Nonmember Territories

In a few cases, the Fund held regular consultation discussions with nonmember territories following voluntary agreements with members. The prototype for this activity was an agreement with the Netherlands, beginning in 1970, to conduct Article VIII consultations with the Netherlands Antilles.²³ The Antilles (then comprising Aruba, Curaçao, and four other Caribbean islands) was an integral political part of the Kingdom of the Netherlands but had a distinct and autonomous economy based largely on tourism and petroleum. It also had its own currency, which was pegged to the U.S. dollar, not to the Netherlands guilder. A local political crisis in 1969 did considerable damage to the economy and to the balance of payments. That induced the Dutch government to ask the Fund to consult with the authorities in the Antilles and offer policy advice.²⁴ From then on, the Fund conducted regular consultations with the Antilles, first under Article VIII and later (after the Second Amendment took effect) under Article IV.

In 1990, the Fund entered into a similar agreement with the United Kingdom for consultations with Hong Kong. The government of the United Kingdom controlled the territory through a 99-year lease from China, scheduled to expire in 1997.²⁵ In April 1990, the British and Chinese authorities ratified an agreement for the return of Hong Kong to China when the lease ran out. Although Hong Kong was a major hub for international finance, its status as a British territory meant that its relations with the IMF had been limited. Beginning in 1979, the Fund included some data in the WEO. From 1987, it provided technical assistance, mainly on balance of payments accounting. Two years later, the British and Chinese teams negotiating terms for the handover agreed that Hong Kong would “continue to participate in the activities of the IMF” after the change in sovereignty. That decision enabled the IMF to begin collecting data more systematically and to initiate annual consultations and other contacts.²⁶

The first Article IV consultation discussions with Hong Kong took place in October 1990, led by Bruce J. Smith (Assistant Director, Asian Department). The staff report praised the authorities’ impressive record of economic success, while offering some low-key advice to strengthen balance of payments statistics, watch out for spending excesses, and consider implementing a value-added tax.²⁷ The staff recommended placing Hong Kong on the bicycle, but the authorities asked for—and got—annual consultations.²⁸

Regional Surveillance

Another approach the staff suggested in 1990 was to make increased use of regional surveillance. For groups of countries with close trade and other economic and financial ties, and especially for those with common currency arrangements, discussing them as a group would bring two benefits. The main advantage would be to generate more meaningful analysis of policy options. In some situations, regional surveillance would also yield economies of scale for the use of time by both the staff and Executive Directors. Groupings that seemed particularly apt included the European Communities, the CFA franc zone in Africa, and the Eastern Caribbean Currency Union.²⁹ The Executive Board cautiously endorsed that approach, though with the proviso that it should be primarily an analytical exercise and not a substitute for bilateral surveillance with individual member countries.

As detailed in Chapter14, the Fund began this sort of regional analysis with a study of the exchange arrangements of the CFA franc zone. That study eventually contributed to the regional understanding essential for successfully managing the devaluation of the CFA franc in January 1994. Soon afterward, when war broke out in the Middle East in response to the invasion of Kuwait by Iraq, the Fund undertook a study of the likely economic effects on the region. That study helped inform an interagency effort to provide advice and financial assistance to the most severely affected countries.³⁰

Subsequent instances of regional surveillance in the 1990s, with the exception of the European issues discussed below, were mostly singular discussions of regional groupings or discrete events with important regional effects. Examples included coverage of regional policy issues in the West African Economic and Monetary Union in 1998; the establishment of the Central African Economic and Monetary Community in 1999; developments in the Eastern Caribbean Currency Union in 1998–99; postmortem analysis of the Asian financial crises of 1997–98; and studies of the economic and policy implications of regional trade agreements.³¹ The opening of the Regional Office for Asia and the Pacific in Tokyo in 1997 also symbolized the Fund’s recognition of the need for a wider focus in that region. And in September 1998, the Fund hosted a special meeting of Latin American finance ministers and central bank governors to develop a regional approach to containing the effects of the financial crises that had already spread from East Asia to the Russian Federation (see Chapter12).

One of the most important and continuing examples of regional surveillance was with regard to the monetary integration of western Europe. The ad referendum signing of the Maastricht Treaty in December 1991 set in motion an integration process aimed at establishing full Economic and Monetary Union (EMU) by 1999. The staff of EUR monitored developments over the next year and then prepared a study of the regional and global implications that was intended to be discussed by Executive Directors around the end of August 1992. The paper noted that the transition from national formulation and implementation of monetary policy to joint implementation by participating European countries would take several years but ultimately would have profound effects on Europe, its trading partners, and the Fund. It proposed that the Fund begin to conduct a regional “Article IV type” surveillance consultation with the new regional monetary authorities (primarily the European Monetary Institute, or EMI, which the Maastricht Treaty would establish as a transitional body to the permanent European Central Bank, or ECB).³²

Unfortunately, the Executive Board’s discussion of the EMU paper was postponed in August 1992.³³ When the exchange rate mechanism (ERM) of the European Monetary System (EMS) came under speculative pressure in September and the British and Italian currencies were forced out, the hole in Fund surveillance was glaringly obvious. Coincidentally, European finance officials were in Washington for the IMF/World Bank Annual Meetings, and some of their negotiations took place either at Fund headquarters or at the convention hotels. The Fund itself, however, was kept on the sidelines, not only by the long-standing reluctance of European officials to seek the Fund’s advice, but by the lack of a clearly articulated institutional position on EMU. Only after the crisis had passed, in December 1992, did the Executive Board finally hold its postponed seminar.

