in 1959. Such a development of U.S. deficits should not have been worrying from the point of view of the technical operation of the international monetary system. Since the overwhelming majority of the Fund’s drawings between 1945 and 1959 had been in dollars (as a consequence of the dollar scarcity), the United States had a large potential claim on the Fund that it might use to finance its deficits.6
In practice, the United States did not need to draw on the Fund until a late stage in the development of the deficit problem, as other central banks held increasingly large dollar reserves. The flow of dollars thus solved the liquidity problem and greatly eased the process of adjustment, but worried those who believed that there should be some regulation of liquidity growth. As it was, from 1959, the world began to face a new dilemma. It was sometimes put in this form: the U.S. dollar was, like the notoriously unsafe (American) Corvair automobile, “unsafe at any speed.” If the growth of dollar reserves occurred at too slow a pace, it would make convertibility much harder for other countries and thus serve as a brake on world economic development. If, on the other hand, dollar creation ran too quickly, it would push inflation across frontiers. At least for the first half of the 1960s, many observers believed that the major problem lay in the likelihood of U.S. deficits continuing only for a short time, as most prognoses envisaged a return to surpluses.7
The academic version of the Corvair thesis appeared first in the context of a criticism of international institutions for not regulating and controlling the development of world liquidity, and of a memory of the early 1930s as a fearful precedent for what might happen if international money were left to the uncontrolled whims of the marketplace. Failure systematically to deal with the liquidity issue would mean a resurgence of doubts about confidence. The analysis was couched in terms highly critical of both the Bretton Woods system as actually applied, as well as of the intellectual legacy of John Williams’s key currencies approach, which had exercised so dominant an influence. Robert Triffin’s meticulous dissection of the intellectual shortcomings of Williams’s vision became ever more compelling during the course of the 1960s. It was an argument that Triffin had begun to formulate already before 1959 and the evident end of the postwar dollar shortage in the context of a general critique of the international financial system, and also before the general transition to convertibility.
In 1957, he wrote of “the over-all failure of the Fund to affect significantly the course of events in the postwar world.” “The Fund’s administrative structure unfortunately constitutes the worst possible instrument for the negotiation of such international agreements and compromises among sovereign countries. Permanently in residence in Washington, cut off from regular policy-making responsibilities in their own countries, the members of the Executive Board inevitably tend to become little more than glorified messenger boys, dependent on instructions from their home countries for all substantive matters submitted to the Board. This does not create an effective forum for negotiation.”8 He believed that the most effective model for world economic integration lay in the legacy of the European Recovery Program, the European Payments Union, and the Organization for European Economic Cooperation and that the IMF should learn from these successes. It “should also lean heavily on the regional organizations and agreements that promote [its] basic objectives, and should openly recognize the justifiable exceptions to the rule of nondiscrimination which may be inseparable from progress toward trade and exchange liberalization.”9
He bitterly criticized the way in which the Williams key currency approach had de facto—and quite contrary to what he held to be the intentions of the Bretton Woods “founders”—emerged as the monetary foundation for the world economy. It would endanger the establishment of convertibility from the outset. A repetition of 1931 would be the inescapable outcome of such an approach: the moment when claims on a key currency such as sterling would be realized, the available liquidity would be exceeded, and in consequence a worldwide rapid deflationary contraction would be induced. “History might easily repeat itself tomorrow if convertibility were to be restored today on the same unorganized gold exchange standard basis. Short-term dollar and sterling liabilities to foreigners … are now about twice as large as total gold holdings outside the United States and the United Kingdom … Their partial conversion into gold, in the event of renewed currency fears, could easily wreck sterling convertibility and might even create serious, although still manageable, financial problems for the United States itself.”10
This gloomy prognosis about the likely consequences of convertibility required only little modification after the introduction of general convertibility between 1958 and 1961; in its modified version of a diagnosis of a general liquidity shortage it came by the mid-1960s to command nearly universal consensus. The theory became known as the “Triffin dilemma”: that either there would be a general shortage of liquidity or the United States might create liquidity through the provision of dollar assets, but that these might become inconvertible in a 1931-like generalized financial crisis. In a “1931,” the realization that the quantity of dollar reserves far exceeded the U.S. holding of assets, and that the dollar was in consequence strictly speaking inconvertible, would in itself induce panic. In such a crisis, the fundamental problem of liquidity shortage would re-emerge.
This view was elaborated in two articles published by Triffin in 1959 and then in 1960 as a book.11 In retrospect, looking back from the perspective of the years after 1971, when dollar convertibility and the Bretton Woods par value system indeed collapsed spectacularly, Triffin’s warning seemed highly prophetic. But the analysis failed to explain ten years of the working of the “system” of convertibility: ten years in which international trade and production expanded, and more and more countries came to adopt domestically an expansionist and allegedly Keynesian approach to demand management at the same time as maintaining the restoration of international convertibility.12 The 1960s were without doubt the heyday of the Bretton Woods system.
