Bretton Woods System

Introduction
The Bretton Woods system is commonly understood to refer to the international monetary regime that prevailed from the end of World War II until the early 1970s. Taking its name from the site of the 1944 conference that created the *International Monetary Fund (IMF) and *World Bank, the Bretton Woods system was history's first example of a fully negotiated monetary order intended to govern currency relations among sovereign states. In principle, the regime was designed to combine binding legal obligations with multilateral decision-making conducted through an international organization, the IMF, endowed with limited supranational authority. In practice the initial scheme, as well as its subsequent development and ultimate demise, were directly dependent on the preferences and policies of its most powerful member, the United States.

The Bretton Woods International Monetary System: A Historical Overview

After twenty years of floating exchange rates, there is now considerable interest, among those concerned over its perceived shortcomings, in an eventual return by the world to a fixed exchange rate regime. This interest has been enhanced by the apparent success of the European Monetary System (EMS) and the prospects for European monetary unification. The Bretton Woods system was the worlds most recent experiment with a fixed exchange rate regime.

Although it was originally designed as an adjustable peg, it evolved in its heyday into a de fact0 fixed exchange rate regime. That regime ended with the closing by President Richard Nixon of the gold window on 15 August 1971. Twenty years after that momentous decision, a retrospective look at the performance of the Bretton Woods system is timely.

This paper presents an overview of the Bretton Woods experience. I analyze the systems performance relative to earlier international monetary regimesas well as to the subsequent one-and also its origins, operation, problems, and demise. In the survey, I discuss issues deemed important during the life of Bretton Woods and some that speak to the concerns of the present. The survey is limited to the industrial countries-the G10 and especially the G7. I do not examine the role of the International Monetary Fund (IMF), the fundamental organization of Bretton Woods, in the economies and international economic relations of the developing nations.

Section 1.1 compares the macro performance of Bretton Woods with the preceding and subsequent monetary regimes. The descriptive statistics on research assistance has been provided by Bernhard Eschweiller and Johan Koenes. nine key macro variables point to one incontrovertible conclusion. Both nominal and real variables exhibited the most stable behavior in the past century under the Bretton Woods system, in its full convertibility phase, 1959-71.

While Bretton Woods was relatively stable, it was also very short lived. From the declaration of par values by thirty-two countries on 18 December 1946 to the closing of the gold window on 15 August 1971, it lasted twenty-five years. However, most analysts would agree that, until the Western European industrial countries made their currencies convertible on 27 December 1958, the system did not operate as intended. On this calculation, the regime lasted only twelve years. Alternatively, if we date its termination at the end of the Gold Pool and the start of the two-tier system on 15 March 1968, it was in full operation only nine years.

This raises questions about why Bretton Woods was statistically so stable and why it was so short lived.

(1) Was Bretton Woods successful in producing economic stability because it operated during a period of economic stability, or did the existence of the adjustable peg regime produce economic stability? Alternatively, was its statistical stability an illusion-belied by the presence of continual turmoil in the foreign exchange markets?

(2) Why did the system crumble after 1968 and end (so far) irrevocably in August 1971? It is the hope of the conference organizers that answers to these questions and many others can be provided by this and other papers to be presented here.

Section 1.2 surveys the origins of Bretton Woods: the perceived problems of the interwar period; the plans for a new international monetary order; and the steps leading to the adoption of the Articles of Agreement. Section 1.3 examines the preconvertibility period from 1946 to 1958, the problems in getting started exemplified by the dollar shortage and the weakness of the IMF, and the transition of the system to convertibility and the gold dollar standard.

Section 1.4 analyzes the heyday of Bretton Woods from 1959 to 1967 in the context of the gold dollar standard and its famous three problems: adjustment, liquidity, and confidence. I review both the problems and the many proposals for monetary reform.

Section 1.5 considers the emergence of a de facto dollar standard in 1968 and its collapse in the face of U.S.-induced inflation.

Finally, section 1.6 summarizes the main points of the paper, discusses some lessons learned from the Bretton Woods experience for the design of a fixed exchange rate regime, and raises questions answered by the other papers in the conference volume.

