This book is the fourth installment in the officially sponsored history of the International Monetary Fund. It begins where the previous installment ended, shortly after the entry into force of the Second Amendment to the Articles of Agreement of the Fund. It ends with the adoption of the Brady plan, which aimed to resolve the debt crisis that dominated the work of the Fund for most of the 1980s, but before the collapse of the Soviet Union, which made new and different demands on the human and financial resources of the institution.
Although this is an officially sponsored history, it is not a mere chronicle. In fact, it is not even organized chronologically. In the first part, James Boughton describes the Fund’s efforts to influence the policies of the major industrial countries. In the second part, he examines the Fund’s involvement in the debt crisis and thus focuses on its relations with the middle-income developing countries of Latin America. In the third part, Boughton discusses the Fund’s attempts to foster growth-promoting reforms in low-income developing countries and provide those countries with concessional financing. In the last part, he discusses the evolution of the Fund as a financial institution, the changing role and attributes of the SDR, the start of the move toward universal membership, the Fund’s relations with other institutions, and changes in its organization and governance.
Previous volumes in this series were written when the Fund was obsessively protective of its confidential relationship with its members. The authors of those volumes, however, were given unrestricted access to the Fund’s archives and could therefore provide outsiders with a window on the Fund’s decision-making processes. The first volume (Horsefield, 1969) relied largely on the minutes of the Executive Board, papers prepared for the Board, and other documents in the Fund’s archives; it did not provide references to individual documents but did quote Executive Directors by name. The second and third volumes (de Vries, 1976, 1985) used the same sources, subject to the same constraints, but exploited outside sources, including the author’s interviews with many officials who had dealings with the Fund. Those volumes also discussed important developments outside the Fund, such as the work of the Group of Ten.1
In 1996, however, the Fund began opening its archives, and most of the documents on which Boughton relies are now available to outsiders or will be available soon. Therefore, Boughton is able to cite individual documents and can also devote less attention to the deliberations of the Executive Board, in order to say more about the roles and views of the Managing Director and staff. Yet he tells us more than we need to know about the chronology of negotiations between the Fund and governments—the travels of staff missions and finance ministers—and less than we need to know about the issues with which they were struggling. Furthermore, he does not always explain why a Fund-supported program “went off track”—a phrase he uses with maddening frequency. Nevertheless, Boughton provides a thorough, thoughtful, and lively account of the Fund’s activities during a turbulent decade. Most readers will probably want to skip some topics or case studies, and Boughton makes that easy by offering frequent summaries of what lies ahead. The patient reader, however, will be well rewarded. This thousand-page book describes some remarkably interesting episodes.
I. An Overview
Boughton takes the title of his book from a passage in a speech by Michel Camdessus, Managing Director of the Fund during the last part of the period covered by the book, in which Camdessus spoke of a “silent revolution” in many developing countries. It is showing itself, he said, in the number of countries that have sought the Fund’s help in devising growth-oriented adjustment programs. But Boughton uses the phrase more broadly. There was, he writes, a shift “away from tendencies toward autarky, insularity, mercantilism, and governmental planning and control over economic activity; and toward a common set of beliefs and policies based on open international trade and finance, competitive pricing and production decisions, and cooperation between countries.” This shift, he says, reflected a basic change in economic philosophy “toward a new classical synthesis in which government has an indirect role in, but not a direct responsibility for, ensuring national economic prosperity; in which private economic activity is promoted through good governance and the development of physical and social infrastructure” (Boughton, p. 3).
Boughton’s book explores the consequences for the IMF—how the silent revolution altered the Fund’s relations with the major industrial countries and how the Fund sought to support it in developing countries. In its dealings with the industrial countries, no longer obliged to defend pegged exchange rates and thus less dependent on IMF financing, the Fund had to struggle for influence. In its dealings with the developing countries, it faced a different problem: how to reconcile the need for painful adjustment and far-reaching reform with the need for weak national governments to claim ownership of their Fund programs.
The book begins with a review of major developments in the world economy, the international monetary system, and the IMF from the end of the 1970s to the start of the 1990s. It also includes brief biographies of Jacques de Larosière and Michel Camdessus, the Managing Directors who led the Fund during those years. The Fund, says Boughton, faced serious internal weaknesses at the start of the 1980s:
The surge in lending to developing countries at the end of the 1970s had come at the expense of maintaining the quality of adjustment programs and of the Fund’s financial portfolio. The failure of the effort to establish a substitution account had weakened the potential for the SDR to play a central role in the international monetary system and had deprived de Larosière of the centerpiece of a strategy for strengthening the institution. The waning of demand for the Fund’s resources by industrial countries, together with the very limited consensus on whether and how to stabilize exchange rates, left the Fund struggling to define a clear role for its core function, surveillance. If the Fund was to remain the world’s premier monetary institution, it would have to nurse its roots back to health or redefine its role. As the decade progressed, it did enough of both to dominate the global economic stage more than ever before (Boughton, p. 41).
Controversial premises lurk within this passage: that the quality of the Fund’s financial portfolio had been impaired by the large volume of low-conditional lending in the 1970s; that surveillance over its members’ policies could have given the Fund significant leverage over those national policies so as to make surveillance the “core function” of the Fund; that replacing dollar reserves with an SDR-denominated asset—the aim of the substitution account—would have given the SDR and, therefore, the Fund a “central role” in the international monetary system. Boughton is right to say that the Fund played a more prominent role during the 1980s than during previous decades. It did so, however, by redefining its role, largely in response to the debt crisis, rather than performing the tasks defined by the Second Amendment—overseeing its members’ exchange rate policies and managing the growth of international reserves.
The Fund’s original Articles of Agreement embodied a remarkable degree of agreement about the appropriate form and functioning of the postwar monetary system. The Second Amendment, by contrast, was crafted in language designed to conceal deep disagreements about the future functioning of the system, and they were not resolved thereafter. They were instead compounded in the early 1980s by deep disagreements among the industrial countries about economic policy in general. Therefore, the Fund was unable to exercise any significant influence over those countries’ policies. Furthermore, the restrictions imposed on SDR creation by the Second Amendment and the collapse of the subsequent effort to devise a substitution account kept the Fund from exercising any appreciable influence over the evolution of international reserves.
The Fund did not merely revive conditionality in the 1980s but greatly broadened its scope in the attempt to foster growth-promoting adjustment. One is led indeed to wonder whether the roles of financing and conditionality had been reversed by the end of the 1980s. Originally, conditionality was viewed as a way of making sure that the Fund would be repaid when it provided financing. During the 1980s, by contrast, the provision of financing seemed increasingly to be viewed as a way of promoting reform. In several cases, moreover, new money from the Fund served merely to roll over the Fund’s existing claims rather than provide additional financing. There is indeed an odd similarity between the Fund’s efforts to mobilize commercial bank lending during the debt crisis and its own concessional lending. The commercial banks’ loans to the middle-income countries served as a roundabout way to capitalize the banks’ interest-income claims; the Fund’s concessional lending to the low-income countries served as a roundabout way to refinance the Fund’s own claims.
II. Surveillance and the Developed Countries
The first main part of Boughton’s book deals with surveillance and, in that context, the Fund’s involvement in the affairs of the major industrial countries. It starts with a chapter on the origins of surveillance, which reviews at length the efforts of the Executive Board to develop guidelines for surveillance. Those efforts, Boughton explains, were hugely complicated by basic disagreements produced by the ambiguities built into the Second Amendment. Was the Fund supposed to discourage its members from trying to manage floating exchange rates, or was it supposed to encourage its members to pursue national policies aimed at achieving exchange rate stability? The chapter also reviews the staff’s efforts to identify equilibrium exchange rates and the evolution of its views about appropriate exchange rate policies for the developing countries. It concludes with accounts of two cases in which the Fund held special consultations with individual countries because other countries had criticized their policies. In 1982, Sweden’s neighbors accused it of devaluing the krone excessively; in 1987, the United States criticized Korea for refusing to let the won appreciate and thus reduce its current-account surplus. Boughton says that the Fund played a useful role in resolving these disputes, although he concedes that its involvement had no obvious influence on Swedish or Korean exchange rate policy.
Boughton turns next to the Fund’s surveillance of the major industrial countries and its formal role in those countries’ sporadic attempts at policy coordination and exchange rate management. The story is well told, but the Fund does not play a very large part. In a footnote, indeed, Boughton reports that his interviews with Group of Seven (G-7) officials failed to identify a single case in which a policy decision was influenced by the Fund’s advice (Boughton, p. 140). Much of his account, moreover, is based on outside sources rather than Fund documents—which is itself revealing. But he describes some interesting episodes involving the Fund directly. Beginning in 1981, the staff and Executive Board began to criticize the fiscal policy of the Reagan administration. In 1985, moreover, the staff responded to a claim made by the Reagan administration that there was no causal link between the U.S. budget deficit and high real interest rates. The Research Department attached a 140-page “supplementary note” to the draft of the World Economic Outlook (WEO), which found that a cut in the budget deficit would reduce real interest rates. But the U.S. authorities insisted that the note be omitted from the published version of the WEO, and they also succeeded in blocking suggestions that the note be published in some other form (Boughton, p. 145).