Over this period the system was subtly changing: in the first place, because of the emergence of new surplus countries, Germany and then later Japan. The United Kingdom and the United States—the principal reserve countries—found their situations increasingly difficult and needed to resort to expedients, devices, gimmicks, and ad hocery in order to maintain their international position. The international monetary system began to require a degree of management that no one had foreseen. By the late 1960s, the system was lurching from one crisis to the next, and the realities of power politics seemed to frustrate any attempts at a systemic reform.
The sharp deterioration of the U.S. balance of payments in 1959 facilitated the process of world economic liberalization and integration, but it also shocked many observers. Per Jacobsson, and many Europeans, interpreted deficits as the result of uncontrolled domestic inflationary pressure in the United States. “I explained that the Europeans expected that the Americans would have to do more or less the same things as they had done themselves—including the holding back of wage increases.”13
For the first half of the next decade, however, the U.S. external problem lay fundamentally on the capital account. In fact, the trade balance gave no indication whatsoever of any sustained shift in the international structure of comparative costs. The U.S. merchandise balance indeed increased consistently—from a low of $1,148 million in 1959 to $6,801 million in 1964. The two major sources of concern were the government sector and private capital flows. Government outflows were primarily a result of military spending, which involved transfers of an average of $3 billion yearly between 1960 and 196414 The outflow of private capital from the United States also rose dramatically. Private flows doubled from 1960 ($2.1 billion) to 1964 ($4.5 billion).15 At the same time, the United States attracted short-term funds from the rest of the world. The development of the U.S. external account reflected national differences in patterns of demand and supply of capital and in saving and investment behavior. The international monetary system performed well here in compensating for national idiosyncrasies. In Europe and elsewhere, there was a substantial demand for long-term capital, but investors were used to maintaining short-term and liquid accounts. In the United States, on the other hand, a broad public was accustomed to investing long term at fixed interest rates. The flows of the early 1960s thus reflected different liquidity preferences.16 Or, to put this point another way, the United States was in practice acting as a banker for the rest of the world in the manner characteristic of many bankers with a mismatch of liquidity: borrowing short and lending long.17
This development contained the danger of a Triffinite collapse. The question arose whether to add more liquidity through some artificial creation, as a substitute for the world’s need for ever more dollars, which would allow and require reduction of the U.S. payments deficit. A change in the gold parity of the dollar would be one way of accomplishing this. The chief economist of the Bank for International Settlements (BIS), Milton Gilbert, continually tried to press behind the scenes for a change in the dollar price of gold, but this option never seemed very realistic. U.S. Treasury officials had discussed devaluing the dollar in respect of other currencies in the later years of the Eisenhower administration, although they were unwilling to change the dollar’s gold parity.18 The British Prime Minister Harold Macmillan told President Kennedy in 1962 that “if the gold price were raised to $70 an ounce, most of the difficulties would be over and done with.”19 But Kennedy believed it his duty to resist inflationism and believed that the notion of a dollar devaluation was profoundly immoral. Arthur Schlesinger, Jr., records that “he used to tell his advisers that the two things which scared him most were nuclear war and the payments deficit.…. He had acquired somewhere, perhaps from his father, the belief that a nation was only as strong as the value of its currency.”20 On July 23, 1962, in the first televised address of a U.S. President carried across the Atlantic, Kennedy announced that he would not alter the gold price. Chapter 4 has already described how before this statement, the U.S. refusal to act made unavoidable the deutsche mark revaluation of 1961. A legacy of Kennedy’s very firm attitude was that the subject of a change in parity became a tabu over the course of the 1960s.
Instead of parity alterations, let alone a systemic reform to tackle the balance of payments problem, the United States adopted a series of essentially technical devices to minimize the extent of disturbance caused by U.S. deficits. Robert Roosa, the Treasury Under-Secretary for Monetary Affairs, called them the “outer perimeter defenses” of the dollar.21 At the same time, he opposed the broader systemic solutions such as dollar devaluation or gold revaluation or Triffin’s idea of creating additional liquidity through a world central bank.22
The Outer Perimeter Defenses
The United States followed a number of strategies for the defense of the dollar. All involved moving responsibility for the management of the international monetary system away from the IMF. As the dollar problem became ever more acute, the United States increasingly felt impelled to act on its own and outside the mechanism provided by the system.
The Gold Pool. At the beginning of the decade, there had been substantial volatility on the international gold market. The fixed price of gold relative to national currencies, combined with a gradual price inflation, made gold look cheaper and led to greater gold use for artistic and industrial purposes. A potential undervaluation of gold then led speculators to test the market. Over the weekend of October 15, 1960, it became clear that John F. Kennedy was very likely to be elected as President of the United States, and many Europeans believed that this would mean an American inflation promoted by expansionist Keynesian economists.