That seminar led to general agreements that the Fund should hold regular discussions of EMU during the transition, that the Fund broadly supported the process despite some misgivings about the wisdom of defending exchange rates between countries with different trend rates of inflation, and that formal consultations under Article IV had to be restricted to sovereign member countries.³⁴ Over the next year, as turmoil continued in the ERM, the WEO and the periodic WEMD sessions became the main vehicles for these discussions. Twice in 1994 and once in 1996, the Board held more formal discussions of the broader implications of European monetary integration. The staff papers for these meetings stressed the potential benefits of EMU but also highlighted the political and technical hurdles that remained to be cleared before the process could be brought to a successful conclusion.³⁵

When Executive Directors met to discuss the subject in March 1996, those hurdles were daunting. Almost none of the candidate countries for EMU met the criteria spelled out in the Maastricht Treaty, and the region was in a recession that appeared to be caused or aggravated by attempts to get fiscal deficits and debt levels down to qualifying levels. A few Directors from outside the European Union (EU) found the prospects worrying, and some were skeptical of whether EMU was really feasible on the envisaged schedule. Shakour Shaalan (Egypt) kicked off the debate by pleading for a delay in the process and expressing doubts about whether a single currency was even in Europe’s best interests. Daniel Kaeser (Switzerland) joined that argument by noting that “adjusting the timetable … [would put the] future European monetary policy, from its inception, on a more solid and thus more robust basis.” Even Directors from EU member countries expressed concerns about the convergence criteria pushing up unemployment rates across the continent. The prevailing view, though, was that no matter how difficult the process turned out to be, its successful conclusion was critically important for Europe and for the rest of the world.³⁶

In March 1997, as the adoption of the Stability and Growth Pact was approaching, the Fund hosted a conference at which a number of outside experts debated the benefits and costs of Europe proceeding with EMU as planned. Despite considerable doubts on the part of many speakers, the general atmosphere was supportive.³⁷ In April, the Interim Committee “welcomed the progress made toward establishing conditions for EMU, the creation of which is one of the most important international monetary developments in the post–Bretton Woods period.” The communiqué called on the Fund to “undertake a broad program to assess the implications of EMU for the international monetary system and for the Fund.”³⁸

From that point on, analysis of all aspects of EMU was an integral part of Fund surveillance. The staff was meeting regularly with counterparts in the EMI. When the ECB replaced the EMI in January 1999, the Fund granted it observer status in the Fund, with the right for its representative to attend relevant meetings of the Executive Board.³⁹ More broadly, the Fund agreed in 1998 that once the EU adopted a common currency (the euro) in January 1999, consultations on monetary and exchange rate policies should take place at the regional level, while Article IV consultations with euro area countries “should focus on fiscal, financial, and structural policies.”⁴⁰ Thus, for the first time, the Fund adopted a specific policy establishing regional surveillance as an ongoing function.

Global Standards

Establishing good medium-term goals and policies was more a political than an economic task. The staff’s technical job was to assess the economic consequences of the political decisions. Consequently, after the 1992 review of surveillance, Camdessus and the Chairman of the Interim Committee (Philippe Maystadt) decided to seek the Interim Committee’s support in defining standards for good policies. That led to a three-year progression of increasingly detailed declarations by the Committee that culminated in what Camdessus later liked to call the “eleven commandments.”

The initial effort was the Interim Committee’s “Declaration on Cooperation for Sustained Global Expansion,” adopted at the April 1993 meeting in Washington.⁴³ This declaration called on all countries and the major multilateral institutions “to join forces in a global cooperative effort to bolster confidence and strengthen prospects for a durable, noninflationary world expansion.” (Emphasis is in the original text.) Specifically, the declaration stressed the importance of a multilateral effort to promote free trade, including a successful conclusion to the Uruguay Round of trade negotiations; continued reform of the transition economies; improved and better-coordinated economic policies in industrial countries, including fiscal consolidation over the medium term and structural reforms to improve the functioning of markets; and an updating of IMF lending facilities, including for concessional lending to low-income countries. The declaration reaffirmed the IMF’s role as “the central international monetary institution.”

This statement of principles was expressed in very general terms. Limited as it was, at that time the 1993 declaration was the most concrete international commitment to promoting market-oriented structural policies, and the most ambitious attempt to define good macroeconomic policies. It thus gave a global blessing to what had become known as the “Washington Consensus.”⁴⁴

The next step was the adoption of the Madrid Declaration at the October 1994 meeting of the Interim Committee in Madrid, Spain. That document asserted that a “strategy based on steadfast implementation of strong programs of macroeconomic adjustment and structural reform” had succeeded in “many developing economies” and should be adopted more widely. To enhance the prospects for success, it called on industrial countries to improve the “global environment” by strengthening their own macroeconomic policies, opening their markets to developing-country exports, reducing the debts of low-income countries, and increasing development assistance.⁴⁵ The Madrid Declaration offered more specific guidance than did its predecessor. That guidance focused mainly on the specific issues the world economy faced at the time, such as the continued need for a gradual reduction in fiscal deficits and for ratification of the trade agreements in the Uruguay Round, but it laid the groundwork for a more comprehensive agreement to follow.

