III. The removal, so far as possible, of all economic barriers and the establishment of an equality of trade conditions among all the nations consenting to the peace and associating themselves for its maintenance.Woodrow Wilson, Fourteen Points (1918)
The International Monetary Fund was forged from failure.
When the heads of government of the great powers met in Paris at the end of 1918, they had before them a blueprint for restoring prosperity and world peace, in the form of U.S. President Woodrow Wilson’s Fourteen Points. Six months later, they agreed on the terms of what would become known as the Treaty of Versailles, but key parts of the blueprint had been cast aside. Within a decade, prosperity was lost. In another decade, peace was gone as well. The most famous failure was Wilson’s inability to convince the U.S. Senate to confirm the country’s membership in the League of Nations. The most disastrous, however, was arguably the failure to lay the groundwork for economic cooperation among the world’s great trading nations. Whether U.S. membership in the League would have slowed the slide toward war in the 1930s is debatable. The effect of the autarkic policies of the 1920s on the collapse of trade and output in the 1930s, however, is well established (Crucini and Kahn, 1996; Irwin, 1998).
When delegations from 44 countries met at Bretton Woods, New Hampshire in July 1944 to establish institutions to govern international economic relations in the aftermath of the Second World War, avoiding a repeat of the failings of the Paris peace conference was very much on their minds. Creation of an International Bank for Reconstruction and Development would help restore economic activity, while creation of an International Monetary Fund would help restore currency convertibility and multilateral trade. Removing the barriers to trade, as envisaged by Wilson a quarter century earlier, was not enough. A more active and institutionalized cooperation was now understood to be needed.
The failure of Paris was only the first of a series of historical events and ideological evolutions to influence the design and work of the IMF and the postwar international monetary system. This paper surveys some of the key twentieth-century events and shifts in economic theory that had the greatest influence on the Fund, in order to draw some general conclusions about the force of history on the international monetary system that now prevails.
I. Ten Events
The first three key events—the Paris peace conference, the Great Depression, and the Second World War—made the creation of a multilateral financial institution possible and largely determined the form it would take. Subsequent events caused the IMF to alter its practices in various ways to stay relevant in a changing world.
A. The Paris Peace Conference
Economics was far from being a high priority at the Paris peace conference. The borders of Europe had to be redrawn one by one, and that task alone took up most of the six months of high-level meetings. A means had to be found to pay the costs of the war and the costs of rebuilding, and solving that problem was about all the economics that any of the leaders had the patience for. They created the League of Nations, but its economic functions were poorly defined and never did gel into an effective role.1 They created the International Labour Organization, but its role was specialized and limited.
The conference’s neglect of economics did not result from a failure to understand the importance of international trade for prosperity and thus for maintaining the peace. As the quotation at the head of this paper shows, Woodrow Wilson had made this relationship clear in his “fourteen points” speech to the U.S. Congress in January 1918. Instead, the neglect of economics occurred largely because the limitations of the invisible hand were not well understood. For a generation or more, the international gold standard had provided a measure of stability with little need for overt cooperation. The challenge seemed to be simply to avoid imposing barriers to trade or otherwise interfering with markets.
In the economic turmoil that followed the war, that passive approach was not nearly enough. Some countries remained on the gold standard, but others did not. Without clear guidance or any institutional check on behavior, competitive devaluations and punitive tariffs became a common temptation for a quick fix to economic ills. Margaret MacMillan (2001) is surely right in arguing that the Versailles treaty cannot be held solely responsible for these and other ills of the twentieth century, but neither can it be absolved from blame.
What does this experience have to do with the IMF? A quarter century afterwards, it was very much on the minds of those who were drawing up the designs for the new institution. In the view of John Maynard Keynes (the head of the British delegation), the “contractionist pressure on world trade” brought on by the “special protective expedients which were developed between the two wars” resulted in large measure from futile efforts “to protect an unbalanced position of a country’s overseas payments.” Creation of an “international clearing union” would obviate the need for such “forced and undesired dodges.”2 Without the clearing union (which eventually morphed into the IMF), the expected persistent creditor position of the United States would depress world economic growth and drive the world back into protectionist policies, regardless of how quickly or well production and trade could be reconstructed after the war.
Harry Dexter White, the chief drafter of the IMF charter for the U.S. delegation, was equally impressed by the need to avoid the passive errors of Versailles. His initial plan noted that during “the last twenty years” (that is, throughout the interwar period), countries had often imposed protectionist policies because they lacked adequate gold reserves, and it warned that the same problems would arise and would constitute a major barrier to the growth of trade after the war. An international monetary fund would enable countries in that position to economize on their gold reserves and thus avoid recourse to trade barriers, payments barriers, and bilateral clearing schemes.3 As early as 1935, when France and Great Britain were contemplating currency devaluations that were aimed at improving their competitive positions but that threatened to spark a vicious cycle of retaliatory actions, White had argued that the U.S. Treasury should intervene by encouraging an international agreement to stabilize exchange rates (Boughton, 2002). That led to the Tripartite Agreement of 1936 and set the stage for more comprehensive and institutionalized agreements later on. When the Articles of Agreement for the IMF were adopted at Bretton Woods in 1944, they specified that one purpose of the institution was “to avoid competitive exchange depreciation.”
It is important to note that for both Keynes and White, the motivating principle for creating the IMF was to engender postwar economic growth by establishing an institution that would prevent a relapse into autarky and protectionism, not just to avoid a recurrence of the Depression. The impetus was less the Depression than the necessity of rebuilding and engendering economic growth after the war.
B. The Great Depression
Although the Great Depression may not have been “defining moment” for the international monetary system (as argued by Bordo and Eichengreen, 1998), it was certainly an important influence on the initial design of the IMF. The Depression amplified the negative consequences of Versailles, as an implosion of international trade interacted with domestic policy errors to deflate both output and prices around the world. It severely tested the confidence of analysts and voters in the efficacy of free markets and strengthened belief in an activist role for the public sector in economic life. It thus became easier and more natural to start discussions on a postwar framework from the assumption that an intergovernmental agency with substantive powers would be beneficial and even essential for the international financial system.