On October 20, after the opening of American markets, the London gold price went far above the U.S. gold parity, to $40 an ounce. The panic demonstrated to central banks and governments the need to intervene in disorderly markets. This situation was managed very differently to the similar problem in the late 1940s. At that time, after an increased demand by gold hoarders in response to inflationary threats, the IMF had used its leverage to discourage the sale of gold at premium prices. It had been able to act quite effectively, in part because the largest Western gold producer, South Africa, had had payments problems and required Fund support. When some gold was sold at premium prices on the private market, the Fund’s response was not to condone this, but rather simply to wait for the market to turn around.23 As confidence returned, the premium price of gold gradually fell, and disappeared by 1953. In the early 1960s, however, the problem was tackled in another fashion: not through the IMF, but through market operations conducted by the central banks of the major European countries and of the United States. In 1961, the United States with Britain, Switzerland, and the countries of the European Economic Community (EEC) agreed jointly to commit $270 million gold, which would be used for interventions on the London market in order to prevent the price rising above $35.20. The participants agreed not to buy gold in the London market, or from the major gold producers (the U.S.S.R. and South Africa). They believed that if speculative demand for gold were eliminated, as a consequence of the existence of the pool and the commitment it represented to the $35 price, in the long run the supply of gold would exceed the demand and there might consequently be a substantial price stability.24 It was hoped that in this way, through a demonstrative commitment of central banks to the $35 price, the need for any adjustment in the dollar price of gold could be eliminated.
The Swap Network. In addition, the world’s major central banks created an international machinery for dealing speedily with temporary imbalances. It allowed the central banks to function collectively and very effectively as an emergency lender of last resort. Already in the 1950s, the BIS had organized within the European Payments Union what amounted CO a swap system for European central bankers. The swap network inaugurated by the Federal Reserve Bank of New York revived in a straightforward way the day-to-day collaboration maintained in the 1920s between Governors Strong and Norman. It established a pre-arranged volume of short-term credits between central banks, which might be used to counter destabilizing market movements (for instance, if the pound was weakening against the dollar, the Federal Reserve would provide the Bank of England with dollars for a support operation). The network amounted to $900 million at the end of 1962 and grew to almost $30 billion by 1978.25 For the first part of the 1960s, this cooperation proved very effective, although there were warnings, particularly from the British side, that the mechanism might fail in the case of a simultaneous attack on the dollar and the pound, and that the machinery might then do nothing more than give “the appearance of two lame ducks helping each other.”26 Such a lender of last resort could only operate well in a small or limited crisis. But crises have a tendency to spin out of control; and it was possible to envisage situations arising in which the resources of the central bank swap network would be inadequate.
Offsets. In addition, the United States tried to induce other countries to take action to support the U.S. balance of payments. From 1962, dollar-denominated 12-to 24-month securities (“Roosa bonds”), as well as foreign currency bonds, were sold to foreign central banks as an alternative to sight deposits of dollars. It was also expected that countries that benefited from U.S. military protection would buy substantial quantities of military equipment. When, for instance, Germany joined NATO in 1955, and established a new army, it bought its supplies almost entirely from the United States. As military sales slowed down after the Bundeswehr reached full size, the United States created substitutes for military sales. After 1960, the United States asked countries in which a substantial amount of U.S. military expenditure was undertaken to make balance of payments “offsets”: funds placed in the United States for over a year in order to balance U.S. capital movements. In addition, countries experiencing large short-term inflows took measures to discourage them. After 1964, Germany, for instance, required banks taking foreign credits to make no-interest deposits at the Bundesbank proportionate to the capital inflow. As a result, foreign borrowing became more costly than borrowing on the German market.
Tax Measures. Measures were taken to slow down the outflow of private capital. In 1963, the United States introduced an interest equalization tax on the purchase of foreign stock and fixed interest securities as a way of counterbalancing the tax incentives to purchase foreign securities that had previously existed. In February 1965, U.S. banks were asked “voluntarily” to limit foreign lending as part of a package to deal with the U.S. balance of payments problem. At the same time, the interest equalization tax was extended. In 1968, the Federal Reserve was given the power to make bank restrictions on foreign lending mandatory, but in practice never exercised this right.
The GAB. A new mechanism was created within the IMF to supply reserves through a credit mechanism in case of a shortage in a major industrial country: the General Arrangements to Borrow (GAB), launched on October 2, 1962. At first, the proposal to create this facility was widely known as the Jacobsson plan.
The General Arrangements to Borrow
General convertibility brought a potential for much greater movements in international reserve levels. Despite the existence of the swap network, a situation might arise in which, faced with a drain on reserves, both the United States and the United Kingdom would need to undertake large-scale short-term borrowing. The IMF would not possess sufficient resources for this purpose. Conceivably, such a possibility could be anticipated through an increase in Fund quotas; but there had already been a quota review and general increase in 1959. In addition, if the borrower in the possible emergency were to be the United States, it would find that there were not sufficient nondollar and nonsterling funds in the IMF to meet the notional U.S. right to borrow. As an alternative, therefore, the IMF envisaged the creation of an additional facility by those industrial countries with an interest in preserving the functioning of the world reserve system.
Jacobsson had a broader purpose in sponsoring the new facility. He saw cooperation between the leading industrial countries as an essential part of world economic stabilization. In 1960, he noted after a lunch with William McChesney Martin, the Chairman of the Federal Reserve Board: “To succeed in an antideflationary policy they must work together—but it is sufficient that the leading industrial countries work together i.e. U.S.A., Great Britain, France, Germany and maybe Italy and japan. These countries are, I think, able to determine together… the price trend—through appropriate open market and wage policies. Fortunately they all by now have more or less adequate reserves.”27 The purpose of the new facility would be to have an emergency mechanism in place should a crisis arise, but that need not be used in other circumstances. Jacobsson argued that “borrowing could not really be improvised in a crisis: some advance arrangements had to be made.” The existence of such a security provision would restore confidence in the two tottering key currencies of what in practice had become a reserve currency system.