This progressive effort culminated two years later in the adoption by the Interim Committee of the document, “Partnership for Sustainable Global Growth.”⁴⁶ By this time, the principles of the Washington Consensus had become so broadly accepted that they had (temporarily, as it happened) ceased to be controversial, at least among the world’s finance officials. Camdessus therefore decided that the time was ripe to redefine the purposes of the Fund beyond those set out in Article I of the Articles of Agreement. He was convinced that the IMF had to promote sustainable economic development, through its surveillance as much as through its conditional lending. Clarification and extension of the Madrid Declaration would make a good start.

The 1996 declaration, the final text of which was hammered out by the Executive Board in two drafting sessions just before the meeting of the Interim Committee, reiterated the validity of the earlier principles and restated them as a set of action-oriented goals.⁴⁷ It added that a goal of the global strategy was to promote the “full participation of all economies, including the low income countries, in the global economy. … Because the sustainability of economic growth depends on development of human resources,” the strategy had to include policies to strengthen the resources for and effectiveness of social spending and programs to protect the poor and alleviate poverty. The committee concluded the statement of 11 goals by encouraging “the Fund to continue to cooperate with other international organizations in all relevant areas.” (Emphasis added.) To underscore the operational implications, the Interim Committee communiqué added a quotation from the Managing Director, stating that the declaration was “the distillation of Fund surveillance by the world’s most representative body of finance policymakers.”

The issuance of these “eleven commandments” turned out to be the apogee of Camdessus’s ambitious effort to base Fund surveillance on this broad strategy. Indeed, it is difficult to discern any direct effect at all from any of these declarations on the treatment of macroeconomic or social policies in Article IV consultations with countries large or small. To a great extent, this failure reflected deep skepticism by many on the staff about whether the Fund was the right agency to promote structural reforms or that such an effort could succeed outside the framework of conditional lending. The difficulty was reinforced in the aftermath of the 1997–98 Asian financial crisis, when public, official, and academic skepticism about global trade and finance began to supplant the optimistic views of the mid-1990s. Another decade would pass before the Fund would successfully rebase its surveillance on specific but universal policy goals, and even then the scope of the reform would be greatly diminished.⁴⁸

Codes of Good Practices

What did bear fruit in the second half of the decade was the use of IMF surveillance to examine whether countries were meeting internationally established standards on institutional development, including data dissemination, financial sector soundness, and the conduct of fiscal policies. The Fund’s oversight of these specific issues is discussed later in this section. The general approach on codes of good practices evolved in response to the 1997–98 financial crisis in East Asia. While that crisis was still unfolding at the end of the summer of 1997, the Interim Committee asked the Fund to take several measures aimed at helping to prevent a recurrence. In that regard, the committee “stressed the importance of openness and accountability of economic policymaking” and asked the Fund to consider “developing a code of good practices.”⁴⁹

As an initial response to that request, the staff prepared a draft code of good practices on “fiscal transparency” in time for the Interim Committee’s next meeting. The final version of the code reflected many modifications made by Executive Directors, who went through the draft line by line before agreeing to send it to the Interim Committee in April 1998. Agreement by consensus was nearly derailed by objections from Russia, which insisted—both in the Executive Board and in the Interim Committee—that the document should be downgraded from a prescriptive “code” to a set of suggested “principles.” Eventually, Maystadt settled the issue by retaining the original title but adding “declaration on principles” as a modifier.⁵⁰ With that one change, the code was adopted and published as an attachment to the Interim Committee communiqué.⁵¹

Throughout 1998 and into 1999, work in this area proceeded in two directions. For one, the staff continued to develop standards for other policies for which it had a primary mandate, including on transparency in monetary and financial policies and on data dissemination.⁵² In addition, the idea began to take hold that the Fund should have a central role in monitoring countries’ adherence to such standards, including those that other agencies were responsible for developing.⁵³ In October 1998, a working party commissioned by the Group of 22 (G22, the forerunner of the G20) called on the Fund to use its Article IV consultations to prepare “transparency” reports on “the degree to which [countries meet] internationally recognized disclosure standards.”⁵⁴ Shortly afterward, the G7 finance ministers endorsed that suggestion and added that the Fund, in cooperation with other agencies, should “monitor … the implementation of these codes and standards,” publish its transparency reports, and advise member countries on how to improve compliance (Group of 22, 1998).⁵⁵

In March 1999, the Fund began preparing “transparency reports,” not as a surveillance activity but as technical assistance to its member countries. It then broadened and renamed them “reports on the observance of standards and codes” (ROSCs). At the outset, there were four ROSC modules: banking supervision, data dissemination, fiscal transparency, and transparency in monetary and financial policy. Although supportive of the staff’s proposals for this initiative, many Executive Directors insisted that participation had to be purely voluntary.⁵⁶ The first country to volunteer for an assessment by the IMF was the United Kingdom, for which the Fund prepared reports on all four categories in March 1999. That was followed by reports on Argentina and Australia in April, and on Hong Kong SAR and Uganda in August. The Fund then produced five ROSCs on Tunisia in September, adding a new category for regulation of the securities market. In addition, fiscal transparency ROSCs were issued on Ukraine in September and on Greece in December.⁵⁷

Pursuit of Orderly Liberalization

Liberalization of currency exchange for current account transactions was a widely shared goal and scarcely controversial. The same could not be said for liberalization of the capital account. When John Williamson coined the phrase “Washington Consensus” in 1990 to describe the generally accepted prescriptions for good economic policies, he pointedly omitted capital account liberalization from the list. Moreover, the Articles of Agreement did not include a mandate for the IMF to promote open capital flows. On the contrary, Article VI prohibited the Fund from lending to finance large or sustained capital outflows, and it empowered the Fund to require member countries to impose capital controls as a condition for borrowing if necessary to prevent a large or sustained outflow.