The combined effects of Versailles (the absence of a stabilizing system in international finance) and the Depression were an important influence on the mandate of the IMF adopted at Bretton Woods in 1944. Article I of the Articles of Agreement, which sets out the purposes of the Fund, includes the objective of using IMF lending to provide member countries “with opportunity to correct maladjustments in their balance of payments without resorting to measures destructive of national or international prosperity.” Article IV set out a system for achieving that purpose by establishing a system of fixed but adjustable exchange rates through agreements to be reached under the auspices of the Fund. U.S. Treasury staff made the case for such a system by evoking the specter of what had occurred throughout the interwar period: “Long before the war, the necessary monetary and financial basis for international prosperity had been weakened by competitive currency depreciation, by exchange restriction, by multiple currency devices,” and the like.4 The new institution would obviate the need for such unilateral and destructive actions.
C. World War II
The third major historical influence on the IMF was the Second World War, which provided both the impetus and the context for reforming the international system. When the United States entered the war in response to the bombing of Pearl Harbor in December 1941, Treasury Secretary Henry Morgenthau, Jr., put Harry White in charge of international economic and financial policy and asked him to come up with a plan for remaking the system once the war was over. As it happened, White had already sketched out a rough plan for an international stabilization fund, and he was able to produce a first draft within a couple of months. On the other side of the Atlantic, Keynes was developing a plan for an international clearing union to be run jointly by Britain and the United States as “founder-States.”5 Though less overtly multilateral than White’s scheme, and based on the British overdraft system rather than on White’s rather complicated proposal for currency swaps (Boughton, 2002 and Boughton 2003b), Keynes’s clearing union was similar in its essence to White’s stabilization fund. Over the next two years of discussion and negotiation, the two plans would meld into a draft for the IMF charter.
The IMF was created in the midst of the war, at the United Nations Monetary and Financial Conference, which convened 44 country delegations at Bretton Woods in July 1944. Keynes had tried to limit the involvement of countries other than Britain and the United States, fearing that a “great monkey house” would result if all of the wartime allies were invited. White, however, insisted on a multilateral conference, partly because he seems to have sensed that the project would otherwise fail and partly because he doubtless wanted to neutralize the force of Keynes’s intellect and personality.
The importance of Bretton Woods as a wartime event was that it took advantage of a window of opportunity to create a multilateral financial system. Both before and after the war, there was too much suspicion and national self-interest for such a sweeping agreement to be possible. Even in 1945, when the U.S. Congress and the U.K. Parliament were to ratify the Articles of Agreement, passage was far from easy (Gardner, 1956). Again, White invoked the specter of Versailles. Asked, in a House of Representatives hearing, what would happen if Congress refused to ratify the agreement, White replied, “I think history will look back and indict those who fail to vote the approval of the Bretton Woods proposals in the same way that we now look back and indict certain groups in 1921 who prevented our adherence to an international organization designed for the purpose of preventing wars.”6 Such arguments carried the day in 1945. Within three years, however, when negotiators tried to complete the system by creating an International Trade Organization, the multilateralists were outmanned, and the proposal failed.
The other major influence of the war on the IMF was that it left the United States in virtual control of the world economy. With Britain heavily dependent on American largesse, Keynes had few cards to play in his efforts to shape the postwar system to his country’s advantage. Of the other major allies, France was equally powerless and the Soviet Union was politically isolated and intellectually detached. As a consequence, the financial structure of the IMF would be based on the U.S. dollar, rather than on an international currency of its own making. Its lending power would be limited in size and scope, and the Fund would lack most of the powers of a central bank. Its headquarters would be neither in London nor even in New York, but in Washington where the U.S. Treasury could exert a strong gravitational pull. For the next three decades, the IMF would be essentially a dollar-centric institution, with the United States providing most of its loanable resources and effectively controlling most of its lending decisions.
D. The Rise of Multiple Economic Centers
Once the war was over and the world economy—and world trade—began to recover, U.S. economic hegemony was gradually eroded. The first to rise from the ashes was western Europe. Through a combination of national drive, international support—from the U.S. Marshall Plan, the World Bank, and eventually the IMF—, and a home-grown multilateralism in the form of the Common Market and the European Payments Union, much of Europe was growing rapidly and increasingly open to multilateral trade and currency exchange by the late 1950s. The Federal Republic of Germany joined the IMF in 1952 and quickly became one of the world’s leading economies. Next came Asia. Japan also joined the Fund in 1952, and by the 1960s it was on its way to joining the United States and Germany on the top rung of the economic ladder. Then the 1970s saw the rise of economic power in Saudi Arabia and other oil-exporting countries of the Middle East. In thirty years, the U.S. share of world exports had fallen from 22 percent to 12, while its share of official international reserves dropped even more dramatically, from 54 percent in 1948 to 12 percent in 1978.
As the balance of economic and financial power became more widely dispersed, more and more currencies became fully convertible for current account and even capital transactions. Trading partners grew at different rates and with different mixes of financial policies. Pressures on fixed exchange rates and on the limited supply of gold and U.S. dollars became increasingly frequent and more severe. The IMF responded in 1969 by amending its Articles and creating Special Drawing Rights (SDRs) as a supplement to existing reserve assets, but that action was too limited to deal with the underlying problem of differential pressures. As a result, even before the first oil shock in 1973, the original Bretton Woods system of fixed but adjustable exchange rates had become unviable. A second amendment, adopted in 1978, acknowledged that exchange rates among key currencies were likely to float or at least be allowed to adjust more frequently than the old system could have handled.
E. The Cold War
Harry White had worked hard in 1944 to persuade the Soviet Union to join the IMF, in the belief that economic cooperation between the Soviet Union and the United States would be the key to postwar peace and prosperity. The Soviet delegation to Bretton Woods did sign the Articles ad referendum, but Joseph Stalin eventually refused to ratify the agreement, apparently because he feared (not without justification) that Fund policies would be largely controlled by the west (James and James, 1994). When that tension segued into the Cold War, White’s vision of universal membership was dashed. Poland withdrew from membership in 1950. Four years later, Czechoslovakia was forced to withdraw. Shortly after taking power in 1959, Fidel Castro pulled Cuba out. For more than three decades after Mao Zedong took control of China, the U.S. government blocked efforts by the People’s Republic to be seated as China’s representative on the IMF Executive Board. Most other countries in the Soviet or Chinese spheres of influence simply did not join. Not until the 1980s would the trend be reversed with the seating of China and renewed membership for Poland (Boughton, 2001a, Chapter 19).