The problem that arose in formulating the new mechanism involved the degree of control that the lenders could exert on the policies of the borrower, or, to use the old concept derived from the arguments before and after the Brerton Woods conference, the degree of automaticity. Initially, for the United States as well as the United Kingdom, the main attraction of the new proposal was that it would be speedy and noninterventionist. It would bring money quickly without “advice” or control from multilateral agencies. Treasury Secretary Douglas Dillon stated that “to be effective, the additional resources must be promptly available in case of need.”28 Jacobsson warned the United States that unless the Fund scheme were accepted, “an obstinate Managing Director might refuse to initiate borrowing procedures before the U.S. budget balanced.”29
Convincing the Europeans involved the exact converse of the exercise: providing an assurance that there would be a sufficient control of the use of the borrowed resources. According to the continental Europeans, the Fund should “have the opportunity, by a collective judgment, to confirm that also in their opinion the special emergency, for which the additional resources have been set aside, has indeed arisen.”30 In the Netherlands and Germany, which were establishing themselves as the bulwarks of monetary orthodoxy, there were worries that the new system would mean a much more relaxed approach to international monetary management. These two countries also argued that it might be difficult for the countries currently operating surpluses (themselves) to draw on the GAB once their own reserves had been committed in GAB loans. French politicians held the belief that the United States might abuse its position and that Britain was trying to borrow on more generous terms than it had been accorded by the European countries between March and July 1961 in the aftermath of the inadequate deutsche mark revaluation. The French Finance Minister Wilfrid Bnumgartner consequently insisted that in each case each potential lending country should decide whether its resources were needed and should be used.
Jacobsson traveled to Europe in the last months of 1961 to obtain a general agreement from the continental governments. He needed to convince the new skeptics that the Bretton Woods institutions had an effective power over not only British but also American behavior. He claimed that the Fund and the World Bank had an important leverage in Congress, because they were considered “sound” and because their reputations had been enhanced by the “attacks of economists wanting more expansion and therefore blaming the Fund for insisting too much on stabilization.” In addition, political circumstances might bring exchange problems to Europe: a crisis over Berlin, for instance, might lead to an outflow of funds from Germany that might quickly also affect France. “Europe might once again one day experience difficulties in its balance of payments—there w[oul]d be little chance of another Marshall Plan—this Fund plan contained nor only provision for Europe providing funds but also advantages for Europe.”31
The result of these negotiations became known as the General Arrangements to Borrow. A $6,000 million sum provided by ten countries or central banks (the United States, the Deutsche Bundesbank, the United Kingdom, France, Italy, Japan, Canada, the Netherlands, Belgium, and Sweden) would be available “to forestall or cope with an impairment of the international monetary system.”32 This group subsequently became known as the Group of Ten. Jacobsson also pressed Switzerland, though not a member of the Fund,33 to join the GAB in the spirit of cooperation between major financial centers (which it eventually did in 1964). Requests for borrowing would be dealt with according to the Fund’s policies.
After discussions with Baumgartner, Jacobsson pledged that before making a call under the GAB, the Fund’s Managing Director would obtain the consent of the participating countries. The agreement was recorded in a letter of Baumgartner’s to the other participants in the GAB.34 These would consult with each other and inform the Fund of the amount they were prepared to lend. The French provisions were accepted by the Fund’s Executive Board on December 18, 1961. A request for support under the GAB required the consent of the GAB members as well as the Executive Board. The principle underlying the GAB of a “double lock” represented a major dent in the Fund’s claim to universality and to a capacity to judge by itself the conditions for assistance in dealing with balance of payments problems.
The GAB and its members rapidly aroused the suspicion that a new ideology of cooperation between industrial countries had replaced the universalist aspirations of Bretton Woods. The G-10 seemed a “very exclusive club.”35 Australia and Portugal immediately protested their exclusion from the “special club” of strong convertible currencies.36 The GAB appeared to mark a division of the world into haves and have-nots, a distinction often endowed with a Manichaean property of the struggle of good against evil. Although the backdrop to the formation of the GAB and the G-10 indisputably lay in the potential problems of the United States and the United Kingdom, the ideology of some of the GAB’s more self-righteous founders presented haves in general as maintaining stability in their balance of payments, while the feckless have-nots, outside the G-10, required constant external support. As a result, the new arrangements were one element in provoking among industrializing countries the elaboration of a counterdoctrine of development, and a feeling that their interests too should be given an institutional form (which to some extent was realized in the creation of the United Nations Conference on Trade and Development (UNCTAD)). In this way, the GAB had perhaps solved a practical problem, but at the same time it appeared to divide the world economy.