For the first quarter-century of IMF operations, private sector capital flows played a small, relatively benign, and mostly passive role in facilitating international trade and economic growth. In the 1970s, that situation began to change. The advent of generalized floating of exchange rates, combined with large-scale international wealth transfers associated with increases in oil prices, spurred rapid growth in capital flows. In a popular phrase of the day, that growth served to “recycle petrodollars.” Oil-exporting countries invested earnings with large international banks that then onlent the funds to oil-importing countries, including many developing countries. Beneficially, those flows limited or delayed the negative consequences of oil price increases on economies dependent on oil imports. That benefit came at a cost, because the inflows were often on highly liquid terms and were denominated in key currencies, not the local currency of the borrower. When short-term interest rates rose internationally in the early 1980s, bank creditors tried to withdraw liquid capital from developing countries. The resulting debt crisis nearly bankrupted sovereign borrowers and private creditors alike.

As detailed in Chapter 9, private capital began flowing back into developing countries in the early 1990s. Much as it had two decades earlier, but with more force and much greater breadth, this inflow fueled economic growth. Rather than coming only from international commercial banks in the form of loans, financial capital in the 1990s came from many different sources and in many different forms. Middle-income developing countries transformed themselves into “emerging markets” where companies could issue bonds and sell equities internationally. Adapting to these new market conditions and meeting the standards for attracting international capital became imperatives for any developing country aspiring to share in global prosperity.

The financial collapse of Mexico in December 1994 unmasked the fragility of this boom period, reminiscent of Mexico’s 1982 debt crisis that brought an end to the first great wave of postwar international capital flows. Most emerging markets nonetheless continued to attract capital for the next two or three years, until the Asian and Russian financial crises forced a major and broadly based pullback.

These shifting fortunes posed a challenge for IMF surveillance.⁶⁴ If a country was failing to establish economic conditions conducive to capital inflows, it would be underperforming with regard to potential growth. If it established conditions that attracted capital at an excessive rate, it could subject itself to destabilizing inflationary pressures and risk facing a sudden cessation of inflows. Either way, the IMF had a responsibility to assess the risks and offer appropriate policy advice.

The official guidance to the staff in the early 1990s did not specifically address these issues, but the Fund was aware of the dangers of uncontrolled inflows. A 1992 staff research paper, published as Calvo, Leiderman, and Reinhart (1993), was a seminal study of the potential for “sudden stops” in capital flows and their devastating effects on emerging markets. Another (Mathieson and Rojas-Suárez, 1993) examined several countries that had experienced destabilizing real exchange rate appreciations after opening their capital accounts. The Executive Board reviewed the issue in July 1993 and concluded that finding ways to stabilize and absorb capital inflows effectively should be a priority for Fund surveillance.⁶⁵ In 1994, at the request of the Asia-Pacific Economic Cooperation (APEC) finance ministers, the staff prepared a set of papers (Khan and Reinhart, 1995) on ways APEC member countries could better manage inflows of financial capital. The initial paper for the 1995 biennial review of surveillance, largely written before the eruption of the Mexican peso crisis, noted that the Fund had to have “clearer recognition of the extreme vulnerability of the recipients … to a sudden cessation or reversal of the inflows.”⁶⁶

In the aftermath of the Mexican crisis, the Fund saw clearly that it had to pay close attention to the dangers of unsustainable capital inflows. Mexico had attracted sizeable inflows throughout 1994 by selling a hybrid form of government securities called tesobonos, which were payable in pesos but in amounts that were effectively denominated in U.S. dollars (see Chapter 10). As the outstanding stock of these notes accumulated, the potential cost of a currency devaluation rose correspondingly. The situation was unsustainable, and it crashed at the end of the year.

When the Executive Board met in mid-February 1995 to conduct the biennial review of surveillance, Huw Evans (United Kingdom) proposed modifying the 1977 decision to include a concern about private capital flows.⁶⁷ The decision already contained a reference to “policies that provide abnormal encouragement or discouragement to capital flows” as one of the factors that could signal the need for a special consultation or other discussions with a member country. The new issue was whether, despite the pursuit of reasonable policies, a country might face circumstances in which private capital flows could become volatile or unsustainable.

After further discussion, the staff proposed, and the Board approved, revising the opening of the third paragraph of the 1977 decision to read (with the new wording in bold here), “The Fund’s appraisal of a member’s exchange rate policies shall be based on an evaluation of the developments in the member’s balance of payments, including the size and sustainability of capital flows, against the background of its reserve position and its external indebtedness.” In addition, Evans insisted, and the Board agreed, that the list of possible signals be extended to include “unsustainable flows of private capital.”⁶⁸ With those amendments, the Fund put itself on record in favor of limiting the size and volatility of capital flows to sustainable levels.

Limiting the size of capital flows did not imply that the Fund was in favor of capital controls. Rather, the Fund was in favor of a world in which (a) emerging-market countries would pursue stable macroeconomic and financial sector policies that did not abnormally encourage capital inflows and (b) advanced economies would pursue stable policies that did not give rise to abnormal incentives for capital to flow out to the emerging markets. Direct controls on capital flows were generally discouraged, although most Fund officials recognized that controls could be helpful in some circumstances if they were well designed and applied only temporarily. The long-run goal of many, especially in the mid-1990s, was a system in which capital would flow freely enough to spur development and growth but would not be so unruly as to destabilize national economies in the process. Unfortunately, the Fund found it difficult to convey this nuanced position with clear policy advice to its member countries.