Design of the Bretton Woods system

The conference that gave birth to the system, held in the American resort village of Bretton Woods, New Hampshire, was the culmination of some two and a half years of planning for postwar monetary reconstruction by the Treasuries of the United Kingdom and the United States. Although attended by all forty four allied nations, plus one neutral government (Argentina), conference discussion was dominated by two rival plans developed, respectively, by Harry Dexter White of the U.S. Treasury and by *John Maynard Keynes of Britain.

The compromise that ultimately emerged was much closer to White's plan than to that of Keynes, reflecting the overwhelming *power of the United States as World War II was drawing to a close. Although, at the time, gaps between the White and Keynes plans seemed enormous especially with respect to the issue of future access to international *liquidity - in retrospect it is their similarities rather than their differences that appear most striking. In fact, there was much common ground among all the participating governments at Bretton Woods. All agreed that the monetary chaos of the interwar period had yielded several valuable lessons. All were determined to avoid repeating what they perceived to be the errors of the past. Their consensus of judgment was reflected directly in the Articles of Agreement of the IMF. Four points in particular stand out. First, negotiators generally agreed that as far as they were concerned, the interwar period had conclusively demonstrated the fundamental disadvantages of Unrestrained flexibility of *exchange rates.

The floating rates of the 1930s were seen as having discouraged trade and investment and to have encouraged destabilizing speculation and competitive depreciations. Yet in an era of more activist economic policy, governments were at the same time reluctant to return to permanently fixed rates on the model of the classical *gold standard of the nineteenth century. Policy- makers understandably wished to retain the right to revise currency values on occasion as circumstances warranted. Hence a compromise was sought between the polar alternatives of either freely floating or irrevocably fixed rates some arrangement that might gain the advantages of both without suffering the disadvantages of either. What emerged was the 'pegged rate' or 'adjustable peg' currency regime, also known as the par value system. Members were obligated to declare a par value (a 'peg') for their national money and to intervene in currency markets to limit exchange rate fluctuations within maximum margins (a 'band') one per cent above or below parity; but they also retained the right, whenever necessary and in accordance with agreed procedures, to alter their par value to correct a 'fundamental disequilibrium' in their *balance of payments. Regrettably the notion of fundamental disequilibrium, though key to the operation of the par value system, was never spelled out in any detail a notorious omission that would eventually come back to haunt the regime in later years. Second, all governments generally agreed that if exchange rates were not to float freely, states would also require assurance of an adequate supply of monetary reserves. Negotiators did not think it necessary to alter in any fundamental way the *gold exchange standard that had been inherited from the interwar years. International liquidity would still consist primarily of national stocks of gold or currencies convertible, directly or indirectly, into gold ('gold

exchange'). The United States, in particular, was loth to alter either the central role of the dollar or the value of its gold reserves, which at the time amounted to three quarters of all central bank gold in the world. Negotiators, did concur, however, on the desirability of some supplementary source of liquidity for deficit countries. The big question was whether that source should, as proposed by Keynes, be akin to a world central bank able to create new reserves at will (which Keynes thought might be called *bancor); or a more limited borrowing mechanism, as preferred by White. What emerged largely reflected U.S. preferences: a system of subscriptions and quotas embedded in the IMF, which itself was to be no more than a fixed pool of national currencies and gold subscribed by each country. Members were assigned quotas, roughly reflecting each state's relative economic importance, and were obligated to pay into the Fund a subscription of equal amount.

The subscription was to be paid 25 per cent in gold or currency convertible into gold (effectively the dollar, which was the only currency then still directly gold convertible for central banks) and 75 per cent in the member's own money. Each member was then entitled, when short of reserves, to borrow needed foreign currency in amounts determined by the size of its quota. A third point on which all governments agreed was that it was necessary to avoid recurrence of the kind of economic warfare that had characterized the decade of the 1930s. Some binding framework of rules was needed to ensure that states would remove existing *exchange controls limiting currency convertibility and return to a system of free multilateral payments. Hence members were in principle forbidden to engage in discriminatory currency practices or exchange regulation, with only two practical exceptions. First, convertibility obligations were extended to current international transactions only. Governments were to refrain from

regulating purchases and sales of currency for trade in goods or services. But they were not obligated to refrain from regulation of capital-account transactions. Indeed, they were formally encouraged to make use of *capital controls to maintain external balance in the face of potentially destabilizing 'hot money' flows. Second, convertibility obligations could be deferred if a member so chose during a postwar 'transitional period.' Members deferring their convertibility obligations were known as Article XIV countries; members accepting them had so-called Article VIII status. One of the responsibilities assigned to the IMF was to oversee this legal code governing currency convertibility.