Cooperation among the G-10 became institutionalized through study meetings of deputies (senior or permanent civil servants at the heads of finance ministries or central banks) from October 1963, chaired initially by Robert Roosa, the U.S. Treasury Under-Secretary for Monetary Affairs, and after 1965 by Otmar Emminger, Vice-President of the German Bundesbank. The group committed itself to “undertake a thorough examination of the outlook for the functioning of the international monetary system and of its probable future needs for liquidity.” It defined its task as “multilateral surveillance,” which it interpreted as an appraisal of the various “means of financing surpluses or deficits” in order to develop “a common approach to international monetary matters.”37
Debate about reform of the system shifted for a time to this new forum. The G-10 suggested that the examination of ways of maintaining balance of payments equilibrium should be conducted through the OECD’s Working Party 3—a group whose country membership was then identical with that of the G-10. It created its own study group on the creation of reserve assets. The industrial countries, with a heavy representation of Europeans, appeared to be on their way to making their own international financial system. Their actions meant a challenge to a more global or universal vision of the workings of the international economy. At the Tokyo meetings in 1964 of the Fund and the Bank, Jacobsson’s successor, Pierre-Paul Schweitzer, defended the role of the Fund in international monetary policy: “Where decisions are taken to create and administer liquidity by deliberate international action, it is particularly important that the advantages of the multilateral institutional approach be kept in mind. An international organization provides the forum for a balanced consideration, and hence the best reconciliation, of the various objectives in the international financial field as they affect all countries.”38
The Fund’s assertion of its role in international monetary relations involved a new quota increase in 1965, with a general increase of 25 percent for all quotas, and special increases for 16 members, reflecting the changing character of the international economy. But it also included the consideration of more ambitious schemes—in particular, for the creation of a reserve unit. Some of this discussion took place within the IMF, but a great deal of the impetus came from the outside.
The problems of the dollar as a reserve currency started a debate about reserve creation, and, in particular, about the possibility of supplementing the stock of “unconditional” reserves rather than the “conditional” reserves that existed through the IMF, its quota system, and its lending operations by means of the GAB. Additional unconditional reserves would obviate the need for the external judgments that needed to be made on states’ policies if they were to draw more than their gold tranche from the Fund, or if they were to borrow under the GAB. They seemed in consequence an easy way to defend national policy autonomy in the international monetary debate.
The most far-ranging of the initiatives for a new reserve system originated outside the United States, although the debate only began to be serious once the United States too was convinced of the need for reform. The British Prime Minister Harold Macmillan had already lectured U.S. President Kennedy very shortly after his election on how a “united Free World was more likely to be achieved through joint monetary and economic policies” than through political or military alliances.39 He believed the prevailing monetary system, in which the United States “had all the marbles,” to be fundamentally unfair. In 1962, the British Chancellor of the Exchequer, Reginald Maudling, set out a scheme for a “multiple currency account,” in which countries would acquire the currency of another country or countries that were temporarily in surplus and as a result establish claims on the account. But this looked too much like an example of British special pleading to attract much support. Roosa commented that “the pity is that one begins to wonder about British wisdom.” At a press conference during the IMF Annual Meeting he declared that there would be no global liquidity problem at least for the next five years. In fact, he feared that any reform would make it harder for the United States to finance its payments deficits, jacobsson thought that the British scheme was “half-baked” and tightly believed the British Prime Minister Harold Macmillan to be behind it. “He still lives in the depression,” Jacobsson noted. “Macmillan is a New Dealer masquerading as a conservative.”40 A similar scheme to the British one came from the veteran Greek Central Bank President, Xenophon Zolotas, but did not encounter any substantially more enthusiastic reception.
In 1963 Edward Bernstein, the former Director of the Fund’s Research Department, had proposed a more adequately balanced composite reserve unit (or CRU) constituted of the 11 currencies of the GAB countries. The IMF took up the debate and produced a study in 1964 that suggested that the need for international liquidity could not be met simply by adding more gold or foreign reserve assets.41 Since October 1963, a G-10 deputies’ study group had been engaged in an analysis of “the functioning of the international monetary system and of its probable future needs for liquidity.” Its 1965 report argued that the world would shortly face a liquidity shortage: “from the considered expectation that the future flow of gold into reserves cannot be prudently relied upon to meet all needs for an expansion of reserves associated with a growing volume of world trade and payments and that the contribution of dollar holdings to the growth of reserves seems unlikely to continue as in the past.”42
The developing countries produced their own proposals in 1965, in a report of an Expert Group on International Monetary Issues, prepared for the newly launched UNCTAD. According to this statement, the provision of long-term aid, rather than the creation of additional liquidity, should be the main priority in reforming the international monetary system. Liquidity provision could only help to solve short-term problems and would not contribute to longer-term stable development. International liquidity creation should be linked to development finance; and developing countries should be represented in discussions of monetary reform. In a similar vein, in May 1966 the Group of Thirty-One Developing Countries insisted that “monetary management and co-operation should be truly international and … all countries, which are prepared to share in both the benefits and obligations of such new monetary arrangements as may be devised, should be eligible to participate in the creation of new reserve assets.”43
In the IMF’s Executive Board, the representatives of developing countries similarly insisted that “all countries, not only the economically big ones,” should be involved in the discussion of reform.44 Schweitzer warned against the division of member countries into “the reliable few and the less responsible many.”45 He was suspicious of proposals for a CRU composed of a few currencies and doubted whether they would work well in practice. By 1965, he was insisting that “international liquidity is the business of the Fund”: that the IMF was central to the operation and surveillance of the international monetary system.46