The obvious effect of the Cold War on the IMF was this limitation on membership. In the terminology of the period, it included the first world and much of the third, but the second was missing from the table. The IMF became largely a capitalist club that helped stabilize market-oriented economies.7 The more subtle and difficult question concerns the effect on the staff and its analytical work. The bulk of IMF analysis has always been mainstream and centrist, viewed from the perspective of the dominant strain of Anglo-Saxon economics. The leading universities of North America, the United Kingdom, and Australia have been the main training grounds for much of its professional staff. Martha Finnemore, a political scientist who has studied a number of large organizations, has even claimed that the Pentagon displays more intellectual diversity than the IMF8.
Would this centrist dominance have been weaker, with a broader range of views on economic policy being represented, if the Fund’s membership had been universal from the outset? That seems unlikely. The shift to universal membership in the 1990s and the corresponding geographic broadening of the staff9—in Finnemore’s terminology, an increase in “passport diversity”—have so far had little analytical impact. Moreover, the influence of Latin American economic thought—exemplified by the dependencia theories of Raúl Prebisch (1971) and others at the UN Economic Commission on Latin America10—was never strong in the IMF despite the presence of large numbers of economists from the region on the Fund staff from the outset. Analytical diversity and internal dissent have been more prominent in the World Bank (with the same membership) than in the IMF, albeit less so than in the nearly universal United Nations secretariat. The influence of mainstream western thinking at the IMF—an influence that the staff itself would regard with some justification as reflecting best practices in the economics profession—is a more deeply seated phenomenon than can be explained by Cold War politics.
F. African Independence
Only three of the IMF’s 40 original members were in Africa: Egypt, Ethiopia, and South Africa. Of those, one was more closely affined to the Middle East, and one was minority controlled and more culturally linked to Europe. Most of the continent was still under colonial rule. That situation began to evolve in 1957, when the newly independent countries of Ghana and Sudan became members. Applications then flooded in, and by 1969 the IMF had 44 members (out of 115) from Africa. By 1990, all of Africa’s 53 countries were in the IMF.11 They comprised nearly a third of the member countries, though their average small size and mostly low incomes meant that they controlled less than 9 percent of the voting power and held only three of the 22 seats on the Executive Board.12
The emergence of Africa as a continent of independent nations had a major effect on the size and diversity of the IMF, and it required a substantial intensification of the Fund’s involvement with and oversight of its borrowers. Most of these countries, especially in sub-Saharan Africa, had and still have very low per capita incomes and are among the least economically developed countries in the world. Their economic problems tend to be structural even more than macroeconomic; rooted in the need for improvements in education, health, infrastructure, and governance rather than finance; and more deeply ingrained and persistent than in other regions. When the Fund began providing financial assistance to large numbers of low-income countries in the 1970s, it had to find ways to subsidize its lending, coordinate its assistance with other official agencies, and develop more extensive and structural policy-reform conditions on its lending. In addition, the Fund sharply increased and broadened its provision of technical assistance to member countries, thereby expanding its work further beyond its original boundaries.
Lending to low-income countries also raised the riskiness of the IMF’s portfolio of sovereign claims. By the mid-1980s, several African countries had fallen into protracted arrears on their borrowings from the Fund, which forced the institution to further reexamine its conditionality as well as its finances. Several countries with protracted arrears—mostly in Africa—have been subject to “remedial” measures leading up to suspension of voting rights. The IMF now lends to low-income countries primarily through a separately funded subsidized window, the Poverty Reduction and Growth Facility, and it coordinates that assistance with the World Bank. To qualify for those loans, countries must develop their own strategies for generating economic growth and reducing poverty. The IMF still emphasizes the need for countries to maintain sound macroeconomic policies, but that traditional focus is only the starting point for most of its work in Africa.
G. The Vietnam War
The intensification of U.S. involvement in the Vietnam War in the 1960s and early 1970s would not by itself have had substantial effects on the IMF, other than the direct effect on Vietnam’s membership. When the government of South Vietnam was about to fall in April 1975, its officials tried desperately to borrow as much as they could from the IMF. The Fund refused to go along, and within a few months it recognized the Socialist Republic of Viet Nam as the successor government (Boughton, 2001a, pp. 766–67). The larger effect, however, was on the U.S. economy and its external payments position. In combination with a sizeable increase in domestic spending on President Lyndon Johnson’s Great Society programs, the rise in external military spending gradually worsened the overvaluation of the U.S. dollar under the Bretton Woods system of fixed exchange rates. In a series of spasms, the system dissolved between 1968 and 1973. With the dollar no longer convertible into gold, the precious metal could no longer serve a central or even a useful function in the international monetary system. The Vietnam War was by no means the sole culprit in this decline, but its catalytic role was certainly substantial (James, 1996, Chapter 8).
H. Globalization of Financial Markets
Private-sector financial flows were of limited scope and importance when the IMF was founded. Trade flows were financed largely by trade credits, and cross-border portfolio flows were considered by most economists to be as much a potential destabilizing nuisance as a potential source of investment capital. Keynes and White therefore agreed that the IMF should be given the power to restrict capital flows in situations where they seemed to be destabilizing. Article VI of the IMF charter prohibited member countries from borrowing from the Fund “to meet a large or sustained outflow of capital,” and it empowered the IMF to “request a member to exercise controls to prevent such use” and to declare the member ineligible to use the Fund’s resources if it failed to comply. More generally, it recognized countries’ right to impose capital controls as long as the controls did not restrict payments for transactions on the current account.
The range and importance of capital flows began to increase in the 1950s as European countries gradually reestablished convertibility. The first big increase, however, came in the 1970s, with the emergence of the Eurodollar and other offshore financial markets. It was driven further by the accumulation of “petrodollars” by oil exporting countries in the 1970s and the recycling of those assets to oil-importing sovereign borrowers through large international banks. By the 1990s, cross-border flows had become an essential source of finance for both industrial and emerging-market economies around the world, and the structure of international financial markets had become so complex that their effective size could no longer be measured, much less controlled.
Largely in an implicit recognition of these developments, the IMF has never invoked the provisions of Article VI enabling it to encourage the imposition of capital controls. Nor has the prohibition on lending to finance a large or sustained capital outflow ever prevented the Fund from acting, simply because it can always be argued that an unchecked capital outflow will eventually cause problems for the current account. That justification was first made in 1956, when the United Kingdom borrowed to stop a speculative attack on the pound sterling in the wake of the Suez crisis (Boughton, 2001b), and it has been taken for granted ever since.