Pleading for the Fifth Amendment

Pursuit of a stable and orderly liberalization of capital was hampered by both the complexity of the message and the lack of a clear mandate. In 1994, the Interim Committee’s Madrid Declaration expressed support for the free flow of capital, and that impelled the Fund to begin thinking seriously about the adequacy of its efforts in that direction. If each country was free to set its own course, and if the IMF acknowledged that capital inflows could have both negative and positive effects, then what advice could it sensibly give? If openness to international capital was a desirable long-term goal, how could the IMF ensure that each country was progressing adequately toward it?⁶⁹

In July 1995, the Executive Board had a preliminary discussion of the possibility of amending the Articles of Agreement to give the IMF a clear mandate to promote capital account liberalization. The staff paper was circumspect, and the strongest push for the idea came from Karin Lissakers (United States). “We would see merit,” she noted, “in an amendment to the Articles of Agreement to bring capital convertibility explicitly within the jurisdiction of the Fund and to introduce obligations regarding the staged liberalization of capital account transactions.” When a number of other Directors expressed reservations, Manuel Guitián (Director, Monetary and Exchange Affairs Department) responded by pleading for action. There was, he argued “general agreement about the economic advantages of such a liberalization,” but “current practice did not provide the Fund with any basis for influencing effectively what a country’s plans might be for capital account liberalization.”⁷⁰

The staff overall was more skeptical than Guitián, and so were most Executive Directors—especially those from emerging markets and other developing countries. The conclusions of the Board in July 1995 were that capital flows should be liberalized gradually, at a pace reflecting each country’s circumstances, and that the Fund had adequate scope under the Articles to monitor and encourage that process. The matter lay dormant for the next year, while the staff conducted more detailed research on the implications of large-scale capital inflows to developing countries. Camdessus kept the issue alive by asking the staff to prepare papers for the Executive Board on economic and legal issues related to “capital account convertibility and the role of the Fund.”⁷¹ The Interim Committee communiqué of September 1996 encouraged the Fund to continue working on it.

Serious consideration of amending the Articles began in February 1997. A staff paper set out the rationale and scope for such an amendment, and the Executive Board held an informal discussion of it. The Board reaffirmed the view that an “orderly and sustainable process” leading over time to “an open and liberal system of capital movements” was desirable and should be promoted by the Fund. Directors were prepared to recommend an amendment that would make capital convertibility part of the Fund’s mandate, but not everyone liked the idea of a prescriptive amendment giving the Fund jurisdiction over countries’ capital account policies.⁷²

Several Directors from developing countries were worried about the Fund pressing them to open their financial systems before they were ready, and some from industrial countries were worried about the Fund stretching into fields in which it lacked a clear rationale for its decisions. Thomas A. Bernes (Canada) suggested that other organizations, such as the Organization for Economic Cooperation and Development or the World Trade Organization, might be better placed for “rule making and negotiating.” Willy Kiekens (Belgium) expressed support for the idea that “the Fund should advise countries on issues connected with liberalizing capital movements.” He noted that his authorities were “not convinced that the Fund should have jurisdiction to decide whether controls are still needed during the transition period.”⁷³ It appeared that any amendment would have to be narrowly written to have any chance of being accepted and ratified by countries holding the requisite 85 percent of the voting power.

The Interim Committee endorsed this ongoing effort in its April 1997 communiqué and asked the Fund to come up with “key elements” for a proposed amendment by its next meeting. In September, the ministerial committee would be meeting in Hong Kong SAR, making for a special occasion and therefore an ideal time to modernize the Fund’s mandate. Throughout the summer, the staff, management, and Executive Directors worked to refine the technical aspects of the proposal. How, and under what circumstances, might the Fund approve capital controls? Should the Fund impose a time limit? What if a country committed itself to an open regime and then subsequently introduced a new restriction? Should all types of capital flows be treated alike, or should direct investment be treated differently? Should the Fund’s jurisdiction be limited to currency convertibility, or should it extend to the underlying transactions? These questions would have to be resolved before a sensible amendment could be drafted, and the Fund devoted a fair amount of time to the task.⁷⁴

A major outcome of these discussions was recognition that an overly narrow amendment would not have the desired effect. Making liberalization of capital flows a purpose of the Fund without simultaneously imposing obligations on all member countries and giving the Fund jurisdiction over those obligations would simply enable the Fund to justify imposing related conditions on its lending arrangements. To be effective, the amendment would have to be comprehensive and applicable to all members.⁷⁵

At the Hong Kong SAR meetings, enthusiasm for the open-capital amendment was at an all-time high but was not unalloyed. At a seminar preceding the formal meetings of Fund and World Bank Governors, Stanley Fischer laid out a carefully reasoned case for the amendment, arguing that it would enable the Fund to guide liberalization in an orderly way, taking account of each country’s circumstances and allowing for as long a transition as was appropriate.⁷⁶ The Interim Committee gave its blessing and trumpeted that it was “time to add a new chapter to the Bretton Woods agreement.” It added a note of prudence, however, that the “Committee sees the Fund’s proposed new mandate as bold in its vision, but cautious in implementation.”⁷⁷