Disagreements about U.S. exchange rate policy were “surprisingly mild” compared with the sharp disagreements about U.S. fiscal policy. In its report on the 1981 consultation with the United States, the staff supported the U.S. Treasury’s announcement that it would no longer intervene on foreign exchange markets. “It is readily apparent,” the staff said, “that a substantial degree of exchange rate flexibility needs to be maintained in present circumstances” and that information obtainable from the foreign exchange market should be used to guide domestic policy. By 1983, however, when the dollar had begun to appreciate strongly, the staff began to speak of the need for “orderly correction in the exchange value of the dollar” (Boughton, pp. 150–1).
While urging the United States to cut its budget deficit, the Fund was telling Japan to slow the rapid pace at which it was reducing its deficit. The contrast, Boughton says, reflected a judgment that the U.S. fiscal problem was far more severe, both from a global standpoint and in relation to the large difference between the two countries’ savings rates. There was also a difference, however, in the Fund’s views of the two countries’ efforts to limit their reliance on fiscal policies for countercyclical stabilization. It told the United States to do precisely that, but it told Japan to show more flexibility. This difference, Boughton says, is harder to explain, as the Japanese economy had shown more resilience to negative shocks. For the most part, however, the Fund’s consultations with Japan were marked by “admiration and support for the authorities’ conduct of growth-oriented and financially stable policies” (Boughton, p. 154). Throughout the buildup of the “bubble economy,” the Fund supported the Japanese policy stance. Even in 1989, when inflation accelerated and the yen weakened, the staff found “no compelling reason” for Japan to tighten its monetary policy (Boughton, p. 163).
There is another remarkable difference between the Fund’s views about two industrial countries. Although the Fund did not disagree strongly with U.S. exchange rate policy in the 1980s, it took frequent exception to U.K. policy. At the start of the Thatcher era, Britain’s policy stance resembled the one adopted thereafter by the Reagan administration. Exchange rates were market prices and should be market determined. The Fund’s staff did not disagree initially. Two years later, however, it concluded that sterling should not be allowed to appreciate further, and it began suggesting quietly that Britain might benefit from joining the exchange rate mechanism of the European Monetary System (EMS) once the overvaluation of the pound had been corrected. But the staff changed its mind again when Britain came closer to taking that step; it concluded that Britain should wait until it had narrowed the gap between British and German inflation rates (Boughton, pp. 182–3).
Turning to policy coordination among the major industrial countries, Boughton explains how the Fund was brought into the process in 1982 and its subsequent efforts to devise “objective indicators” as a framework for multilateral surveillance. Formal endorsement of the Fund’s role took place at the Versailles Summit in 1982, which made these commitments:
- 1. We accept a joint responsibility work for the greater stability of the world monetary system. We recognize that this rests primarily on convergence of policies …
- 2. We attach major importance to the role of the IMF as a monetary authority and we will give it our full support in its efforts to foster stability.
- 3. We are ready to strengthen our cooperation with the IMF in its work of surveillance; and to develop this on a multilateral basis …
- 4. We rule out the use of our exchange rates to gain unfair competitive advantage.
- 5. We are ready, if necessary, to use intervention in exchange rates to counter disorderly conditions, as provided for under Article IV of the IMF Articles of Agreement.
This declaration, Boughton says, was consistent with the U.S. view that exchange rate stability depends on policy convergence, but it was also consistent with the French view that exchange rate stability is important and requires a mutual commitment to that goal. “That it was possible to draft a series of apparently firm declarations that equally supported two diametrically opposed visions is a remarkable tribute to the obfuscatory skills of the sherpas and their deputies” (Boughton, p. 194). He should perhaps have wondered why those skills seem to thrive in the neighborhood of Paris. The accord that paved the way for agreement on text of the Second Amendment to the Fund’s Articles of Agreement was reached at the Rambouillet Summit, seven years earlier. Its ultimate incarnation was the strange phrase in the present text of Article IV, which commits the Fund’s members to “promote a stable system of exchange rates.” No one has been able to give operational meaning to that phrase, but the agreement to put the word “stable” in an unnatural place played a key role in defusing the conflict between American and French views about the future of the monetary system. (See de Vries, 1985, pp. 739–49.)
Finally, Boughton describes the Fund’s role in the episodic efforts at exchange rate management—the Plaza Agreement of 1985 and Louvre Accord of 1987. He does his best to emphasize the Fund’s involvement, but that is not easy. In fact, he describes its role as peripheral. In January 1987, the G-5 deputies met in Zurich, and the Fund’s representative participated in the deputies’ discussion of the objective indicators drafted by the Fund. After that, however, he left the room so that the deputies might discuss exchange rate policy. At the Louvre meeting itself, Camdessus, the new Managing Director, submitted a note that dealt “more openly” with exchange rates than had been customary, and his predecessor, de Larosière, having replaced him as Governor of the Banque de France, took part in the subsequent discussion of exchange rate policy. (He was, one participant said, “the ghost of the IMF.”) Unfortunately, Boughton is unable to resolve the dispute that still surrounds the substance of what was actually decided at the Louvre (Boughton, pp. 128–9).
In the last chapter of this part, Boughton offers a thorough account of the way that the staff provided analytical support for the Fund’s efforts to influence national policies, not only in the context of surveillance but in broader forums too. He traces the evolution of the WEO from the 1960s, when it served as a background document for an informal discussion by the Executive Board, to the first published version in 1980, and on to the current regime, semiannual publication, in 1984. More importantly, he traces the evolution of the methodology used to prepare the WEO.2 Buried in this chapter, however, is a four-page section that belongs elsewhere. In fact, it deserves a chapter of its own. It begins this way:
The controversy that more than any other could be characterized as a battle for the heart and soul of the IMF was the debate over fixed versus floating exchange rates. As with other economic issues, this debate took place on two levels: the field, especially in the policy recommendations given to countries requesting to borrow from the Fund, and at headquarters, especially research and policy papers (Boughton, p. 247).
The discussion that follows, however, is disappointing. There is little here or elsewhere about the course and consequences of the internal debate. Although Boughton alludes to it often and in several contexts, he does not explain how it impinged on the work of the Fund—whether and how it divided the staff, how it affected the design of Fund-supported programs, and how it cropped up in the Board. A visitor to the Fund in the 1980s and 1990s soon became aware of the internal controversy. There were disagreements between departments about the viability of the EMS and, thereafter, the feasibility of European monetary union, and the various area departments seemed to hold quite different views about the comparative merits of fixed and flexible exchange rates.3 Yet Boughton does not discuss these matters. In fact, he says very little about debates within the staff, making it seem like a monolith.
III. The Debt Crisis
The second main part of Boughton’s book deals with the debt crisis. It begins with a brief overview, which tracks the debt crisis back to its origins—the surge of bank lending to developing countries after the first oil shock of 1973–74—and then outlines the evolution of the debt strategy orchestrated by the Fund after the crisis erupted. Boughton describes that strategy briefly in the first chapter of his book:
The effects of the debt crisis were contained in the early 1980s through a “case-by-case” strategy in which external financial support was provided to countries willing to adjust their economic policies. That strategy succeeded in preventing a series of defaults on sovereign debts, but it did not lead to an early resumption of normal relations between debtors and creditors. A driving assumption in the development of the strategy, much debated in the subsequent literature, was that such defaults could have led to multiple bankruptcies of major commercial banks and possibly to a collapse of the international banking system. The debt strategy also succeeded in greatly reducing the payments deficits of many developing countries, but it did so by forcing a reduction in imports rather than fostering growth in exports.
By 1985, a consensus was forming among officials in creditor countries that new approaches were needed for a more favorable and more sustainable solution to the crisis. The focus of the debt strategy during the next few years was to encourage indebted countries to undertake growth-oriented structural reforms financed largely by longer-term loans from the World Bank and regional developing banks, and to encourage commercial banks to resume net lending to countries undertaking such reforms. That effort failed to generate either long-term growth or even much long-term financing, and by 1988, almost all of the countries hit by a debt crisis several years earlier were still struggling to escape from it. For much of the developing world, the 1980s were to be a “lost decade” for economic growth.
The denouement of the debt strategy arrived when the realization took hold that a high-growth equilibrium could be attained only through debt reduction. The debt-relief approach had two prongs: one aimed at the low-income countries that owed most of their external debt to creditor governments, and the other aimed at the middle-income countries that were heavily indebted to commercial banks (Boughton, p. 31).
That same chapter, moreover, makes a point not stressed enough in Boughton’s detailed account—that there was a huge drop in nonfuel commodity prices in the 1980s. It can, he says, be called “the most severe decline in commodity prices in recorded history” and, for the low-income countries, the “most devastating event of the decade” (Boughton, p. 24).
Elsewhere, Boughton notes that the debt crisis was not wholly unanticipated by the Fund. In the spring of 1981, the WEO had provided a five-year projection of the debt burdens of the oil-importing developing countries; it described them as “worri-some” and warned that many countries “would soon find themselves unable to finance their deficits” (quoted by Boughton, p. 234). In January 1982, moreover, several months before the crisis, de Larosière had addressed a group of leading bankers and urged them to insist that borrowing countries confront their economic problems rather than use bank financing to postpone adjustment. He also urged the banks to maintain or increase their exposure levels when countries adopted appropriate policies (Boughton, p. 268). Yet the sudden onset of the crisis was unanticipated.