Eventually the United States reacted to the flood of initiatives for international monetary reform, for its own reasons. As the use of the dollar as a reserve by other countries grew—in other words, as U.S. liabilities to the international system grew—the United States faced a reserve shortage. A new artificial reserve asset might be a way of providing much-needed additional liquidity for the United States. In July 1965 U.S. Treasury Secretary Henry H. Fowler indicated his partial conversion to Triffinism when, in a speech to the Virginia State Bar Association, he called for an international monetary conference to consider “what steps we might jointly take to secure substantial improvements in international monetary arrangements.” The speech surprised the Europeans. When the central bank governors next gathered in their regular meetings at the BIS in Basle, according to one observer, there was “astonishment, puzzlement and resentment.” In particular, the Europeans believed that the G-10 was the most suitable forum for the reserve discussion and that there was absolutely no need for a new Bretton Woods. The Fund immediately engaged itself in the controversy. It started an ultimately successful offensive to establish itself as the appropriate forum for this monetary discussion, rather than either the G-10 or a new conference. In the course of these debates, it prepared its own suggestions for monetary reform and the creation of liquidity by a mutual exchange of claims among members of the Fund.47
The IMF arranged joint meetings between the G-10 deputies and the Fund’s Executive Board as a forum for the discussion of reserve creation in 1966. The large gathering that resulted—an assembly of 105 people—perhaps perversely appeared to make reaching agreement rather easier than if the same discussions had taken place in a more intimate setting with a more frank interchange of views. But it made the discussion of reform proposals very cumbersome and encouraged the use of deliberately obfuscating linguistic formulations in order to secure a basis for agreement. The most obvious of these concerned the name of the new international reserve: composite reserve unit suggested a clear challenge to the position of the U.S. dollar and might offend American sensibilities.
There existed a double problem regarding international reserves. The dollar in the 1960s had formed an increasing proportion of the world’s reserves. But the growth of dollar holdings depended on U.S. actions. If the United States reduced its payments deficit, there would be a slowing in the growth of exchange holdings in other countries. In 1965, the U.S. balance of payments deficit had been lower than in any year since the 1950s, and the U.S. administration also cut back the budget deficit quite effectively. It seemed that the confidence crises of the early 1960s had come to an end: but equally, that the long-awaited world reserve shortage would be realized. Other countries would begin to suffer from the consequences of an insufficiently large creation of dollars and would not be able to build reserves to finance their own potential deficits.
At the same time, the pivotal position of the United States in reserve creation could seem like a use or abuse of power. Germany and particularly France produced a vigorous criticism of the U.S. position as being openly dominating. Jean-Jacques Servan-Schreiber’s book, The American Challenge, presented the capital flows of the 1960s as establishing a potential U.S. dominance over the European continent. Since 1958, he stated, U.S. corporations had invested $10 billion in Western Europe, over a third of all overseas investment. “The Common Market has become a new Far West for American businessmen. Their investments do not so much involve a transfer of capital, as an actual seizure of power within the European economy.”48
French politicians focused on the reserve issue and insisted on the pre-eminent role of gold as a means of countering the American challenge. Metallic reserves, and an attack on the key currency concept, meant for them the assertion and preservation of national sovereignty in the face of the challenge of internationalization. But it also meant the rediscovery of the international financial system as a rules-based order that would prevent the hegemonic country drawing additional privileges from the pre-eminent position of its currency. Gold as the basis of the system would prevent the kind of manipulation that might be undertaken in a reserve currency system. In this way, it offered an escape from the conflation of economic and national security concerns that had, in the French interpretation, come to dominate and paralyze the international financial system. A popular way of putting this was to claim that the use of the dollar as an international reserve currency gave the United States illegitimate “seigniorage” advantages. It could finance its military expenditure overseas painlessly by simply applying green ink to paper. (In fact, since holdings of dollar reserves bore interest, the U.S. outflow could not truly be said to be a pure seigniorage gain for the United States.)
The French Finance Minister Valéry Giscard d’Estaing announced at the Annual Meetings of the IMF and World Bank in 1964: “The world monetary system must be set in concentric circles: the first one being gold, and then, the second, if necessary, recourse to deliberate and concerted creation of either reserve assets or credit facilities. The inner circle is gold. Experience in recent years has shown us that, aside from any theoretical preference, gold remains the essential basis of the world payments system.”49 In a press conference on February 4, 1965, General de Gaulle attacked the U.S. abuse of its “exorbitant privilege” under the gold exchange standard. “The convention whereby the dollar is given a transcendent value as international currency no longer rests on its initial base, namely, the possession by America of most of the gold in the world. The fact that many States accept dollars as equivalent to gold, in order to make up for the deficits of any American balance of payments, has enabled the United States to be indebted to foreign countries free of charge.”50 (For de Gaulle, suspicions about the political manipulation of paper currency by Americans had a long history: he had been involved in a bitter clash in 1944 over what he regarded as the overissue of U.S. military currency in France.)51 In these circumstances, France could only hope to reform the system by shaking it. A sustained attack on the dollar might demonstrate the nature of the fundamental systemic flaws. France had already begun at the outset of 1965 to reduce its dollar holdings by converting them into gold. Foreign exchange held by France fell from $284 million at the end of 1964 to $112 million by 1966, while gold holdings increased from 106.54 million ounces to 149.66 million ounces over the same period.