A second effect of financial globalization was that IMF financing became quantitatively marginalized for many potential borrowers. In the early days of the IMF, countries facing a financing gap in their balance of payments could often close it solely by borrowing from the Fund. By the 1980s, the object was more often to “catalyze” other capital inflows by borrowing relatively small amounts from the Fund in support of an agreed package of policy reforms and thereby hoping to convince other creditors that the country was a good prospect. What mattered was not so much the quantity of money as the quality of the reforms. Globalization thus fundamentally altered the relationship between the IMF and its borrowing members and between the IMF and other official and private creditors.
A third effect was to weaken the “credit union” character of the IMF as a membership institution. The original idea was that most countries would probably undergo periods as creditors and other periods as debtors. In the 1950s and 1960s, most of the large industrial countries fit that description. Of the seven largest economies, only Germany and the United States consistently maintained creditor positions in the Fund. By the 1980s, however, the more advanced economies all were able to finance their external payments with private flows, and the membership of the IMF became divided into persistent creditor and debtor groups. The presumed commonality of interests among members was correspondingly diminished.
I. The International Debt Crisis
In August 1982, a two-year gradual worsening of conditions in international debt markets suddenly accelerated and precipitated a major economic and financial crisis. A scattering of countries including Hungary, Morocco, Poland, and Yugoslavia had already seen their bank creditors turn their backs in 1981 and the first half of 1982. When the banks suddenly pulled out of Mexico, the crisis took on systemic proportions. Within a few months, Argentina, Brazil, and Chile were also in trouble, and the crisis was continuing to spread. Not until 1990, when world interest rates were settling down and the bank debts of the most heavily indebted developing countries were being replaced by Brady Bonds, would it be possible to declare the crisis over (Boughton, 2001a, Part II).
The debt crisis had a transforming impact on the IMF, catapulting it into the role of international crisis manager. Previous international crises—Suez in 1956, the breakdown of the official gold market in 1968, the oil shocks of the 1970s—had intensified the demand for IMF lending without fundamentally changing the way the IMF worked (Boughton, 2000). The 1982 crisis was different because the range and diversity of creditors involved made it unlikely that it could be resolved without the active involvement of an outside agent. The Fund’s Managing Director, Jacques de Larosière, intervened personally by refusing to approve stand-by arrangements for the crisis-hit countries until he received written assurances from bank creditors that they would share the burden by increasing their lending exposure. This “concerted lending” tactic was the first instance of what later became known as “private sector involvement” in debt workout procedures. Over time, the specific tactics changed in response to evolving circumstances, but the role of the IMF as the central agency for coordinating the resolution of financial crises remained. For better or worse, the Mexican peso crisis of 1994-95, the Asian crises of 1997, and those that hit Russia, Brazil, Argentina, and Turkey in the next few years all brought the IMF to the forefront of efforts to coordinate temporary official financing, reform policies in the affected countries, and attempt to restore confidence and commitment on the part of creditors and investors.
J. Collapse of Communism
The fall of the Berlin wall in 1989 and the dissolution of the Soviet Union in 1991 enabled the IMF at last to become a (nearly) universal institution. In three years, membership increased from 152 countries to 172, the most rapid increase since the influx of African members in the 1960s. Many of the new members needed to borrow from the Fund, and all of them needed technical assistance and regular consultations. Consequently, the size of the IMF staff increased by nearly 30 percent in six years, with 15 new countries represented. The Executive Board expanded from 22 seats to 24 to accommodate Directors from Russia and Switzerland, and some existing Directors saw their constituencies expand by several countries. As discussed above, this development had little impact on the philosophical underpinnings of the Fund’s work. It did, however, broaden the range of issues with which the staff had to struggle. How could formerly centrally planned economies best be transformed and integrated into the world market economy? What structural reforms were needed, and in what sequence? How could price levels be stabilized when individual prices were still so far out of equilibrium and large excess money balances were still outstanding? How important for stabilization was the independence of the central bank from government control? For the Fund to stay reasonably within its mandate of stabilizing economies and strengthening macroeconomic policies while meeting the genuine needs of its expanding membership required a balancing act that became harder and harder to sustain.
II. Ten Ideas
While these events were shaping the IMF and in some cases forcing it to adapt to changing circumstances, economic theories were also evolving. Events and ideas often overlapped in their effects on the IMF and the international monetary system.
From the outset, three economic concepts have formed the bedrock of thinking at the IMF and have been the basis for much of the Fund’s operations: Keynesian macroeconomics, the monetary approach to the balance of payments, and the open-economy macro model. Two of Milton Friedman’s great ideas from the 1950s—monetarism and the case for floating exchange rates—were impossible to ignore and had some influence on the IMF as well. More recently, several developments shifted the economics profession and the Fund away from a Keynesian fixation on demand management as a means of stabilizing and strengthening national economies.
A. Keynesian Macroeconomics
The IMF was conceived essentially as a Keynesian institution. This linkage should not be surprising, since Keynes was one of its founding fathers and the other (White) was a New Deal economist who had championed the use of countercyclical monetary and fiscal policy as early as 1932 (Laidler and Sandilands, 2002). The U.S. Treasury’s case for creating the Fund stressed that the goal was to use and coordinate macroeconomic policies to prevent recessions and unemployment. “Only through international cooperation,” they wrote, “will it be possible for countries successfully to apply measures directed toward attaining and maintaining a high level of employment and real income which must be the primary objective of economic policy.”13 These objectives were accordingly included in Article I, along with world economic growth (“development of the productive resources of all members”) and avoidance of contractionary policies (“measures destructive of national or international prosperity”).
Some critics of IMF policies have argued that the Fund has drifted away from Keynesian principles by seeming to emphasize fiscal and monetary discipline over growth. Joseph Stiglitz (2002, p. 38) put this argument starkly, writing that the IMF “has taken on the pre-Keynesian position of fiscal austerity in the face of a downturn, doling out funds only if the borrowing country conforms to the IMF’s views about appropriate economic policy, which almost always entail contractionary policies leading to recessions or worse.”
This argument is based on a fundamental misconception of both Keynesian macroeconomics and IMF policy advice (Rogoff, 2003). Countries that are unable to finance their external payments position on affordable terms, regardless of whether the initial source of the difficulty was a fiscal excess, an adverse terms of trade shock, or other developments, have to restore balance if they are to maintain full employment and growth. Keynes himself acknowledged in his General Theory (1936, p. 332) that the early stages of Roosevelt’s New Deal, involving “curtailment of current output” through a reduction in unwanted inventories, were “a phase which had to be endured. … Only when it had been completed was the way prepared for substantial recovery.”