The Onset of the Crisis
On August 12, 1982, Mexico’s finance minister, Jesús Silva Herzog, informed the IMF, the U.S. Treasury, and the Federal Reserve that Mexico could not repay debt coming due at the start of the following week. Kraft (1984) describes the event as “a bombshell that shook the entire universe.” The Fund’s staff was aware of serious problems in Mexico and of its need for large-scale borrowing. Four weeks before the onset of the crisis, moreover, the Board had discussed the Mexican situation, and several Directors had expressed concern about the buildup of debt. But no one had predicted the imminent cessation of new foreign lending. There was, indeed, “an assumption—or a hope—that commercial financing would continue to be available until the necessary policy adjustments could be made” (Boughton, p. 288), and those adjustments could not be made until the end of the year, when a new Mexican president would take office. Lending did not continue, however, and Mexico’s reserves were too small to pay off its maturing debt. It had either to default or seek immediate help. But the Mexican authorities had reason to hope for immediate help because a default would threaten the solvency of major banks in major countries. Angel Gurría, Mexico’s debt negotiator, put the point vividly:
We didn’t crawl to the international financial community as debtors seeking relief through some minor adjustment that could be made backstage. We walked in through the front door. We said we had a major problem with a capital P. We didn’t say the problem was a particular debt. We said it was everybody’s problem (quoted by Kraft, 1984, p. 3).
The Fund was prepared to help quickly, but Mexico had first to find a way of avoiding default by making its interest payments and reaching agreement with its bank creditors on a rescheduling of the maturing principal.
After arduous negotiations, Mexico obtained emergency funding from the United States and the Bank for International Settlements (BIS), as well as a 90-day rollover of maturing principal from its commercial bank creditors. But Mexico had still to reach agreement with the Fund on an adjustment program before it could obtain IMF financing, and the amount for which it could qualify would not fully meet its needs. Hence, Mexico would have to borrow some US$5 billion of “new money” from the banks. The Fund made it clear, moreover, that all of the creditor banks would have to be involved. Meeting with the major banks, de Larosière announced that he would not ask the Executive Board to approve the planned drawing by Mexico unless each bank agreed to provide its pro rata share of the US$5 billion. This was the birth of concerted lending—the insistence by the Fund that commercial banks agree collectively and individually to provide enough new lending to guarantee the full financing of a Fund-supported program.
Having explained how the debt crisis erupted in Mexico and how the Fund responded, Boughton goes on to explain how Argentina, Brazil, and Chile were swept into the crisis. Those countries, he argues, were not the innocent victims of contagion. They had serious homegrown problems.
Argentina had introduced a new policy regime in 1978 aimed at reducing inflation and stimulating growth. Its best-known feature was the tablita—a pre-announced schedule governing the gradual depreciation of the Argentine peso. The program reduced inflation, but not by enough to prevent a large real appreciation. Hence, exports and output stagnated, capital formation collapsed, and the balance of payments deteriorated sharply. Argentina became very heavily dependent on loans from foreign banks. Thereafter, the budget deficit grew hugely, fueling inflation, and the peso was devalued twice in 1981. In the spring of 1982, moreover, Argentina tried to seize the Falkland Islands, Britain and Argentina froze each other’s financial assets, and other countries imposed economic sanctions. Argentina could no longer service its external debt. The debt crisis thus came to Latin America several months before the Mexican crisis. But the Mexican crisis made it clear that Argentina could not expect to borrow again, even with a speedy end to the unique problems caused by the Falklands War. Weeks after the Mexican crisis, then, Argentina sought to draw on the Fund.
The problems produced by the war were not resolved quickly, and they hugely complicated Argentina’s situation. Because it had barred all payments to British banks, fancy footwork was required to stave off an Argentine default and arrange a Fund-supported program. As in the Mexican case, moreover, new money would be needed from the banks, and de Larosière insisted that the banks commit themselves before he would take the Argentine program to the Executive Board. When the program was actually brought to the Board, some Directors questioned its viability—whether, as one Director put it, the required adjustment effort “could impose such a heavy burden on the government as to exclude the possibility of complete observance” (quoted by Boughton, p. 334). Nevertheless, the program was approved, and it seemed to be on track in early 1983.
Boughton also downplays contagion as the explanation for Brazil’s debt problem. It was due, he says, to the country’s “dependence on a global economy that was no longer strong enough to generate the growth rate to which the country had become accustomed, and a serious domestic imbalance manifested in a persistently high inflation rate” (Boughton, p. 337). Yet Boughton himself finds no reason to believe that Brazil’s situation was unmanageable, and contagion appears to have played a larger role in the Brazilian case than the Argentine case. In fact, it produced a stronger reaction from the foreign banks. Some of them threatened to pull out completely. After an unfortunate three-month delay, due to the need to await the outcome of a congressional election, Brazil was quick to reach agreement with the IMF on an adjustment program and began negotiating with the commercial banks. In the Brazilian case, however, de Larosière declined initially to call for concerted lending, and it took a long time to wrap up the whole package.4
There were, in fact, two reasons for the delay. Soon after their initial agreement with the Fund, the Brazilian authorities began to back away from it. In February 1983, they devalued the cruzeiro abruptly, although they had said initially that they would not do so. In addition, some commercial banks were refusing to accept a key component of the financing package—a commitment to maintain their interbank credit lines—and de Larosière had finally to opt for a modified form of concerted lending. As he could not persuade the banks to promise firmly that they would restore their interbank lines, he settled for an informal assurance that the banks would consider enlarging their new medium-term loan to Brazil to cover the financing gap that would be produced by the run-down of interbank claims.
More problem arose thereafter. In the first three months of its IMF program, Brazil breached the ceilings on domestic credit growth and the fiscal deficit, causing the Fund to bar the next scheduled drawing. In this instance, Boughton explains why a Fund program went off track:
Without question, the authorities had undertaken a serious and substantial adjustment effort, after the disastrous delay in the fall of 1982. The trade surplus in the first half of 1983 was above target, and—aside from the effect of inflation on the government’s domestic interest payments—the fiscal accounts had also strengthened. To some extent, the program targets may have been misconceived in the first place, owing to the very poor quality of the data and to the quite different conceptions of the problem and the underlying economic model held by the Fund staff and the authorities. To some extent, the implementation of the program was weakened by the authorities’ conviction that inflation could not (indeed, should not) be sharply reduced in the short term. And to some extent, the management of the Fund overestimated the ability of the banks to deliver. Even with the strongest adjustment effort, the program very likely would have gone off track owing to delays in obtaining the promised financial support from the banks (Boughton, pp. 344–5).
Boughton concludes this chapter with the Chilean case, which differed appreciably from those of the three big countries. First, Chile had embarked on free-market reforms a decade before the debt crisis. Second, its external debt was owed mainly by Chilean banks, not by the government, with the result that Chile experienced a disastrous banking crisis. There was indeed a striking resemblance between the Chilean crisis of 1982 and the Thai crisis of 1997.
Chile began to experience serious problems well before the Mexican crisis. The Chilean peso had been pegged to the dollar in 1979 with the aim of reducing inflation. Although inflation fell thereafter, there was a large real appreciation of the peso, as in the Argentine case, and the current account deficit widened hugely. In mid-1982, the peso was devalued, but not by enough to stem capital flight, and the peso was allowed to float on August 5, a week before the outbreak of the Mexican crisis. Boughton describes the consequences in language resembling a standard account of the 1997 Thai crisis. The continuing depreciation coincided with crumbling domestic demand and put unbearable pressure on the weak banking system. Domestic firms had taken out U.S. dollar-denominated bank loans while the peso was pegged, and they could not service them after it depreciated. Therefore, the banks could not service their debts to foreign banks without borrowing more dollars, and they could not continue to borrow after the Mexican crisis erupted. Rather remarkably, Chile persuaded foreign banks not to run down their claims on Chilean banks, but that did not suffice to stave off a banking crisis. In January 1983, Chile declared a bank holiday, liquidated three major banks, and intervened in the operations of five others.
The rest of the Chilean story resembles those told earlier about the three big countries, with one interesting difference. When Chile breached the terms of its initial IMF program, foreign banks became reluctant to put up new money or reschedule old loans. Instead of replacing the IMF program, however, the Fund staff negotiated a “shadow program” aimed at helping the Chilean authorities reach agreement with the foreign banks. The banks, in turn, asked that the Fund promise to reactivate Chile’s IMF program if Chile adhered to the shadow program, and de Larosière agreed to reinstate the program once the banks had reached a firm agreement with Chile and had disbursed the first tranche of a new money loan. In July 1983, a Fund mission found that Chile was meeting the terms of the shadow program, the banks signed on to the loan agreement, and the Fund reinstated Chile’s original program.