The French dollar conversion campaign inevitably made the discussion of replacement of the reserve role of the dollar highly politically sensitive. To the United States, the attraction of reform lay in taking the strain off the dollar, and avoiding new dollar crises, by the provision of additional liquidity. France’s main concern was to stop the “abuse of the system.” After a speech by Giscard d’Estaing at the IMF Annual Meeting in October 1963, in which he had criticized the composition of currency reserves, France set out in the meetings of the G-10 deputies a proposal for a collective reserve unit (CRU). In June 1965, Giscard prepared a detailed program. The CRU would be linked to gold and be composed in fixed proportions of the currencies of the G-10 countries.52 Such a course, French policymakers now believed, offered the only way of raising the price of gold and thus restoring an international monetary order based on the “inner circle” of value. General de Gaulle convinced himself that the planned reserve unit might constitute a defense “contre l’inflation américaine.” It should be linked to a requirement that the United States reduce the volume of outstanding dollar liabilities. The main purpose of the reform in French eyes would be to provide additional credit facilities, so that new reserves would not be needed. This would be conditional rather than unconditional liquidity. The credit facilities would require a tight institutional control. All these aspects of the French vision required a much more powerful role for the IMF. In early 1967, France set about convincing the other members of the EEC of this position and also demanding that the voting system of the IMF should be reformed in order to give the Six a veto on important decisions.53 As a result of French demands, in July 1967, the word “reserve” was dropped from the discussion of the composite reserve unit and the concept was now rephrased as a “special drawing right” or SDR.
There needed to be an additional security against abuse of the SDR, the continental Europeans believed. De Gaulle believed that only a link with gold would give such security, but the United States resisted this precisely out of fear that it might lead to French pressure to change the gold parity of the dollar.54 The other Europeans followed de Gaulle in insisting that approval by the IMF’s Board of Governors of the issue (and potentially the cancellation) of SDRs should require a majority of the Fund’s voting power so large as to give Europe a veto. The eventual stipulation required a vote of 85 percent of the voting power, including at least half of the major creditor countries. With 16.5 percent of the voting power, the EEC countries thus obtained their veto on SDR allocation. The result made voting very complex: the same 85 percent requirement was applied to the vote needed for increases in quotas after quota reviews. Majority voting in international institutions had become a highly sensitive issue, especially in Europe. France had just fought a bruising campaign against the wish of other members of the EEC to adopt a majority voting principle. In both the European and the international monetary case, concerns about national sovereignty led to a limitation of decision taking by majorities and of the capacity of international bodies to work as effectively as they might otherwise have done; and in both cases, the excessive sensitivity provided an obstacle to further development.
In September 1967 the Fund’s Executive Board approved a draft outline for the creation of the SDR. But the debate still went on. The United States and the United Kingdom insisted that the new asset would constitute “frontline reserves.” The United States urgently needed additional reserves in order to lessen the strain on the dollar. France asserted that it could not be a new currency designed to “replace gold” but merely represented “the possible extending of credit facilities,” and eventually decided not to cast a vote for the SDR in the IMF’s Board of Governors. The original French version of the plan had been undermined by the alternative American suggestion, and then defeated. As a witticism of the time had it, “the CRU (cru = raw) had been cooked.”55 Across the Atlantic de Gaulle gave an agreeable interpretation. He told the story of two men who went out into the woods on a search for treasure, and after a long and apparently futile search they concluded that they had not found the elusive gold. But instead, they had discovered “l’amitié.” The German Economics Minister gave a stage whisper: the friends had discovered the SDR.
The SDR was approved by the Governors at the twenty-third Annual Meeting in 1968, and the amendment to the Articles of Agreement came into force on July 28, 1969. The purpose of SDRs (Article XXIV, Section 1(a)) was stated as follows: “In all its decisions with respect to the allocation and cancellation of special drawing rights the Fund shall seek to meet the long-term global need, as and when it arises, to supplement existing reserve assets in such manner as will promote the attainment of its purposes and will avoid economic stagnation and deflation as well as excess demand and inflation in the world.”
The intense and highly politicized debates that preceded the creation of the SDR ensured that the new instrument was surrounded by a protective web of regulations. It had no backing, and it did not represent a claim on any other asset. It could be regarded as the first international currency to be created in the manner of a national paper currency—purely through a series of legal obligations to accept it on the part of members of the system. As a result, it was widely suspected as “funny money.”56
SDRs were administered as a separate account in the Fund. Each member was entitled but not required to participate in the scheme. The provisions creating the SDR made an allowance for the fact that SDRs would not be viewed as a desirable reserve. A limit was placed on the obligation to accept SDRs to meet the fear that with only a limited number of holders (participants in the SDR facility and other members and nonmembers approved by the Executive Board of the IMF), and no private holders, it could not be flexibly used in the way often required of reserves. The “designation” mechanism, by which the Fund staff would determine by means of a formula which creditor country was in the best position to provide its currency in return for SDRs, represented a further recognition of the probable unpopularity of the new instrument. Each participant designated to do so was obliged to provide convertible currency in exchange for SDRs, but only up to three times its net cumulative allocation. A participant would be designated to supply currency if it had a strong balance of payments and reserve position; and a designation plan would be drawn up at quarterly intervals.