In the months after the meetings in Hong Kong SAR, crises spread from Thailand to Indonesia and Korea and threatened to engulf the whole East Asian region. Although the fundamental causes derived from weaknesses in the financial sectors of the primarily affected countries, the shock and the crisis consisted of a sudden withdrawal of private capital. Belatedly, Fund staff realized that the “carry trade” in Japanese yen—in which speculators borrowed yen at extremely low interest rates and invested the proceeds in emerging markets with much higher yields—had greatly magnified the consequences of domestic policy shortcomings in emerging-market countries.⁷⁸ In a January 1998 speech on the Asian crises, Fischer noted that

developments in the advanced economies and global financial markets contributed significantly to the buildup of the imbalances that eventually led to the crises. … Large private capital flows to emerging markets, including the so-called “carry trade,” were driven, to an important degree, by these phenomena and by an imprudent search for high yields by international investors without due regard to potential risks.⁷⁹

The fear of exposure to the vagaries of international capital markets had begun to outweigh the desire to reap the uncertain benefits.

From that point on, the debate over the amendment was muddled. Camdessus and Fischer tried to project the message that their goal was to guide liberalization gradually, country by country. In their view, each country should first strengthen its economic, legal, regulatory, and other institutional policies. The longer-run goal of full convertibility would help speed up and guide that process, and the IMF was the agency best situated to monitor and lead it. They were not aiming to induce countries to open their financial systems to capital inflows before the preconditions were in place. After the financial crisis in Asia, that message got lost. Critics focused on the dubiety of both the goal and the means. Were capital controls necessarily bad? Was the IMF the right agency to promote their elimination? As discussed in Chapters 11 and 12, the Fund was mistakenly accused of having forced Asian countries to open their capital accounts prematurely, thus aggravating the crisis. That perception added to the difficulty of having a reasonable discussion of the Fund’s role and mandate.⁸⁰

The bureaucracy of the IMF continued to work on preparing an amendment, and Camdessus continued to press for a successful conclusion to what he called a “major historic opportunity.” The staff organized a two-day seminar in March 1998 at which a number of senior government officials spoke alongside leading academic economists. Although most speakers and participants in the discussions agreed capital liberalization should be made a purpose of the Fund, they were much less enthusiastic about giving the Fund jurisdiction over it in any way comparable to the Fund’s existing jurisdiction over current account restrictions.⁸¹ Even inside the Fund, much of the steam behind the amendment effort had dissipated.

The Executive Board met on April 2, 1998, to consider a proposal from Camdessus to amend the Fund’s purposes as set out in Article I and to agree in principle on “the general rule [with specific exceptions] that members are prohibited from imposing restrictions on international capital movements without Fund approval.” No formal vote was taken, but enough Directors spoke out against the jurisdictional rule that it had to be shelved.⁸²

For the April 1998 meeting of the Interim Committee, Camdessus therefore submitted a proposal for a possible fifth amendment that would merely revise Article I to add the “orderly” liberalization of capital flows as an additional purpose of the Fund. Specifically, the purposes of the Fund set out in Article I would read as follows, with the new language in bold here:⁸³

• The purposes of the International Monetary Fund are:

  • (i) To promote international monetary cooperation through a permanent institution which provides the machinery for consultation and collaboration on international monetary problems.

  • (ii) To facilitate the expansion and balanced growth of international trade in goods and services and an efficient international allocation of capital, and to contribute thereby to the promotion and maintenance of high levels of employment and real income and to the development of the productive resources of all members as primary objectives of economic policy.

  • (iii) To promote exchange stability, to maintain orderly exchange arrangements among members, and to avoid competitive exchange depreciation.

  • (iv) To assist in the establishment of a multilateral system of payments in respect of current and capital transactions between members, in the orderly liberalization of international capital movements, and in the elimination of foreign exchange restrictions which hamper the growth of world trade and investment.

  • (v) To give confidence to members by making the general resources of the Fund temporarily available to them under adequate safeguards, thus providing them with opportunity to correct maladjustments in their balance of payments without resorting to measures destructive of national or international prosperity.

  • (vi) In accordance with the above, to shorten the duration and lessen the degree of disequilibrium in the international balances of payments of members.

The Interim Committee put the best possible lipstick on the proposal by endorsing this language for Article I while noting that amendments to the more operational articles would also be required as part of the package. The matter was thus back in the hands of the Executive Board, which remained reluctant to push forward. During the rest of 1998 and 1999, the Board held several meetings to discuss various aspects of capital controls and liberalization, but its only further formal discussion of a possible fifth amendment was in an inconclusive seminar in July 1999.⁸⁴ Even the uncontroversial extension of purposes quoted above was never submitted to the Board of Governors for consideration.

Monitoring the Markets

Aside from developing policies on capital flows, the IMF had to determine the best way to monitor what was going on. When international banking flows exploded in the early 1970s, the Fund began tracking the implications by holding special discussions with the monetary authorities of countries with major financial centers. In 1974, the Executive Board reviewed developments in what were then called euro-currency markets and concluded that the staff should continue to build up a knowledge base about these markets. For the next five years, market developments were reported to the Board through the WEO. Then in 1979, staff in ETR began preparing a separate annual report, initially under the name “International Capital Markets: Recent Developments and Near-Term Prospects.” Whereas the 1974 report was based on meetings only with country officials, the preparatory missions in 1979 met with private sector bankers as well.⁸⁵ These capital markets missions and reports became an annual exercise for the Fund. The following year, the Fund began publishing the reports, the first one being Williams (1980).⁸⁶

The debt crisis of the 1980s brought Fund staff into much more regular contact with commercial and investment bankers, foreign exchange traders, and other market participants. The acceleration and diversification of capital flows in the 1990s pulled the staff in even more deeply. For the 1999 report, staff missions traveled to 17 countries, including all of the major international financial centers as well as such emerging markets as Argentina, China, Hungary, Malaysia, and Turkey. As this work intensified, the missions and the reports were elevated to a more central role in Fund surveillance.