Boughton devotes four more chapters to the debt crisis. The first describes the time-buying effort at crisis containment in 1983–85. The second discusses the Baker Plan of 1985–87, aimed at reviving growth in the debtor countries. The third describes the shift to debt relief in 1987–89 and the implementation of the Brady Plan. The fourth assesses the whole debt strategy. In each of the first three chapters, moreover, Boughton combines his account of the evolving debt strategy with an account of the way that the crisis played out in Argentina, Brazil, and Mexico and of the Fund’s dealings with those countries. He does discuss events elsewhere—how the Fund tried to apply “enhanced surveillance” to Yugoslavia, for example, and how debt reduction came to Bolivia, Costa Rica, and Venezuela. The three big countries, however, occupy much space and in a way that is doubly disconcerting. Each country’s story is told in installments, one in each chapter, making the reader retrieve from memory the relevant bits of the previous installment, and Boughton’s account of the Fund’s dealing with those three big countries interrupts his account of the way in which the whole debt strategy evolved.
I cannot claim to know what Boughton should have done instead. Clearly, the Fund’s dealings with Mexico deserve detailed treatment. Many innovations in the debt strategy were worked out first for Mexico. That is true even of the debt relief phase, although Mexico was not the first debtor to secure a Brady Bond deal. There is less reason, however, for dwelling at similar length on Argentina and Brazil and their various policy failures. A strong case can be made, moreover, for saying more about some of the other debtor countries, including Peru, which unilaterally limited its debt-service payments to 10 percent of its foreign-exchange earnings and thus ran up arrears to the Fund and to other creditors. Boughton tells the Peruvian story in his chapter on arrears, where he departs from his usual practice of treating country cases as examples and tells the story of each country that ran up arrears. Some of those stories deserve to be told, especially the story of the Fund’s troubled and troubling relations with Zaïre during the Mobutu kleptocracy, but they are half buried in the chapter on arrears.
Containing the Crisis: 1983–1985
Having described the onset of the debt crisis and its initial manifestations in Argentina, Brazil, and Mexico, Boughton explains how the crisis was contained in 1983–85. This was a time-buying effort, aimed at restoring normal relations between the debtor countries and their foreign creditors. By pursuing rigorous adjustment policies, the debtors would be able to reduce their need for massive amounts of new money and, equally important, induce their foreign creditors to roll over their existing claims voluntarily. No one wanted or expected a revival of the massive borrowing that had occurred in the decade after the first oil shock. What was wanted and expected was a return to modest amounts of voluntary lending and, of course, a resumption of growth in the crisis-stricken countries, as it would reduce the relative magnitude of their existing debts and of their debt-service payments. The banks also needed time to strengthen their own balance sheets, so as to protect themselves from any future breakdown of the debt strategy.
Unfortunately, there was no growth in the crisis-stricken countries, and there was therefore no reduction in the weight of the debt burden. In fact, the time-buying strategy, involving as it did the de facto capitalization of the debtors’ interest payments, increased their debt burden. In 1982, the external debts of net-debtor developing countries totaled US$781 billion, of which US$390 billion was owed to foreign banks; by 1988, the total had risen to US$1,162 billion, of which US$484 billion was owed to the banks.5 In the interim, however, the banks strengthened their balance sheets, making way for the final phase of the debt strategy—the shift to debt reduction in 1988.
During the time-buying phase, however, there were modifications in the initial debt strategy. The first was the move to multiyear rescheduling agreements (MYRAs). At the start of 1984, Wilfried Guth of Deutsche Bank suggested to de Larosière that it was time to consider a longer-term approach to Mexico’s financing needs. Large amounts of Mexican debt would mature in 1985 and 1986 and would have to be rescheduled. At first, Boughton writes, de Larosière was skeptical, believing that the banks would be reluctant to make firm financial commitments extending beyond the end date of Mexico’s IMF program. But he came up with a way for the Fund to overcome the banks’ reluctance—a procedure that came to be known as “enhanced surveillance,” under which the Fund would monitor informally a country’s policies and performance, much as it would if the country had a conventional Fund-supported program. In June 1984, de Larosière suggested that the banks consider a MYRA for Mexico, which looked at the time to be moving toward an eventual resolution of its debt problem. He did not commit the Fund to undertake enhanced surveillance; Mexico would have to decide the form of the Fund’s future involvement.
The banks were divided initially. Some banks feared that debtor countries such as Argentina might seek similar treatment although they had made much less progress than Mexico. Other banks saw merit in enhanced surveillance, and their view prevailed. The banks undertook to negotiate a MYRA for Mexico once Mexico and the Fund had given operational meaning to the notion of enhanced surveillance. The Mexican authorities did not want to negotiate a shadow program like the one devised for Chile, as it would involve a formal review by the Executive Board. They also had reservations about any arrangement that would breach the confidentiality normally surrounding the Fund’s consultations with a member, and the Fund’s staff had similar reservations. But that issue did not have to be resolved immediately, because Mexico still had a Fund program in place. It was therefore agreed that the Fund would conduct semiannual consultations with Mexico, that the Board would review the consultations, and that results of the review would be transmitted to the banks in a manner not yet specified. The banks then began to negotiate a MYRA with Mexico. It would cover nearly US$50 billion of debt, including debt that was not due to mature until 1990. Maturities were lengthened, the repayment schedule was smoothed, and interest rate spreads were reduced.6
During this period, moreover, the Fund began to back away from its insistence on concerted lending, because it posed serious problems. First, “it gave the banks a virtual veto over the approval and financing of adjustment programs” and they sometimes insisted on significant changes in those programs, delaying approval and implementation (Boughton, pp. 406–7). Second, concerted lending was ineffective in small-country cases, as the banks had little at stake. The Fund began to rely instead on two fallback strategies: In some cases, it asked merely for an assurance that the banks collectively had reached an agreement with a debtor country, rather than insisting that a “critical mass” of individual banks subscribe formally to the agreement before the Fund’s Board would approve a Fund program. In other cases, the Board approved a program in principle but withheld disbursements until other financing arrangements had been completed. These fallback strategies, however, had another defect. On several occasions, “final agreements with banks took a year or more to conclude after the Fund arrangement was in place. Hence many of the negotiated programs were underfinanced, and the countries were forced either to intensify their policy adjustments or seek waivers and modifications to the original terms” (Boughton, p. 404).
Trying to Revive Growth: 1985-1987
By 1985, it was clear that the debt strategy was not achieving all of its objectives. The time-buying strategy had greatly reduced the risk of a systemic financial crisis. First, the banks had built up liquid reserves against potential losses. By 1986, indeed, banks in continental Europe had made loan-loss provisions averaging more than 20 percent of their claims on developing countries with debt problems; by 1987, moreover, U.S. banks had begun to provision heavily. Second, the banks had begun to reduce their exposure to the heavily indebted countries, despite their involvement in concerted lending and other commitments made in conjunction with IMF programs. Net bank lending to those countries dropped sharply in 1985, and it turned negative in 1986, by which time there was little new lending in conjunction with Fund programs. Third, the banks had developed a secondary market in sovereign debt, which gave the smaller banks a way to reduce their claims on the debtor countries.
Some debtor countries had also made progress. A few had been able to negotiate MYRAs that stretched out and smoothed their amortization schedules, and some were coming closer to external stability. In too many cases, however, they had reduced their current-account deficits by import compression, not export expansion. Hence, output and employment were not growing, but social unrest was rising, especially in the less successful countries. A new initiative was needed.
There were, of course, two ways to go: abandoning the basic premise of the initial strategy that debts must be repaid in full and espousing debt reduction; or reaffirming the basic premise but providing more financing and refocusing adjustment on structural reforms aimed at reviving growth. But there was not yet any official support for outright debt relief, not even in the debtor countries. Some debt-reducing schemes had already been devised, and there were occasional calls for the debtor countries to unite in support of a common approach to limit their debt-service payments.7 But the larger debtor countries showed no apparent interest in debt reduction or in a so-called debtors’ cartel; the orthodox aim of restoring normal market access was still viewed as the right way to go. Therefore, the official community tried in 1985 to reinforce the initial strategy by providing more financing and refocusing adjustment. That was the Baker Plan. Debt reduction did not come until 1989. That was the Brady Plan. The fact that each plan bears the name of a U.S. Secretary of the Treasury testifies to the obvious—that the debt strategy could not be altered without the support of the United States.
James Baker unveiled his plan at the Bank-Fund Annual Meeting in October 1985. It was designed to address the “lending fatigue” of the banks and the “adjustment fatigue” of the debtors (Boughton, pp. 417–8). Large numbers of small banks were resisting calls for new lending, and the practice of treating all banks equally was delaying the completion of financing packages. Furthermore, the large banks believed that they were having to shoulder too much of the burden relative to official creditors. In the debtor countries, moreover, political leaders were finding it increasingly hard to justify the need to run trade surpluses and use large amounts of their countries’ export revenues to make interest payments to foreign creditors. Finally, within the IMF, there was a growing belief that Fund programs should include structural measures aimed at fostering growth, although that conviction was tempered by two concerns—that the contribution of structural reform to economic growth had not been clearly identified and that, in any case, structural reform was the job of the World Bank and regional development banks, not the IMF.