The IMF, or any institution acting in real time to solve economic crises, often gets the required extent of adjustment wrong, and a case could be made that the Fund is biased on the side of caution (IEO, 2003). But that is different from asserting that it has the basic idea wrong.
B. The Monetary Approach to the Balance of Payments
A longstanding building block of IMF policy advice is the version of the monetary approach to the balance of payments developed by Jacques Polak in the 1950s, when he was the Deputy Director of the IMF Research Department. Polak’s model emphasized the effects of fiscal policies and credit creation on the balance of payments, working primarily through a Keynesian multiplier process. This exposition contrasted with the “Chicago” version of the monetary approach developed by Harry Johnson around the same time, which emphasized the “essential” role of monetary policy (Polak, 2001). In the classic situation, a country with a fixed or managed exchange rate and an external payments deficit can resolve the imbalance by reducing domestic credit of the banking system by either fiscal or monetary means. This simple model became the basis for the specification of macroeconomic policy advice and conditionality by the IMF staff. To some extent, it is still an important building block, though in today’s world program design extends well beyond its confines (IMF, 1987; Polak, 1998).
C. The Open-Economy Macro Model
Within a few years of the introduction of the Polak model, two of Jacques Polak’s staff—Marcus Fleming and Robert Mundell—separately developed the strands of what Rudi Dornbusch would later weave together into the Fleming-Mundell model (Boughton, 2003a). In the early 1960s, Fleming was a division chief in the Research Department (he later became its Deputy Director); Mundell, in a two-year hiatus from his ascending academic career, was an economist in Fleming’s division. Fleming extended the IS/LM framework into an open-economy model that was capable of explaining the distinct effects of fiscal and monetary policies under either fixed or flexible exchange rates. Mundell developed a simpler alternative version of the model and focused on sorting out the dynamic effects of macro policies under varying conditions.
The Fleming-Mundell model had a great intellectual impact from the time the seminal articles were published. Its emphasis on the effects of capital mobility clearly undermined the intellectual basis for Article VI, which treated the capital and current account as independent phenomena. The model’s practical implications then became increasingly apparent after the advent of generalized floating and the growth of capital mobility a decade later. Monetary and fiscal policies were no longer seen as alternative and more or less equivalent means of stabilizing income, as they had been in the Keynesian analyses of the 1950s. Their effects were now known to be distinct and to depend crucially on the exchange regime and the degree of capital mobility. Largely as a consequence of this insight, IMF policy advice gradually expanded to incorporate a broader range of macroeconomic policy actions. The “twin deficits” arguments that the IMF used in the 1980s to criticize the United States for its explosion of fiscal and external deficits derived from this line of reasoning. More generally, the econometric forecasting models developed in the Fund’s Research Department in the 1980s were essentially sophisticated variants of the Fleming-Mundell model, including the rational-expectations elements introduced by Dornbusch (1976).
The emergence of monetarism as a theory of aggregate demand (Friedman, 1956; Brunner, 1968) probably had less impact on the IMF than on the economics profession at large, and its influence was felt primarily in efforts made to examine and ultimately to reject it. In its crudest form (as contrasted with the more nuanced versions discussed in Gordon, 1974), the theory stated that the velocity of money was so stable that policy-induced changes in the supply of money would be reliably transmitted to changes in the price level, and that other influences on aggregate prices could be safely ignored. To economists steeped in an open-economy Keynesian tradition and accustomed to looking for patterns in cross-country analyses, none of the elements of this syllogism seemed particularly persuasive. Studies at the IMF tended to show that for most countries one could estimate a reasonably stable equation linking some measure of the money stock to prices in a form that was reasonably consistent with the theoretical construct of a demand function. Those equations, however, were functions of interest rates and other variables subject to influences other than monetary policy, and they displayed few properties that were consistent across countries or over time (e.g., Argy, 1970; Crockett and Evans, 1980; Boughton, 1991). Similarly, the supply of money could not be assumed to be completely policy-controlled, particularly when the exchange rate was fixed or actively managed.
Despite these limitations and misgivings, monetarist theory had a forceful pull when high inflation became a nearly global phenomenon in the late 1970s. Even if the sources of that inflation extended beyond excessive monetary growth, controlling it would require reining in monetary growth through a tightening of monetary policy. Heterodox alternatives such as incomes policies had little appeal in the Fund, and even fiscal policy was generally seen as insufficiently forceful and constrained under the circumstances. When Paul Volcker, as Chairman of the Federal Reserve System, imposed a seemingly monetarist discipline on U.S. monetary policy starting in late 1979, with dramatic effects on inflation, it was hard to resist being swept along. Nonetheless, the staff persisted with the view that inflation could be controlled through either fiscal or monetary means—preferably both—and that rigidities in the former meant that “monetary policy has borne a disproportionate share of the burden of such restraint” (IMF, 1983, p. 27).
E. The Case for Floating Exchange Rates
Long before the collapse of the par value system in 1973, economists had begun to examine whether exchange rates had to be fixed in order to contribute to economic stability and the growth of international trade. Until the early 1950s, “convertibility” was generally interpreted to mean that a currency could be converted into something else (often, gold) at a fixed price. Milton Friedman (1953), Gottfried Haberler (1954), and James Meade (1955) challenged that view and established an intellectual position that floating and convertibility could be consistent and that floating need not be destabilizing. Friedman’s argument was directed specifically at the Bretton Woods par value system, which he argued was “ill suited to current … conditions.” Floating, in his view, was “absolutely essential for … unrestricted multilateral trade.”
The case for floating took a long time to influence thinking in the IMF. As long as the major industrial countries were committed to maintaining a system of fixed rates anchored on a gold-convertible U.S. dollar, the priority in the Fund was to make that system work as well as possible. Canada’s decision in 1950 to float its currency was viewed with concern in the Fund as a possible threat to systemic stability (Horsefield, 1969, Vol. I, pp. 272–75). Even after the fixed-rate system collapsed, the committee of IMF Governors known as the Committee of Twenty spent two futile years trying to formulate a viable replacement system. Only when that effort failed did interest shift toward examining how a stable system could emerge in a world without stable exchange rates. That led to the idea of IMF surveillance over countries’ exchange rate policies, exercised through regular consultations and supplemented by periodic World Economic Outlook (WEO) reports. From that point on, the Fund took an eclectic case-by-case view on what constituted an appropriate exchange rate regime for any particular country (Mussa and others, 2000, Appendix IV). Along with the rest of the economics profession, the Fund staff continued to debate and reflect on whether any general principles could be applied in practice (Rogoff and others, 2004).