The growing importance of the capital markets report led to a subtle change in its subtitle in 1992, to “Developments, Prospects, and Key Policy Issues.” (Emphasis added.) At the same time because the report was becoming more analytical in scope, primary responsibility for its preparation was shifted from ETR to RES (which was also responsible for preparation of the WEO). Throughout the 1990s, the capital markets report took its place alongside the WEO as one of the two flagship publications of the IMF.

The clearest example of the influence of the capital markets report came in 1995, when the report included a detailed and (by IMF standards) unusually blunt retrospective analysis of the Mexican peso crisis. A debate had been raging in Mexico about the role of foreign investors and speculators in precipitating the crisis. Taking sides, to the consternation of the authorities, the Fund report concluded that “the pressure on Mexico’s foreign exchange reserves … came not from the flight of foreign investors or from speculative position-taking by these investors, but from Mexican residents.” The ensuing controversy eventually became an important element in the evolving understanding of how to avoid such a crisis in the future.⁸⁷

The capital markets report was also a major channel through which the staff analyzed other crises in the 1990s, including the ERM crisis of 1992–93 and the Asian crises of 1997–98. As discussed in Chapter 11, the capital markets mission to Korea in April 1997 raised an alarm about how deep and widespread the weaknesses in the Korean financial system were, well before the crisis erupted in November. Unfortunately, that message was not absorbed by either the area department or senior management until much later. One of the critical lessons from the Asian crisis was the need for much better integration of capital markets and macroeconomic analysis, especially for vulnerable emerging markets.

Financial Sector Soundness

Certain structural policies linked naturally to the Fund’s traditional focus on macroeconomics and finance. The most important of those was the soundness of a country’s banking system. Until the mid-1990s, the institutional structure of national finance was not systematically reviewed by the IMF unless a country requested technical assistance on the subject.⁸⁹ If a country had manifest problems or if it was undertaking major institutional changes, the staff would take note of it in its Article IV consultation reports, but the Fund did not generally attempt to uncover institutional weaknesses in anticipation of a crisis. The limitations of this approach are best illustrated by the Swedish banking crisis.

Sweden. Throughout the 1980s, Sweden carried out major structural reforms aimed at making the financial sector more competitive, both domestically and internationally. As nonbank finance companies flourished in response to these reforms, banking regulations were removed or reduced to enable banks to compete. In the second half of the decade, a number of Swedish banks merged with one another, and large international banks established offices in Sweden. These changes generated substantial capital outflows, as residents took advantage of new opportunities to invest in foreign currencies and in foreign financial institutions, but both the authorities and the Fund staff viewed this as a benign side effect of the transition to a more open and vibrant system. Assessing the situation in mid-1990, the staff “welcomed” the liberalization of the financial sector, through which the “highly regulated financial sector had been transformed into a market system.”⁹⁰

In fact, the “market system” had a fatal flaw, similar to the one that gave rise to the crisis in U.S. Savings and Loan institutions in the late 1980s and that later would fuel other financial crises such as those in Thailand in 1997 and worldwide in 2008.⁹¹ Although interest rate and some prudential regulations had been softened, allowing financial institutions to offer high rates to attract deposits they could onlend to finance risky investments, depositors and investors generally viewed those institutions as implicitly protected by the regulatory authorities. This underpricing of risk led to excessive borrowing and hence to a bubble in asset prices, notably in commercial real estate. When the deterioration in asset quality became evident, sizeable government assistance was required to prevent a wave of insolvencies from undermining the payments system.

The imbalances in the Swedish financial sector were already evident by 1990, but IMF surveillance at that time was focused on macroeconomic data. Bank assets were growing rapidly and aggregate profits were holding up well, but the underpricing of risk meant that profit margins were being squeezed. As the staff would later acknowledge, banks were already becoming increasingly vulnerable to risks and increasingly dependent on ever more costly sources of funds.⁹² Real estate prices had peaked in 1989, but the declines were not yet large enough to precipitate a crisis, and the staff saw no reason to raise an alarm.

In December 1990, the Swedish parliament authorized the government to apply for membership in the European Communities, which would soon be reconstituted as the EU. The following May, as one step toward membership, Sweden changed its exchange rate policy from a trade-weighted basket peg to a peg against the European Communities’ unit of account, the European currency unit (ECU). Although that shift had the effect of hardening the currency at a time when the real economy was beginning to soften, the staff noted that it immediately produced an inflow of portfolio capital and thus led to a decline in interest rates. It was expected to bring a drop in inflation as well, and the staff accordingly fully endorsed it. The 1991 staff report also continued to praise the way the authorities were overseeing the financial sector. “Sweden has had considerable success in carrying out structural reforms,” the mission wrote in August 1991. “The tax reform and the deregulation of financial markets have made important contributions.”⁹³

When the next staff mission went to Stockholm in May 1992, it was to conduct only a bicyclic “interim” consultation, which would be concluded without a meeting of the Executive Board. That decision implied that the Fund viewed Sweden’s circumstances as benign, not posing a serious macroeconomic threat either to itself or to its neighbors. The staff mission was led by the third chief in as many years, reflecting the demands being placed on the staff by work on the large number of new European member countries.