Finally, negotiators agreed that there was a need for an institutional forum for international cooperation on monetary matters. Currency troubles in the interwar years, it was felt, had been greatly exacerbated by the absence of any established procedure or machinery for intergovernmental consultation. In the postwar era, the Fund itself would provide such a forum - in fact, an achievement of truly historic proportions. Never before had international monetary cooperation been attempted on a permanent institutional basis. Even more pathbreaking was the decision to allocate voting rights among governments not on a one-state, one-vote basis but rather in proportion to quotas. With one-third of all IMF quotas at the outset, the United States assured itself an effective veto over future decision-making.

The Performance of Bretton Woods in Comparison to Alternative

Monetary Regimes The architects of the Bretton Woods system wanted a set of monetary arrangements that would combine the advantage of the classical gold standard (i.e., exchange rate stability) with the advantage of floating rates (i.e., independence to pursue national full employment policies). They sought to avoid the defects of floating rates (destabilizing speculation and competitive beggarthy- neighbor devaluations) and the defects of the fixed exchange rate gold standard (subordination of national monetary policies to the dictates of external balance and subjection of the economy to the international transmission of the business cycle). As a consequence, they set up an adjustable peg system of fixed parities that could be changed only in the event of a fundamental disequilibrium.

The architects derived their views of an ideal international monetary arrangement from their perception of the performance of the pre-World War I classical gold standard and of the sequence of floating rates and gold exchange standard that characterized the interwar period. As background to the historical survey of Bretton Woods, I compare descriptive evidence on the macro performance of the international monetary regime of Bretton Woods with that on the performance of preceding and subsequent regimes.

The Bretton Woods period (1946-70) is divided into two subperiods: the preconvertible phase (1946-58) and the convertible phase (1959-70).* The comparison also relates to the theoretical issues raised by the perennial debate over fixed versus flexible exchange rates. According to the traditional view, adherence to a (commodity-based) fixed exchange rate regime, such as the gold standard, ensured long-run price stability for the world as a whole because the fixed price of gold provided a nominal anchor to the world money supply. By pegging their currencies to gold, individual nations fixed their price levels to that of the world. The disadvantage of fixed rates is that individual nations were exposed to both monetary and real shocks transmitted from the rest of the world via the balance of payments and other channels of transmission (Bordo and Schwartz 1989). Also, the common world price level under the gold standard exhibited secular periods of deflation and inflation reflecting shocks to the demand for and supply of gold (Bordo 1981; Rockoff, 1984).

One important caveat is that the historical regimes presented here do not represent clear examples of fixed and floating rate regimes. The interwar period is composed of three regimes: general floating from 1919 to 1925, the gold exchange standard from 1926 to 193 1, and a managed float to 1939.3 The Bretton Woods regime cannot be characterized

as a fixed exchange rate regime throughout its history: the preconvertibility period was close to the adjustable peg envisioned by its architects; the convertible period was close to a de facto fixed dollar ~t a n d a r dF.~in ally, although the period since 1973 has been characterized as a floating exchange rate regime, at various times it has experienced varying degrees of management.

Table 1 . 1 presents descriptive statistics on nine macro variables for each country, the data for each variable converted to a continuous annual series from 1880 to 1989. The nine variables are the rate of inflation, real per capita growth, money growth, short- and long-term nominal interest rates, shortand long-term real interest rates, and the absolute rates of change of nominal and real exchange rates. The definition of the variable used (e.g., M1 vs. M2) was dictated by the availability of data over the entire period. For each variable and each country, I present two summary statistics: the mean and the standard deviation. For inflation, I also show (in parentheses) the standard deviation of the forecast error based on a univariate regression. For all the countries taken as a group, I show two summary statistics: the grand mean and a simple measure of convergence measured asthe mean of the absolute differences between each countrys summary statistic and the grand means of the group of countries. Comment the statistical results for each variable.