F. Supply-Side Macroeconomics
The term supply-side economics has taken on a variety of meanings over the past quarter-century. In the 1970s, it referred to efforts to model the supply side of the economy as an adjunct to Keynesian analysis of the demand side. That line of reasoning, exemplified by the stagflation model developed by Michael Bruno and Jeffrey Sachs (1981, 1985), was influential in the Fund and was reflected in the WEO and other studies as well as in the Fund’s policy advice and conditionality. In the 1980s, the term was hijacked by tax-cut advocates who argued either that lowering tax rates would raise tax revenues by stimulating economic activity (Canto, Joines, and Laffer, 1983) or that a shift from taxes to deficit financing would have no real effects (“Ricardian equivalence”; Barro, 1974). By the 1990s, it had branched out to encompass advocates of low interest rates and monetary expansion, on the grounds that inflation would be held in check by productivity growth stimulated by easy money (e.g., Kemp, 2001). These radical views never took hold in the Fund.
G. New Classical Economics
The theoretical development that probably had the biggest post-Keynes impact on the IMF was the reformulation of the micro foundations of macroeconomics in the 1970s and early 1980s. Rational expectations theory seemed to undermine the basis for countercyclical demand management policy. In its place came the case for stable policies and nominal anchors to underpin stable expectations. The economics of information was being independently developed around the same time, and that work would eventually lead to a synthesis in which the countercyclical effects of monetary and fiscal policies could be more clearly understood. In the meantime, the new classical concepts held the floor.
The Fund did not develop a doctrine on this issue, but its surveillance activities (both in the WEO and in consultations with individual countries) did shift toward laying greater stress on the desirability of a medium-term policy framework and toward skepticism about the efficacy of countercyclical policies. In the early 1980s, it was still possible for the staff working on Japan to advise the government to take expansionary fiscal action to counter a slowdown, while the staff working on the United States were endorsing the eschewal of such policies by the Reagan administration (Boughton, 2001a, Chapter 3). The clearest example of the shift in thinking, however, was in the annual consultations with Germany, where the staff gradually abandoned the view that persistently high unemployment was due to weak demand and increasingly focused on rigid labor markets and other supply-side issues as the source of the problem (Ibid.).
H. The Silent Revolution
Until the late 1980s, state socialism—government control over economic activity—had a dominant role in driving economic development in many parts of the developing world, in economies as diverse as India, Mexico, and Tanzania. After Julius Nyerere stepped down as President of Tanzania in 1985, his successors gradually liberalized the economy and moved away from policies such as the “villagization” of agriculture and the nationalization of banks. Mexico began liberalizing its international trade policies in the mid-1980s, a move that led to GATT membership in 1986 and prepared the way for more comprehensive economic reforms in the following decade. Under Prime Minister Rajiv Gandhi, India also initiated a major liberalization process in the second half of the 1980s. By the end of the decade, economic liberalization had become a seemingly universal and unstoppable force.
The major effect on the IMF of this “silent revolution”—as the Fund’s Managing Director, Michel Camdessus, called it—was to help ease longstanding tensions between the institution and many of its borrowing members and make it easier to negotiate adjustment and reform programs that the Fund could support. By the early 1990s, there was enough of a consensus about the broad features of desirable economic policies that the Fund’s high-level governing body, the Interim Committee, could unanimously adopt a series of resolutions embodying principles of economic liberalism. The Committee’s “Madrid Declaration,” for example, noted that the “recent success of many developing countries illustrates … the validity of a strategy based on steadfast implementation of strong programs of macroeconomic adjustment and structural reform. The Committee urges other countries to follow a similar bold strategy …” (IMF Annual Report, 1995, pp. 207–208). That appeal, however, was issued just a few months before the Mexican peso crisis led off a series of financial crises that would force a reevaluation of such policy advice, particularly regarding the liberalization of international capital flows.
I. The Washington Consensus
In 1990, John Williamson labeled the type of policy advice meted out by the IMF and the World Bank—supposedly with the encouragement of the U.S. Treasury—as the “Washington Consensus.” Much of what Williamson included in that rubric was similar to the indigenous revolution or evolution in thinking in developing and developed countries around the world. His terminology was therefore more a catchy phrase than an accurate pinpointing of the source of these ideas, as he himself has acknowledged (Williamson, 2000). Nonetheless, it did catch on, and after the flurry of financial crises in the second half of the nineties it became a lightning rod for criticism of globalization in general and the IMF in particular. Although liberalization of capital flows was only an incidental adjunct to Williamson’s consensus list, that aspect of policy reform became especially controversial. The same countries that had benefited from inflows after the debt crisis faded away now were reeling from the effects of sudden losses of confidence and corresponding withdrawals of capital. By the turn of the century, the phrase “Washington Consensus” had become a synonym for a narrow-minded and excessive zeal for laissez faire market economics.
It is certainly true that the IMF—both officially and in the individual views of most of its professional staff—embraced the policies that Williamson collected under the umbrella of the Washington Consensus. As Stanley Fischer (the Fund’s First Deputy Managing Director from 1994 to 2001) put it recently (2003, p. 6), the Washington Consensus is “a useful shorthand description of a desirable basic policy orientation.” It is also true that the Fund went through a phase in the 1990s in which the free mobility of capital was seen as an essential ingredient in economic policy, though it always recognized that liberalization had to be underpinned by sound financial systems and prudential supervision of markets. The Interim Committee’s communiqué of October 1995 noted that “increased freedom of capital movements and globalized markets bring significant benefits to all countries,” and it attributed volatility in capital flows to inconsistent implementation of “firm economic policies.” In April 1997, the Committee “agreed that the Fund’s Articles should be amended to make the promotion of capital account liberalization a specific purpose of the Fund and to give the Fund jurisdiction over capital movements.” After the Asian crisis, however, that idea gradually lost support. By April 1999, the Committee had toned down its request to encourage the Fund “to continue its work on the appropriate pace and sequencing of capital account opening and … to explore further issues related to the IMF’s role in an orderly and well-supported approach to capital account liberalization.”14
J. Inflation Targeting
The use of monetary policy to pursue price stability (meaning a low rate of inflation) as a single target instead of as part of a broader strategy to balance inflation and employment objectives, and the direct targeting of inflation rather than relying on intermediate indicators such as interest rates or monetary aggregates, captured the imagination of central bankers and economists in the 1990s. The trend began in New Zealand in 1989, was picked up by Canada a year later, and by the end of the 1990s had spread to at least a dozen other countries (Schaechter, Stone, and Zelmer, 2000).