What did Baker propose? First, debtor countries should adopt “comprehensive macroeconomic and structural policies” to promote growth and balance of payments adjustment and to reduce inflation. Second, the Fund should continue to play a central role “in conjunction with increased and more effective structural adjustment lending” by the multilateral development banks. Third, commercial banks should increase their lending in order to support comprehensive adjustment programs. And Baker went on to quantify the second and third objectives. The World Bank and Inter-American Development Bank should raise their disbursements to the principal debtor countries by about 50 percent, and commercial banks should lend an additional US$20 billion over the next three years.
There were then long discussions in the Fund about implementation of the Baker Plan and its implications for the IMF. The staff continued to express reservations about embarking on efforts to foster growth-oriented adjustment, but the Executive Board did not. It agreed that Fund-supported programs should include supply-side measures to foster domestic saving and promote investment and that the Fund should collaborate more closely with the multilateral development banks.8 The commercial banks expressed reservations initially about the Baker Plan but soon changed their tune. Their subsequent support, however, was based on the belief that they would lend jointly with the World Bank and that their risks would thus be covered by some sort of guarantee. When disabused of that belief, their enthusiasm waned. Nevertheless, the plan went forward.
Did the Baker Plan succeed? Boughton gives two answers. The multilateral development banks came quite close to providing the required amounts of financing, but there is disagreement about the commercial banks’ performance. The Fund’s staff found that the banks did not provide any net lending in 1986–88, whereas one well-known outside study cited by Boughton found that they lent an additional US$18 billion over the three-year period (Cline, 1995, p. 209). The Fund’s figure, however, represented the net change in the banks’ claims on the 15 debtor countries featured in the Baker Plan, whereas other studies, including the one just cited, looked at lending flows. The difference between them is due largely to the banks’ debt-reducing operations—debt sales, swaps, write-downs, and buy-backs.9 But Boughton concludes that the Baker Plan failed, in that the 15 countries did not return to robust growth (Boughton, p. 429).
Boughton goes on to discuss the Fund’s relations with the three big countries during the years of the Baker Plan and describes their attempts to quell inflation: the introduction by Mexico of the so-called pacto, under which wages were set by tripartite agreements involving the government, unions, and employers and were based on projected inflation rather than backward-looking indexation; the “heterodox” Cruzado Plan adopted by Brazil in 1986, which introduced a new currency tightly pegged to the dollar and imposed a temporary price freeze and looser controls over wages, but did not attempt to reduce the public-sector deficit; and the equally heterodox Austral Plan adopted by Argentina in June 1985. He also discusses Brazil’s sudden decision to halt interest payments on its medium- and long-term debt to the foreign banks and the complex efforts needed to unwind it later.
Turning to Debt Reduction: 1987-1989
Nicholas Brady assumed a major role in reshaping the debt strategy weeks after he became Secretary of the Treasury in August 1988. The plan that bears his name, however, reflected a process that began much earlier.
By 1987, many observers were convinced that the debt strategy had to be more forcefully separated from the interests of commercial banks. The most heavily indebted countries might never be able to shed their burden without substantial relief from the contractual obligations they had undertaken in the carefree days before 1982. Moreover, most banks had already dodged the threat of bankruptcy, the risk of a systemic collapse was long past, and bank creditors no longer had a clear incentive to increase their exposure to developing countries (Boughton, p. 478).
There was opposition to this new view in the official community, especially from U.S. officials, and the merits of debt relief were debated vigorously within the academic community and within the IMF. This is indeed one of the rare instances in which Boughton describes disagreement within the institution. In 1987, he says, an intense debate began, involving members of the Research Department and the Exchange and Trade Relations Department (ETR). Some staff members believed that debt relief would benefit the creditors, not the debtors, and would delay the return to normal relations between them; others believed that the large stock of debt was an insuperable obstacle to the revival of voluntary lending (Boughton, p. 481).10
There was, however, significant movement toward debt relief in 1987. First, the Fund began to encourage the use of a menu of financing options, including debt-equity swaps and exit bonds, in lieu of concerted lending. Second, several major banks, led by Citibank, announced that they were building up their loan-loss reserves. This step was widely seen to mean that the major banks would stiffen their resistance to concerted lending. As Boughton points out later, however, it also enhanced their ability to engage in debt-reducing operations (Boughton, p. 544). Third, three heavily indebted countries, Bolivia, Costa Rica, and Mexico, sought to achieve debt relief on their own.
Late in 1986, Bolivia offered to buy back part of its debt at a deeply discounted price, using cash contributed by donor governments. As the plan moved forward, the price of Bolivian debt began to rise in the secondary market. Earlier, it had sold at 6 cents on the U.S. dollar; on the eve of the buyback in January 1988, it was selling at 11 cents, and this became the buyback price. As a result, Bolivia reduced the face value of its debt by US$240 million at a cash cost of only US$26. But the buyback had little effect on the market value of Bolivia’s debt. The increase in the market price had offset the reduction in the quantity of debt outstanding. Therefore, Bulow and Rogoff (1988) concluded that Bolivia had gained nothing from the buyback. Their argument, however, rested implicitly on the supposition that the market value of a country’s debt represents the burden of that debt, viewed from the debtor’s standpoint, and that is debatable. The market value of a country’s debt may well reflect the creditors’ expectations about the prospects for repayment, but it would be wrong to assume that the Bolivian government took the same probabilistic view. Most debtor countries tried very hard to honor their obligations fully because of the risk that default would bar future borrowing.
At about the same time that Bolivia made its offer, the Fund’s staff concluded that Costa Rica needed debt reduction. Soon thereafter, Costa Rica proposed what the Fund’s staff viewed as a realistic plan: the country’s whole debt would be rescheduled over 25 years, and interest payments would be capped at 1 ½ percent of Costa Rica’s GDP. But the banks’ advisory committee rejected the plan because of the precedent it might set, and the resulting deadlock lasted for two years. When debt relief came into vogue, however, Costa Rica made a new proposal, and negotiations resumed. The result was not a full-fledged Brady Plan deal of the sort described below, but it provided substantial relief. Costa Rica bought back part of its debt at 16 cents on the dollar and converted the rest into bearer bonds.
At the end of 1987, Mexico announced an ambitious effort to reduce its debt. In a deal worked out with Morgan Guaranty Bank, it offered to convert part of its debt into long-term bonds, which would be issued at a discount determined by an auction. The principal of the bonds would be guaranteed; the U.S. Treasury would issue zero-coupon bonds to Mexico, which would in turn deposit them with the Federal Reserve Bank of New York. There was no guarantee of the interest payments, but it was hoped that Mexico’s record of on-time interest payments would make the bonds attractive (Boughton, p. 490). The results, however, were disappointing. Fewer than 100 banks took up the Mexican offer, and the size of the discount, at 30 percent, was smaller than expected. But the Mexican scheme strongly influenced the design of the Brady Plan.
In early 1988, Camdessus tried to interest U.S. officials in a plan to promote the securitization of debt, and the Fund’s staff produced a paper suggesting that the Fund might seek to facilitate the “sharing of the discount” on outstanding debt through buybacks and other transactions (Boughton, p. 483). These efforts failed. Nevertheless, support was building for some sort of debt relief. In the spring of 1988, Japan advanced a proposal, the Miyazawa Plan, aimed at furnishing debt relief by forgiving interest payments on part of the outstanding debt. It was discussed inconclusively by the G-7 governments and the Fund’s Interim Committee at its Berlin meeting in September 1988. Nevertheless, the Interim Committee gave convoluted endorsement to the principle of debt reduction:
While recognizing that new money continues to be of primary importance in financing packages for countries undertaking adjustment, but remains difficult to secure, the Committee agreed that the menu approach should be broadened further, including through voluntary market-based techniques which increase financial flows and which reduce the stock of debt without transferring risk from private lenders to official creditors.
Soon thereafter, moreover, U.S. officials began to develop a new debt strategy, and Brady described the main elements publicly in March 1989, shortly before presenting the plan to the next meeting of the Interim Committee. First, commercial banks should be asked to waive the sharing and negative pledge clauses in their loan agreements, so that no individual bank would be prevented from taking part in debt-reducing operations. Second, the IMF and World Bank should use some of their lending to finance debt-reducing plans, and they should provide additional financing to help debtor countries collateralize some of their interest payments—what Mexico failed to do in 1987. Finally, the Fund should reconsider its insistence on firm financing assurances before approving an IMF program and, going further, should be prepared to lend to a country even when it had built up arrears during its negotiations with its bank creditors.
When the Executive Board discussed the new strategy, objections were raised to the innovations that would involve the Fund directly—earmarking some of its lending to finance debt reduction, helping to collateralize interest payments, lending into arrears, and deciding how to identify countries that should be allowed to use Fund resources for debt-reducing operations. Boughton devotes four pages to a contentious Board meeting that wrestled with those issues far into the night. Thereafter, he explains how the Philippines, Mexico, and Venezuela obtained debt relief under the Brady Plan. He then ends his history of the debt crisis by completing his account of the Fund’s relations with Argentina and Brazil.
Grading the Debt Strategy
Boughton considers several objections to the debt strategy of the 1980s and dismisses most of them as misplaced or exaggerated.