Countries were expected to use SDRs “only to meet balance of payments needs,” and not to change their reserve composition (for instance by removing the SDRs, or swapping SDRs for dollars). In accordance with the French preference, the SDRs used in this way bore a resemblance to a credit, rather than to a reserve, in that they had to be partly repaid over a fixed time period (while if a country sells its gold reserves, or any other reserve asset, it is under no obligation to repurchase them). There was an obligation to make repayments—to “reconstitute” the SDR holding so as to maintain over a five-year period an average daily balance. The time horizon of five years for SDRs was in this way made compatible with the longest period permitted far the Fund’s stand-by assistance. The word used in the new Article, “reconstitution,” rather than repayment or repurchase, was chosen to make the reserve element as clear as possible and to camouflage the partial credit element implied in the instrument. The SDR was linked to gold, with a value identical to that of the U.S. dollar (until August 15, 1971), and bore an interest rate of 1.5 percent (again emphasizing the partial credit element: this was lower than prevailing interest rates, while gold reserves obviously carried no interest).
In the long run, however, the disappointing history of the SDR, and its failure to emerge as a genuine major international reserve unit, was fundamentally not a consequence of the interest rate limitations or of the reconstitution requirement, which were respectively modified and dropped. The problem of the SDR was twofold: First, the issue of the international currency was controlled through the voting requirements that had been an essential part of the difficult compromise about the new reserve unit. As a result of the 85 percent stipulation, a relatively small number of countries could block new issues, and thus obstruct the emergence of the SDR as a reserve currency. As a result, the innovative proposals made several times during the 1970s that the SDR could come to replace national reserve currencies (and particularly the dollar) never materialized (see below, pages 296–97). Second, a synthetic currency whose potential ownership was so excessively restricted could not really be expected to play a major role in the process of reserve-building through private sector credit operations. It was the private capital markets that created increasing levels of dollar reserves, and at very much faster rates in the early 1970s.
The activation of SDRs occurred in 1969 in the light of what was believed to be the tight reserve situation of that year. In addition to the momentary shortage, the long-term trend of declining reserves relative to world trade seemed clearly established (Figure 6-3). In 1969 the IMF explained the rationale for the SDR: “Since about 1964 … growth of reserves flattened markedly, the ratio of reserves to trade declined more rapidly, the transfer of reserves from deficit to surplus country ceased to act as a force tending to equalize reserve ratios, and there was increasing resort to international credit as a means of relieving the tightness of reserves.”57 The calculation of the appropriate size of the SDR allocation occurred on the basis of a historical comparison of the ratio of reserves to imports over the period 1954–68, with the conclusion that a $9.5 billion allocation between 1970 and 1972 would restore the rate of reserve growth to its historical trend.
Figure 6-3.World Reserves and Their Composition as a Share of World Imports
Source: International Monetary Fund.
In retrospect, the timing of 1969 appears highly ironic. Over the next two years the policies of the U.S. administration resulted in an explosion of dollar liquidity. The increase in reserves exceeded any historical growth path; and the primary task of the SDR in supplying adequate global liquidity had been made very evidently redundant. The SDR’s second function, as a contribution to increasing confidence in the convertibility of the U.S. dollar in the light of shrinking U.S. gold holdings, was at least visibly endangered by the large U.S. deficits.
The SDR as a result turned out to be something quite different from the instrument originally envisaged. It had lost a role and began to look for a new mission. It provided a useful instrument, but it was equally clearly not at the center of the international monetary system.58 In retrospect, as it was actually realized in the circumstances of the late 1960s, the SDR appeared as the last and most controversial of the gadgets devised to deal with the weakness of the U.S. payments position, rather than the beginning of a new approach to managing the international order. The long and intricate debate about the character of the SDR and the circumstances in which it might be issued turned out to be a rather large red herring. It was the SDR debate that prompted the observation that “they could not see the Bretton Woods for the Bretton trees.”59
Increased capital and money movements had made a “system” and provided incentives to join it, since this would give access to new resources, and would help to sustain a trade liberalization that was generally viewed as beneficial. But the existence of a system also generated greater vulnerability, to which the response was the call for higher reserve levels. Just as this discussion was advancing, and as it had produced an outcome, the system itself, with its increased flows, began to generate its own liquidity in an uncontrolled fashion, through the private markets. The new liquidity explosion (in which the SDR represented only a very small addition) destroyed the delicate balance on which had depended the international economy of the “golden age.” Previously, liquidity provision had helped to cushion shocks. Now, excessive liquidity turned out to make the shocks more intense, and their international transmission not harder but infinitely easier. This was a direct consequence of the failure of international coordination, and of the resulting adoption of mutually inconsistent national economic strategies.