The spread of inflation targeting as a monetary policy strategy provided new opportunities and challenges for the IMF. The opportunity was to try to use this strategy to encourage these and other countries to adopt more stable monetary policies. In general, the Fund did so, though with the caveat that the right conditions—well developed financial markets, sound fiscal policies, and an overall stable macroeconomic environment—should be in place before inflation targeting can be expected to contribute to economic performance. Since 1995, IMF staff have published some two dozen working papers on inflation targeting, most of which focus to some extent on establishing the preconditions for successful implementation either generally or in specific countries.
The operational challenge for the Fund has been to adapt program design and conditionality when borrowing countries are targeting inflation rather than conventional monetary policy instruments. In these cases, variables that are usually the subject of IMF policy conditions, particularly floors on net international reserves and ceilings on domestic credit expansion, are not separately controllable by the central bank. Setting conditions on the inflation rate itself would weaken the Fund’s ability to monitor policy implementation because of the lag between policy changes and inflation effects (Blejer and others, 2002). To date, the Fund has tried to steer a middle course, adhering to its conventional instruments while monitoring inflation and other indicators as a further check on implementation and consistency. Whether inflation targeting will continue to gain in applicability and acceptance and/or will force changes in the Fund’s operations is too early to judge.
III. Conclusions: How Has History Shaped the IMF?
The IMF was created at a particular time in world history—during the Second World War—and was given a structure and mandate that reflected that time and those circumstances. The institution has changed greatly in the six decades since Bretton Woods. Much of the volume of its lending has become crisis-driven, and the Fund’s involvement in crisis prevention and resolution has correspondingly intensified. It has effectively become divided into groups of creditor and debtor countries whose membership changes little over long periods of time. Its membership is much larger, more diverse, and nearly universal, and its responsibilities in global governance have correspondingly increased. The breadth of its involvement in policymaking in member countries, especially borrowing countries, has vastly increased.
If the events and ideas chronicled here had not affected the IMF along these lines, the institution would have become marginalized and even irrelevant. The evolution of the Fund has been driven almost entirely by shifts in demand—shifts in world economic and political conditions—not by forces from within seeking to reinvent the institution so as to hang onto a role once the original purpose had faded away. The challenge for the IMF has always been to hold on to its vital center (the original narrow mandate to promote orderly payments adjustment and global financial stability) while adapting its activities to new circumstances and new ideas. Meeting that challenge became increasingly difficult in the 1970s and 1980s, when the advent of generalized currency floating, financial globalization, the need for multilateral crisis management, and financial demands from low-income countries all pressed new functions and responsibilities onto the Fund. By the 1990s, when the Fund had to deal with all of those issues plus the need for rapid structural reforms in formerly centrally planned economies—including Russia, with its great geopolitical importance—“mission creep” became inevitable.
Even accepting that most of the changes in the Fund have occurred for good reasons and could not have been avoided in any case, the argument for adhering to a consistent mandate and mission is not diminished. Institutions have limited resources and employ staff with specific skills and experience, and diffusing those resources imposes substantial costs. The recent commitments by the IMF (2001, 2002) to streamline and refocus its policy conditions and strengthen its cooperation with the World Bank were taken in recognition of that imperative. As much as the world has changed, the raison d’être for the IMF—compensating for the limited global reach of the invisible hand, the goal that first led Keynes and White to create institutions to promote multilateral cooperation—remains as vital as ever.
- Search Google Scholar
- Export Citation
- Search Google Scholar
- Export Citation
- Search Google Scholar
- Export Citation
ArgyVictor1970 “The Role of Money in Economic Activity: Some Results for 17 Developed Countries” IMF Staff PapersVol. 17 (November) pp. 527–62.
ArnonArie and WarrenYoung(Editors) 2002The Open-Economy Macro-Model: Past Present and Future (Dordrecht: Kluwer Publishing).
BarroRobert J.“Are Government Bonds Net Wealth?” Journal of Political EconomyVol. 82 (November/December) pp. 1095–1117.
BlejerMario I.Alfredo M.LeonePauRabanal and GerdSchwartz2002 “Inflation Targeting in the Context of IMF-Supported Adjustment Programs” IMF Staff PapersVol. 49No. 3 pp. 313–38.
BordoMichael D. and BarryEichengreen1998 “Implications of the Great Depression for the Development of the International Monetary System” in BordoGoldin and Whitepp. 403–53.
BordoMichael D.ClaudiaGoldin and Eugene N.White(editors) 1998The Defining Moment: The Great Depression and the American Economy in the Twentieth Century (Chicago: University of Chicago Press).
BoughtonJames M.1991 “Money Demand in Five Major Countries: Estimating and Interpreting Error-Correction Models” in Mark P.Taylor(editor) Money and Financial Markets (London: Basil Blackwell).
BoughtonJames M.2000 “From Suez to Tequila: The IMF as Crisis Manager” Economic JournalVol. 110 (January) pp. 273–91.
BoughtonJames M.2001aSilent Revolution: The International Monetary Fund 1979–1989 (Washington: International Monetary Fund).
BoughtonJames M.2001b “Northwest of Suez: The 1956 Crisis and the International Monetary Fund” IMF Staff PapersVol. 48No. 3 pp. 425–46.
BoughtonJames M.2002 “Why White, Not Keynes? Inventing the Post-War International Monetary System” in Arnon and Youngeds. (2002) pp. 73–96.
BoughtonJames M.2003a “On the Origins of the Fleming-Mundell Model” IMF Staff PapersVol. 50No. 1 (2003) pp. 1–9.
BoughtonJames M.2003b “New Light on Harry Dexter White” Journal of the History of Economic Thoughtforthcoming.