The Origins of Bretton Woods

The planning that led to the creation of a new international monetary order at Bretton Woods was predicated on the belief that the mistakes of the interwar period were to be avoided. These mistakes included, first, wildly fluctuating exchange rates after World War I and the collapse of the short-lived gold exchange standard; thereafter, the international transmission of deflation and the resort to beggar-thy-neighbor devaluations; and, finally, trade and exchange restrictions and bilateralism.

The goal was the negotiation of an international monetary constitution based on stable exchange rates, national full employment policies, and cooperation. Three issues dominated the perception of the interwar experience: the flaws of the gold exchange standard; the case against floating exchange rates; and bilateralism. I survey each in turn.

The Gold Exchange Standard

The gold exchange standard began de facto in 1925 after Britain returned to the gold standard at the prewar parity, followed in the next two years by most countries. The case for gold economy was made at the Genoa conference in 1922. The world could avoid a massive gold shortage, produced in large part by wartime inflation, by two measures: first, by nations adopting a gold bullion standard and demonetizing gold as a national currency; second, by using foreign exchange (primarily pounds and dollars) as international reserves and stockpiling monetary gold in the key reserve centers of London and New York. The gold exchange standard lasted only six years. It ended with Britains suspension of gold convertibility in September 193 1. The subsequent literature focused on the three problems that later plagued Bretton Woods: adjustment, liquidity, and confidence (Johnson 1972a), although Keynes and White, the architects of Bretton Woods, paid attention only to the first two(Dam 1982, chap. 3; Eichengreen 1990b). The adjustment problem involved two issues: first, asymmetric adjustment between deficit (Britain) and surplus countries (the United States), which led to a deflationary bias, and, second, the failure by all countries to follow the rules of the game. The United States and France (with an undervalued gold parity after 1926, the Bank of France being prevented by law after 1928 from holding reserves other than gold against its domestic note issue) together absorbed 53% of the worlds monetary gold reserves by 1924 (Eichengreen 1990a, table 10.1).

As surplus countries, they sterilized gold inflows and avoided expansionary monetary policy and inflation. At the same time, thesteady gold drain from Britain (with an overvalued parity)I4 and other deficit countries ultimately caused them to contract their

money supplies. Sterilization by the surplus countries only worsened the gold drain from the deficit countries. In addition, according to Nurkse (1944) and Eichengreen (1990b), in an attempt to shield the domestic economy from foreign disturbances, virtually every central bank actively or passively followed a policy of offsetting changes in international reserves by changes in domestic credit, hence breaking the rules of the game. Such policies at best slowed the adjustment of relative national prices and incomes required to restore balance of payments equilibrium under the gold standard. The interwar liquidity problem involved gold supplies (at the prevailing set of gold parities) inadequate to finance the growth of world output and trade and to serve as gold cover to back national currencies. To economize on gold, peripheral countries used the key currencies as international reserves, but, as the ratio of their holdings of the key currency (especially the pound) rose relative to the center countrys monetary gold stocks, the peripheral countries reduced their foreign exchange holdings, fearing a convertibility crisis (Triffin, 1960). The interwar conjdence problem involved two components: a shift of currency holdings between the two key centers, London and New York (and to a lesser extent, at the end of the period, a shift to Paris), and a shift between the key currencies and gold. Shifts of foreign currency balances from weak to strong reserve centers weakened the system because they increased the likelihood of a confidence crisis in the weak center, which, if it materialized, would then put pressure on the other center (Nurkse 1944). A shift between the key currencies and gold occurred when, fearing the reserve centers inability to convert their outstanding liabilities into gold at the fixed parity, foreign holders of key currency balances staged a run on the bank, precipitating a suspension of convertibility, as happened to Britain in September 1931