Consider, first, the charge that the Fund mistook a solvency crisis for a liquidity crisis and thus pursued an inappropriate time-buying strategy that had the unfortunate effect of raising the stock of debt rather than reducing it. Boughton trots out the usual answer—that it is very hard to distinguish clearly between the two types of crises. In any case, he says, the problem facing the debtor countries was neither insolvency nor illiquidity. It was instead the sudden need to mobilize real resources in order to service their debts when they could no longer cover their large current-account deficits by additional borrowing. The amount of domestic adjustment required was not politically feasible in the time available.
This answer, however, raises another question: Why did the Fund underestimate the amount of adjustment required? Boughton explains that the Fund was unduly optimistic about the growth prospects of Latin America. Adjustment programs in the first phase of the crisis “were predicated on forecasts of a rapid resumption of economic growth that would gradually bring down debt service ratios to sustainable levels” (Boughton, p. 540; emphasis in original). That forecasting error, in turn, reflected the failure of the Fund—and of Latin American governments—to recognize “the breadth of structural reform that was necessary if macroeconomic stabilization was to be implemented without stunting growth.” Furthermore, strong domestic opposition prevented some countries from implementing the programs to which they had agreed, and the Fund could not freely acknowledge that risk when forecasting Latin American growth (Boughton, p. 541).
It is hard to quarrel with Boughton’s explanation for the Fund’s forecasting error. It should be noted, however, that faster growth would not have reduced the need for substantial adjustment at the start of the debt crisis, when it was no longer possible for the debtor countries to cover their current-account deficits by additional borrowing from the foreign banks. There was only one way to obviate the need for sudden, substantial adjustment—much more money from the Fund. It must also be noted that structural reform is a time-consuming process. It takes time to implement and more time to influence economic performance. Even if the need had been recognized, then, and the right reforms had been adopted, several years would have elapsed before faster growth had offset the debt-raising effects of the initial time-buying strategy.
There was, of course, another reason to opt for a time-buying strategy, and Boughton invokes it in his response to a different attack on the initial debt strategy—that it was chiefly designed to make sure that banks were paid interest on time. To be sure, the provision of new money and deferral of debt service afforded by rescheduling helped them make their interest payments. Nevertheless, the debtors had still to make net payments to the banks. Hence, critics suggest that the debtors would have been better off financially if they had defaulted. Had they defaulted, however, they could have triggered a collapse of the international banking system, which was not in their interest. Boughton stresses this point strongly elsewhere in his book, but he slides over it here to make a different, more controversial point. The adjustment required of the debtor countries was beneficial in its own right, because “maintaining debt service is one element of a strategy to prevent a slide into autarky” (Boughton, p. 542).
The record on policy reform, he says, supports this generalization. Countries that adopt unilateralist policies on debt service, such as Peru under Alan García and Brazil under José Sarney, tend to experience rapid economic deterioration, and countries that delay adjustment enjoy only brief short-term benefits (Boughton, p. 543). This argument makes me uncomfortable. It reinforces my concern that the roles of financing and adjustment were being reversed in the 1980s. Financing was used to buy adjustment in the belief that the Fund and creditor-country governments knew what was best for the debtor countries. I would have been happier had Boughton rested his case on the simpler assertion he made earlier—that all concerned, including the debtors, sought rightly to prevent a collapse of the banking system.
That defense, however, does not shield the Fund from another criticism—that the Fund was too slow to espouse debt relief once it had contained the immediate threat to the banking system. Boughton disagrees. The public stance of the Fund, he says, was that of an intergovernmental institution. It could not get ahead of political leaders in the creditor countries without provoking a backlash. He also reminds us that Camdessus began to urge the creditor countries to consider debt reduction in 1987 and that members of the staff, especially in the Research Department, were early advocates of debt reduction. But the Fund must bear some of the responsibility for the long delay. The real costs of the debt crisis would have been far smaller if the official community had moved directly from the initial time-buying strategy to the Brady Plan, without a four-year detour via the Baker Plan.
Boughton winds up his assessment by asking whether the adjustment strategy was appropriate for Latin America. No one, he writes, can object to the Fund’s basic premise that financial responsibility and stability require market-oriented policies, low fiscal deficits, and limits on the growth of domestic credit. “The issue is whether these prescriptions were applied rigidly in cases where they were not strictly appropriate” (Boughton, p. 549). Questions arise, he says, with regard to countries where the state played a large role in promoting development, so that market-oriented stabilization policies might have been ineffective; where inflation was due largely to structural or inertial forces, so that conventional stabilization might have imposed high costs; and where there was little confidence in the domestic currency or domestic financial system, so that the dismantling capital controls might have been destabilizing. As the Fund worked with a model that paid little attention to these possibilities, it had trouble devising stabilization programs that governments could implement in the face of domestic opposition. But Boughton claims that these problems were partially resolved during the 1980s. On the Fund’s side, there was increasing flexibility; it even helped to design heterodox adjustment programs, such as Argentina’s Austral Plan. On the governments’ side, the silent revolution weakened their belief in state-dominated development and the need for capital controls. It would perhaps be more accurate to say that the debt crisis was resolved by the Brady Plan and the resulting revival of capital inflows, which helped to create an environment in which the debtor countries could mobilize domestic support for the structural reforms required to catalyze self-sustaining growth.
IV. Conditionality, Structural Adjustment, and Concessional Lending
At this point, we are halfway through Boughton’s book but more than halfway through the space allotted for this review. Fortunately, some of the subjects he must cover in his comprehensive institutional history need not be covered here—the two reviews of Fund quotas in the 1980s; how the Fund borrowed to supplement its limited resources; the shrinking role of the SDR as a reserve asset and the unsuccessful attempt to revive it by replacing dollar reserves with SDR-denominated claims; the treatment of countries with large, long-lasting arrears to the Fund; the growth of IMF membership; the Fund’s relations with the World Bank; and the internal governance of the institution. Those stories have interesting episodes, including some that illustrate the strong political pressures to which the Fund was often subject. (See, for example, Boughton’s discussion of the Fund’s dealings with South Africa during the apartheid era, the problems it faced when the United States invaded Panama, its difficult dealings with Zaïre, and the way that it dodged a request that the Palestine Liberation Organization be granted observer status.) Three other stories, however, deserve more attention, because they have had enduring effects—how the Fund tightened and broadened conditionality during the 1980s, how it became heavily involved in concessional lending to the low-income developing countries, and how it used the leverage afforded by that lending to foster structural adjustment.
Tightening and Broadening Conditionality
In the second half of the 1970s, three-quarters of the Fund’s total disbursements involved little or no conditionality; in the first half of the 1980s, the fraction was only two-fifths; and it continued to fall in the second half of the decade (Boughton, p. 561). What explains this shift?
Part of the answer is simple. By the start of the 1980s, the Fund had already committed the money it had borrowed to finance the oil facilities set up to cope with the first oil shock, and few conditions were attached to drawings on those facilities. Hence, countries seeking to draw on the Fund in the early 1980s had to apply for stand-by arrangements that took them into the upper tranches of their Fund quotas, where they faced stiffer conditions. In 1979, at the time of the second oil shock, the Fund was under great pressure to lend more to the oil-importing developing countries, which were hit hard by that shock, and it was widely thought at the time that the Fund had loosened the policy conditions attached to upper-tranche drawings so as to meet those countries’ needs.11 But Boughton denies that there was a deliberate easing of conditionality. He notes, for example, that the Managing Director stressed the importance of conditionality in his meetings with the staff.12 Boughton does concede, however, that conditionality “was not as strong or effective as it later became” (Boughton p. 564).
But did conditionality become more effective? Boughton cites a staff study of 34 countries that had entered into conditional arrangements in 1983. By 1986, those countries had been involved in 71 stand-by or extended arrangements, many of which had been interrupted by the countries’ failure to meet program conditions. Only 7 of the 34 countries were deemed to be relatively close to viability; 6 others would have serious problems for several more years; and the other 21 might achieve viability within five more years, but some would have to make major policy changes (Boughton, p. 569). This sad record, however, tells us little about the quality of conditionality. Most of the countries, Boughton notes, began in dreadful circumstances, and some also suffered natural disasters. Furthermore, the mid-1980s brought a hostile environment—high real interest rates, falling commodity prices, sluggish export growth, and little if any increase in bilateral assistance from the industrial countries. Many countries, moreover, fell far short of meeting their policy commitments to the Fund.13 But the Fund had also to ask whether adjustment programs had been poorly designed, and this was debated at length by the Board. There were suggestions for reform, but they focused chiefly on ways in which the Fund monitored countries’ compliance with their policy commitments, not on the basic nature of those policy commitments. The Fund, says Boughton, was hemmed in by the need to balance conflicting objectives:
The Fund required a borrower to develop a program to restore external viability but encouraged it to do so in a way that would promote sustainable growth. The Fund wished to promote structural reforms while respecting the social and political choices of the country. The Fund tried to design programs that were politically feasible but that could be carried out quickly and strongly. The Fund tried to take account of the special circumstances facing each country while ensuring uniformity of treatment. The Fund sought to collaborate with the World Bank in recommending structural reforms while avoiding cross-conditionality. The Fund had to monitor the implementation of programs but tried to minimize interference with the authorities’ management of the economy. Changing the guidelines or the procedures would do little to resolve these conflicts (Boughton, pp. 571–2).