BrunnerKarl1968 “The Role of Money and Monetary Policy” Federal Reserve Bank of St. Louis ReviewVol. 50 (July) pp. 8–24.
BrunoMichael and Jeffrey D.Sachs1981 “Supply versus Demand Approaches to the Problem of Stagflation” in HerbertGiersch(editor) Macroeconomic Policies for Growth and Stability (Kiel: Institut für Weltwirtschaft).
BrunoMichael and Jeffrey D.Sachs1985The Economics of Worldwide Stagflation (Cambridge, Massachusetts: Harvard University Press).
CantoVictor A.Douglas H.Joines and Arthur B.Laffer1983Foundations of Supply-Side Economics: Theory and Evidence (New York: Academic Press).
CrockettAndrew D. and Owen J.Evans1980 “Demand for Money in Middle Eastern Countries” IMF Staff PapersVol. 27 (September) pp. 543–77.
CruciniMario J. and JamesKahn1996 “Tariffs and Aggregate Economic Activity: Lessons from the Great Depression” Journal of Monetary EconomicsVol. 38 (December) pp. 427–67.
DornbuschRudiger1976 “Exchange Rate Expectations and Monetary Policy” Journal of International EconomicsVol. 6 pp. 231–44.
FischerStanley2003 “Globalization and Its Challenges” American Economic ReviewVol. 93 (May) pp. 1–30.
FriedmanMilton1953 “The Case for Flexible Exchange Rates” in his Essays in Positive Economics (Chicago: University of Chicago Press).
FriedmanMilton1956 “The Quantity Theory of Money—A Restatement” in MiltonFriedman(editor) Studies in the Quantity Theory of Money (Chicago: University of Chicago Press) pp. 3–21.
GardnerRichard N.1956Sterling-Dollar Diplomacy: Anglo-American Collaboration in the Reconstruction of Multilateral Trade (Oxford: Clarendon Press).
GordonRobert J.(editor) 1974Milton Friedman’s Monetary Framework: A Debate with his Critics (Chicago: University of Chicago Press).
HaberlerGottfried1954Currency Convertibility (New York: American Enterprise Association).
HarrodRoy1951The Life of John Maynard Keynes (London: Macmillan).
HorsefieldJ. Keith1969The International Monetary Fund 1945–1965. Vol. I: Chronicle. Vol. II: Analysis. Vol. III: Documents (Washington: International Monetary Fund).
Independent Evaluation Office of the IMF (IEO)2003Fiscal Adjustment in IMF-Supported Programs (Washington: International Monetary Fund).
International Monetary Fund1983World Economic Outlook: A Survey by the Staff of the International Monetary Fund. Occasional Paper 21 (May).
International Monetary Fund1987Theoretical Aspects of the Design of Fund-Supported Adjustment Programs. A Study by the Research Department of the International Monetary Fund. Occasional Paper 55 (September).
International Monetary Fund2001 “IMF Concludes Discussions on Strengthening IMF-World Bank Collaboration on Country Programs and Conditionality” Public Information Notice No. 01/92 (September4); on the internet at http://www.imf.org/external/np/sec/pn/2001/pn0192.htm.
International Monetary Fund2002 “Guidelines on Conditionally” (September25) on the internet at http://www.imf.org/External/np/pdr/cond/2002/eng/guid/092302.htm.
IrwinDouglas A.1998 “From Smoot-Hawley to Reciprocal Trade Agreements: Changing the Course of U.S. Trade Policy in the 1930s” in BordoGoldin and White(1998).
JamesHarold1996International Monetary Cooperation Since Bretton Woods (IMF and Oxford University Press).
JamesHarold and MarzennaJames1994 “The Origins of the Cold War: Some New Documents” Historical JournalVol. 37No. 3 pp. 615–22.
KempJack2001 “The Greenspan Recession” Jewish World Review (August17); on the internet at http://www.jewishworldreview.com/cols/kemp081701.asp.
KeynesJohn Maynard1936The General Theory of Employment Interest and Money (London: Macmillan).
LaidlerDavid E. W. and Roger J.Sandilands2002 “An Early Harvard Memorandum on Anti-Depression Policies: Introductory Note” History of Political Economysummer.
MacMillanMargaret2001Paris 1919: Six Months that Changed the World (New York: Random House).
MeadeJames E.1955 “The Case for Variable Exchange Rates” Three Banks Review; reprinted in Susan Howson (editor), The Collected Papers of James Meade, Vol. III: International Economics (London: Allen & Unwin, 1988).
MussaMichaelPaulMassonAlexanderSwobodaEstebanJadresicPaoloMauro and AndyBerg2000Exchange Rate Regimes in an Increasingly Integrated World Economy. Occasional Paper No. 193 (Washington: International Monetary Fund).
PaulyLouis1997Who Elected the Bankers? Surveillance and Control in the World Economy (Ithaca, NY: Cornell University Press).
PolakJacques J.1994Economic Theory and Financial Policy: The Selected Essays of Jacques J. PolakVol. 1 (Aldershot, England: Edward Elgar Publishing Ltd.).
PolakJacques J.1998 “The IMF Monetary Model at 40” Economic ModellingVol. 15 pp. 395–410.
PolakJacques J.2002 “The Two Monetary Approaches to the Balance of Payments: Keynesian and Johnsonian” in Arnon and Youngeds. (2002) pp. 19–41.
PrebischRaúl1971Change and Development: Latin America’s Great Task (Washington: Inter-American Development Bank).
RogoffKenneth S.2003 “The IMF Strikes Back” Foreign Policy 134 (January/February).
RogoffKenneth S.Aasim M.HusainAshokaModyRobinBrooks and NienkeOomes2004 “Evolution and Performance of Exchange Rate Regimes” IMF Occasional Paperforthcoming.
SchaechterAndreaMark R.Stone and MarkZelmer2000Adopting Inflation Targeting: Practical Issues for Emerging Market Countries. Occasional Paper No. 202 (Washington: International Monetary Fund).
SkidelskyRobert2000John Maynard Keynes: Fighting for Britain 1937-1946 (London: Macmillan).
StiglitzJoseph E.2002Globalization and its Discontents (New York: W. W. Norton).
WilliamsonJohn(editor) 1990Latin American Adjustment: How Much Has Happened? (Washington: Institute for International Economics).
WilliamsonJohn2000 “What Should the World Bank Think about the Washington Consensus?” World Bank Research ObserverVol. 15 (August) pp. 251–64.