Although declining to rewrite the guidelines themselves, the Board intensified and broadened conditionality. Like one of Stephen Leacock’s characters, the Fund leapt on its horse “and rode madly off in all directions.”14
On the one hand, it increased the number of performance criteria used to monitor a country’s compliance with the traditional targets of a Fund-supported program—restraining domestic credit creation, limiting the size of the budget deficit, and forbidding the introduction or intensification of exchange or trade restrictions. During the 1980s, it introduced subceilings on various forms of credit creation, limits on various forms of foreign borrowing, restrictions on the accumulation of arrears to foreign creditors, and various rules pertaining to exchange rate policy, such as a floor beneath the net foreign assets of the monetary authorities to keep them from running down their reserves to defend an overvalued currency, or, alternatively, a rule requiring adjustments in the nominal exchange rate to offset domestic inflation. In 1983, all upper-tranche IMF programs had at least five performance criteria; by 1986, all of them had at least eight, and some had several more. There were 15 such criteria in the 1984 Yugoslav program (Boughton, pp. 603–5).
On the other hand, the Fund introduced structural conditions into several programs, because “lending directed to countries with deep-seated structural problems could no longer be evaluated solely in terms of macroeconomic stability” (Boughton, p. 588). The process began modestly in the early 1980s, when the staff encouraged countries to implement liberalizing reforms but did not usually insist that borrowing rights be conditioned on the implementation of those reforms. Even in 1986, after the adoption of the Baker Plan, which emphasized growth-oriented adjustment, de Larosière concluded that “the staff should continue to explore ways of monitoring structural reform outside the realm of formal performance criteria” (Boughton, p. 589; emphasis added).
Here, Boughton draws a distinction between structural reforms of the sort that the Fund had long favored, such as trade liberalization, and reforms more closely identified with a growth-oriented adjustment strategy, such as supply-side measures aimed at raising saving and encouraging investment. Both the staff and the Board had reservations about committing the Fund to the latter. Their reservations reflected the belief that these were matters best left to the World Bank, as well as a broader concern about the risk of real or apparent conflict between the immediate need to achieve balance-of-payments adjustment and the longer-term need to foster economic growth. As the Fund became more deeply involved with the low-income countries, however, it became harder to distinguish between traditional liberalization and growth-promoting reform. Unfortunately, Boughton says very little about the nature or number of the structural conditions that found their way into Fund programs during the 1980s or about the extent to which countries fulfilled those conditions. If, as Boughton says, structural reform was a principal manifestation of the silent revolution, we need to know more about the form it took and its impact on economic growth in the developing countries.
The tightening of conditionality had one more manifestation. The Compensatory Financing Facility (CFF) was “stripped of its usefulness” (Boughton, p. 724). The CFF was established in 1963 to furnish additional funding, with little conditionality, to countries facing temporary shortfalls in their export revenues because of circumstances beyond their control. It was designed chiefly for countries that depended heavily on exports of primary products, because those products’ prices are notoriously volatile.15 There were thereafter three liberalizations of the CFF, which raised the amount that a country could draw relative to its IMF quota. In the case of large drawings, however, the Fund had to be satisfied that the country was cooperating with the IMF to solve its balance-of-payments problem. Many countries met this requirement easily, because severely affected countries often sought stand-by arrangements with the Fund, and the approval of a stand-by was deemed to be prima facie evidence of cooperation. In other cases, however, it was harder to decide whether the requirement had been met.
Therefore, the Fund raised the bar in 1983. A country seeking to make a small CFF drawing would be visited by a Fund mission. If, in the light of the mission’s findings, the Fund concluded that the country’s policies were “seriously deficient,” the country would have to take corrective action before it could draw on the CFF. A country seeking to make a large drawing would have to convince the Fund that its balance of payments was satisfactory, apart from the effect of the export shortfall, or would have to qualify for a highly conditional stand-by arrangement. This change had two effects. First, it meant that countries could no longer count on speedy access to the CFF. Second, it meant that financing from the CFF might no longer be truly additional to other forms of IMF financing; a country seeking a stand-by arrangement in order to qualify for a CFF drawing might find that the staff had reduced the size of the stand-by to allow for the CFF drawing. The country would get more money up front but might get smaller amounts thereafter.16
The 1983 decision helps to explain why there was a sharp fall in subsequent use of the CFF, despite the huge drop in commodity prices during the 1980s. But there were additional reasons. The collapse of commodity prices was not temporary, and countries facing a permanent deterioration in their terms of trade had to adjust to that shock. Furthermore, the countries most strongly affected by the collapse of commodity prices were, in the main, low-income countries, and they were therefore eligible for low-cost financing from another Fund facility set up for their use.17
The Fund and the Low-Income Countries
In the 1970s, the Fund undertook to provide concessional assistance to the low-income developing countries. In 1975, it established an Oil Facility Subsidy Account, financed by contributions from 25 countries; it was used to reduce the interest cost of borrowing from the Oil Facility itself. In 1976, it established a Trust Fund, financed by some of the profits obtained from the Fund’s gold sales; it was used to make low-interest loans. By 1981, however, the assets of the Trust Fund had been fully committed. Furthermore, some of the loans had begun to mature, and the rest were due to mature during the next ten years. The Fund had therefore to make two decisions—how to replace the concessional lending that had come from the Trust Fund, and what to do with the repayments of the Trust Fund loans. It might, of course, have extended the life of the Trust Fund and thus turned the repayments around by making new Trust Fund loans. But there were other options—converting the old loans into grants, or using the repayments of Trust Fund loans to subsidize the interest costs of ordinary drawings on the IMF.18
An interim solution was adopted in 1981. Some of the repayments of Trust Fund loans were used to create a new subsidy account for use by low-income countries in reducing the interest costs of drawings on the Supplementary Financing Facility (SFF). Those interest costs were almost twice as high as the interest costs of ordinary IMF drawings, because the SFF was financed with borrowed money on which the Fund itself had to pay market interest rates. But the SFF would soon run out of money. Hence, there was need for a new way to recycle repayments to the Trust Fund. Two issues had to be resolved, however, before a new facility could be designed. Which countries should have access to it? How much conditionality should be attached to its loans? These issues were debated in 1985, at the same time that the Fund was debating its role in the implementation of the Baker Plan.
It was, of course, agreed that the new facility should be designed for the low-income countries, and the staff proposed initially that eligibility should be confined to those defined by the World Bank as “IDA-only” countries (i.e., those that were eligible for loans from the Bank’s concessional facility, the International Development Association, but not also eligible for conventional World Bank loans). This proposal would have excluded China and India, the largest potential users, which would have allowed all other countries to borrow more from the new facility. But it was unacceptable to China and India. The problem was finally resolved when those countries agreed voluntarily that they would refrain from using the new facility if they were not formally precluded from doing so.
The question of conditionality was harder to resolve, because the debate about it took place at a time when the Fund was tightening conditionality and was also trying to decide how heavily it should emphasize structural reform. The staff of the Fund recommended against replicating the terms of access to the Trust Fund because those terms of access had put financing ahead of adjustment at a time when the low-income countries were facing severe deterioration in the economic environment. As a result, many recipients of Trust Fund loans wound up worse off than they had been initially. Hence, the staff favored tighter conditionality. The United States, moreover, favored a strong emphasis on structural reform. In fact, the U.S. Executive Director, Charles Dallara, proposed that countries borrowing from the new facility be required to adopt two-year programs aimed at macroeconomic adjustment and structural reform and that the programs be devised jointly by the Fund, the World Bank, and the countries’ governments. Executive Directors from developing countries objected strenuously; they were worried about the risk of cross-conditionality (i.e., that one institution—the Fund or the Bank—might withhold a loan until a country had met the requirements of the other institution). Finally, there were disagreements about the way to monitor conditionality.
There was therefore a compromise. The Fund and Bank would collaborate with a country in drafting what came to be called a Policy Framework Paper. Thereafter, however, each institution would negotiate and monitor the terms of its own loan. The programs would emphasize structural reform, and “quantified policy benchmarks could be required as long as they were distinguished in some fashion from performance criteria” (Boughton, p. 650). The new facility would thus be called the Structural Adjustment Facility (SAF), and it came into being in 1986.
Although the SAF got off to a slow start, Boughton concludes that it was modestly successful. Many low-income countries did not apply for SAF loans or were unable to devise adequate adjustment programs. Over its ten-year life, however, 37 of the 61 eligible countries qualified for SAF loans. Yet ten SAF programs were abandoned before the second or third year, either because the adjustment programs had not been fully implemented or because the country had built up arrears to the Fund (Boughton, p. 654). Furthermore, the small size of the SAF kept it from making a major contribution to the financing needs of the low-income countries. Therefore, the new Managing Director, Michel Camdessus, began an ambitious campaign to raise more money, and Boughton has high praise for him. “His goal initially seemed hopeless to almost everyone, but not even he could have imagined how far the quest would carry him and how fundamental it ultimately would be to history’s judgment of his years in office and of the Fund’s ability to contribute to the alleviation of the world’s worst poverty” (Boughton, p. 663).
In May 1987, Camdessus met with the heads of state and senior officials of the G-7 countries, and they responded favorably. The communiqué issued one month later at the Venice Summit welcomed the Managing Director’s proposal for a “significant increase in the resources of the Structural Adjustment Facility for the three years from January 1, 1988” (quoted by Boughton, p. 664). Even before that, moreover, Camdessus had set a specific target—tripling the resources of the SAF in order to establish what came to be called the Enhanced Structural Adjustment Facility (ESAF). Boughton explains this ambitious goal:
In mid-1987, the 34 SAF-eligible countries in Africa alone faced scheduled repayments to the Fund totaling SDR 5.3 billion through 1990, approximately triple the amounts that the Fund could extend to them through the SAF over that same period. It was crucial both for the Fund and for the indebted countries that those repayments be made. A few countries already were in arrears to the Fund, and if that problem snowballed, then bilateral aid would dry up just when the wisdom of the Fund remaining involved in sub-Saharan Africa would be increasingly called into question. Without something close to a tripling of resources or an equivalent amount of other forms of aid, the economic plight of many very poor countries would be imperiled. It is no exaggeration to conclude that the economic future of Africa was at a critical juncture (Boughton, p. 665).
At several other points, Boughton repeats a point made in this passage—that the low-income countries needed concessional lending in order to repay the Fund. In the case of Uganda, for example, “the money it was getting from the SAF and CFF was not even sufficient to cover the payments that were still due on credits from the Fund” (Boughton, p. 682). In fact, Uganda even agreed to accept the initial disbursement of its ESAF loan in SDRs and maintain a significant balance in its SDR account in order to facilitate timely repayments to the Fund (Boughton, p. 683). In the case of Ghana, the 1998 ESAF loan “committed the Fund to maintain its loan exposure for several years and freed the authorities from having either to repay large sums or to negotiate a series of short-term arrangements” (Boughton, p. 678). The same wording, of course, could be used to describe the effects of a MYRA. That is why, at the start of this review, I compared the Fund’s concessional lending to the low-income countries with its efforts to mobilize new money from commercial banks in order to stave off default by the middle-income countries. Both were time-buying exercises, and both were concerned with the plight of the creditor as well as the plight of the debtor.
When it came time to consider the design of the new concessional facility, the Executive Board had to revisit some of the issues it had debated in connection with the SAF. The management and staff, however, had already concluded that inadequate conditionality had impaired the effectiveness of the SAF as a vehicle for structural adjustment. Therefore, the Board agreed that ESAF loans would be disbursed semiannually, not annually like SAF loans, and that performance criteria would be used to monitor compliance with both macroeconomic and structural conditions. But it was proving hard to raise the necessary money. The United States was reluctant to contribute, because it wanted to concentrate initially on securing congressional approval for the U.S. contribution to the eighth replenishment of the IDA. In the end, the Board decided to create two accounts—a Loan Account to receive loans from donor countries that would then be lent to low-income countries and a Subsidy Account to receive grants from donor countries that would then be passed on to low-income countries as interest subsidies. The United States contributed to the Subsidy Account, along with 23 other countries, but not to the Loan Account. The ESAF made its first loan in 1988 and became the principal source of concessional credit from the IMF. In fact, it would become the only source of IMF financing for the low-income developing countries.
How does Boughton assess the performance of the ESAF? It was, he says, “destined to become a success story of assistance to desperately poor countries in an era when such success was elusive.” Nearly 40 low-income countries would receive ESAF credits and “would have the opportunity to establish a track record of economic progress that would bring in bilateral assistance and in some cases private capital flows and direct investment” (Boughton, p. 684). Boughton concedes that donor countries may have reduced their contributions to the replenishment of the IDA, but says that most of them tried to avoid offsetting reductions in bilateral aid programs. Yet Boughton does not confront the issue raised by his own history—whether the recipients of ESAF loans would have been better off if the money supplied by the donor countries had been used to repay directly the recipients’ debts to the Fund. Presumably, Boughton would say “no” because he believes that ESAF loans have brought important reforms in the borrowing countries, whereas outright debt relief might not have done that, even if countries were made to meet demanding preconditions before they could qualify for debt relief. We may not know the answer to this question, however, until we can assess the long-term effects of the current HIPC initiative, which aims to reduce the debt burdens of the heavily indebted poor countries.
It is not too early, however, to raise questions about the Fund’s ambitious attempt to force the pace of structural reform in developing countries and Boughton’s strong endorsement of it. I have argued elsewhere (Kenen, 2001) that the Fund’s insistence on far-reaching reform during the Asian crisis of the 1990s may have made the crisis worse by convincing panicky creditors that the crisis-stricken countries could not begin to recover unless they dealt swiftly and decisively with all of their deep-seated problems. The Fund was right to insist that some things be done promptly. Insolvent banks had to be closed or recapitalized, and the problem of corporate debt had to be addressed although it could not be solved quickly. The Fund went too far, however, by insisting on many other reforms, such as removing restrictions on the foreign ownership of domestic banks, abolishing domestic monopolies, and liberalizing trade. Reforms of that sort could do little in the short run to reduce the cost or duration of the Asian crisis, and we cannot claim to know what they could do in the long run to foster the further success of the Asian economies. Governments cannot be expected to take ownership of policies that are not clearly efficacious.
Fortunately, a not-so-silent reformation has followed the silent revolution. Under the recent revision of the Fund’s Guidelines for Conditionality, policy conditions will be formulated parsimoniously and will normally focus on macro-economic variables and structural measures that are within the Fund’s core areas of responsibility: macroeconomic stabilization; monetary, fiscal, and exchange rate policies; and financial system issues. Excursions into other policy domains will not occur unless they are critically important. This is surely the right way to refocus the Fund.
BoughtonJamesM.2001Silent Revolution: The International Monetary Fund 1979-1989 (Washington: International Monetary Fund).
BulowJeremyM. andKennethRogoff1988“Buyback Boondoggle,”Brookings Papers on Economic Activity: 2Brookings Institution pp. 675–704.
ClineWilliamR.1995International Debt Reexamined (Washington: Institute for International Economics).
de VriesMargaretG.1976The International Monetary Fund 1966-1971: The System Under Stress (Washington: International Monetary Fund).
de VriesMargaretG.1985The International Monetary Fund 1972-1978: Cooperation on Trial (Washington: International Monetary Fund).
DooleyMichaelP.1989“Market Valuation of External Debt,”in Analytical Issues in Debted. byJacobA.FrenkelMichaelP.Dooley andPeterWickham (Washington: International Monetary Fund).
FrenkelJacobA.MichaelP. Dooley andPeterWickhameds.1989Analytical Issues in Debt (Washington: International Monetary Fund).
HorsefieldKeithJ.ed.1969The International Monetary Fund 1945-1965: Twenty Years of International Monetary Cooperation (Washington: International Monetary Fund).
KenenPeterB.1990“Organizing Debt Relief: The Need for a New Institution,”Journal of Economic PerspectivesVol. 4 (Winter) pp. 7–18.
KenenPeterB.2001The International Financial Architecture: What’s New? What’s Missing? (Washington: Institute for International Economics).
KraftJoseph1984The Mexican Rescue (New York: Group of Thirty).
KrugmanPaulR.1989“Market-Based Debt-Reduction Schemes,”in Analytical Issues in Debted. by Jacob A.FrenkelMichaelP. Dooley andPeterWickham (Washington: International Monetary Fund).
LeacockStephen1911“Gertrude the Governess,”in Nonsense Novels (London: John Lane).
PolakJacquesJ.1957“Monetary Analysis of Income Formation and Payments Problems,”IMF Staff PapersInternational Monetary FundVol. 6 (November) pp. 1–50.
PolakJacquesJ.1991“The Changing Nature of IMF Conditionality,”Essays in International Finance No. 184 (Princeton: International Finance Section, Princeton University).
Ul HaqNadeem andMohsinS.Khan1998“Do IMF-Supported Programs Work? A Survey of the Cross-Country Empirical Evidence,”IMF Working Paper 98/169 (Washington: International Monetary Fund).
WilliamsonJohn1983“The Lending Policies of the International Monetary Fund,”in IMF Conditionalityed. by JohnWilliamson (Washington: Institute for International Economics).
In statistical matter throughout this issue,
- dots (…) indicate that the data are not available;
- a dash (—) indicates that the figure is zero or less than half the final digit shown, or that the item does not exist;
- a single dot (.) indicates decimals;
- a comma (,) separates thousands and millions;
- “billion” means a thousand million; and “trillion” means a thousand billion;
- a short dash (–) is used between years or months (for example, 1998–99 or January–June) to indicate a total of the years or months inclusive of the beginning and ending years or months;
- a slash (/) is used between years (for example, 1998/99) to indicate a fiscal year or a crop year; and
- components of tables may not add to totals shown because of rounding.
The term “country,” as used in this publication, may not refer to a territorial entity that is a state as understood by international law and practice; the term may also cover some territorial entities that are not states but for which statistical data are maintained and provided internationally on a separate and independent basis.
Design: Luisa Menjivar-Macdonald and Sanaa Elaroussi