In the course of the 1992 discussions, the authorities informed the staff they were having problems with the banking sector, notably in the form of large losses at two major banks attributable to “the steep falls in commercial property prices.” Those losses were sufficiently large to require substantial government support. Most banks, however, still satisfied international capital adequacy standards, and neither the authorities nor the staff expressed much anxiety. The overall appraisal in the mission’s report did not mention the issue at all.⁹⁴

The Fund’s concern about the banking situation—not only in Sweden but in the whole Nordic region—began to grow during the summer of 1992. In August, Camdessus stopped in both Finland and Sweden for some quiet discussions about steps that might be needed if more banks got into trouble. Although it was highly unlikely that either country would ask for loans from the IMF, he stressed to the authorities that the Fund was ready to help if needed.

The Fund’s apprehension about financial stability in Sweden heightened considerably when the ERM crisis stuck on Black Wednesday, September 16, 1992. Although Sweden did not participate in the EMS, it pegged the krona to the ECU and thus was just as vulnerable to a speculative attack. The central bank, the Riksbank, mounted a classic and vigorous interest rate defense, briefly raising interest rates as high as 500 percent. When that defense failed to stem the attack, the authorities were forced to abandon the peg in November and allow the currency to float.

During this period, the Fund twice responded to invitations from the Riksbank to send staff—led by Desmond N. Lachman (Assistant Director, European I Department)—to Stockholm to advise the authorities. The first visit, in mid-September, coincided with the ERM crisis and focused on how to insulate Sweden from the attack. The Riksbank was determined to defend the exchange rate peg, but Lachman argued that the first priority should be to reduce Sweden’s burgeoning fiscal deficit. Even with a “bold fiscal package,” he warned that the exchange rate would still be overvalued and a float was close to inevitable. His second visit came right after the Riksbank was indeed forced to float the krona on November 19. Again the staff’s main concern was the lack of political will to get the government’s fiscal accounts in order.

The 1993 consultation finally provided an opportunity for the staff to respond comprehensively to the ongoing emergency, which it called Sweden’s “worst economic crisis since the 1930s … characterized by … a marked decline in output and employment, a ballooning fiscal deficit, and a banking system under severe strain.”⁹⁵ Moreover, the staff now acknowledged that the crisis had been brewing for several years:

During the second half of the 1980s, in a climate of financial market deregulation, Sweden’s banking sector overextended credit in an effort to gain market share, thereby contributing to an asset price boom. The subsequent slowing in the economy and puncturing of the asset price bubble resulted in substantial credit losses in the banking sector. Between 1990 and 1992, Swedish bank groups had to make provisions for bad debts totaling … almost 12 percent of total bank lending, while the State had to intervene directly in support of a number of important banks. In order to maintain public confidence in the banking system, the Government was obliged to provide direct budget assistance to individual banks totaling … almost 2 percent of GDP, in late 1991 and 1992.⁹⁶

By then, the authorities were already well on their way to resolving the problems in the banking sector. After 1993, no further government support of the troubled banks was required, and real GDP was growing again.⁹⁷ For the rest of the decade, consultations with Sweden continued to focus primarily on deficit reduction and related fiscal policy issues. More generally, however, the main lesson for the IMF from this episode was the need to examine the health of the banking system as an integral part of bilateral surveillance. In the Swedish case, the Fund was able to assess the extent and the implications of the sectoral weakness once the problem had come to a head, but not before. Only by broadening the annual discussions so as to examine the financial sector routinely could this shortcoming be alleviated.

Toward a systematic analysis. The Swedish banking crisis stimulated the staff to begin thinking in general terms about the potential for similar problems in other advanced economies, but it did not immediately alter the conduct of surveillance. The Fund’s treatment of financial sector issues in its surveillance activities began to evolve only in 1995, in the wake of the financial crisis in Mexico. As pressures on the peso built up during 1994, one of the reasons the Mexican authorities were reluctant to raise interest rates was a fear that major banks would suffer large losses. At the time, IMF staff had no way to evaluate this threat. As the realization set in that this gap was a substantial hindrance to the effectiveness of surveillance, the Fund set out to strengthen its procedures.

In April 1995, the Interim Committee “noted the risks attached to overreliance on easily reversible capital inflows, and invited the Fund to pay more attention to members’ financing policies, and the soundness of their financial sectors, in its surveillance activities.”⁹⁸ The staff intensified its work in this field, mainly in the Monetary and Exchange Affairs Department (MAE), where Carl-Johan Lindgren (Chief of the Banking Supervision and Regulation Division) led a team effort. The findings in the staff report were alarming, beginning with the opening sentence: “Since 1980, over 130 countries, comprising almost three fourths of the [IMF’s] member countries, have experienced significant banking sector problems” (Lindgren, Garcia, and Saal, 1996, p. 3).⁹⁹ In response, the Board decided in March 1996 that Fund surveillance should be reoriented to focus more directly on bank soundness and especially on the interactions between the banking sector and the macroeconomy.¹⁰⁰

The emphasis on examining banking sectors increased further in 1997, after the staff produced two follow-up studies. In February, the Board agreed that the Fund was “uniquely placed,” because of its surveillance and technical assistance roles, to alert member countries to weaknesses in their banking systems and to encourage countries to adhere to internationally accepted standards. In March, the Board asked the staff to develop a more detailed analytical framework for assessing the strength of countries’ banking and financial sectors. In July, management approved a revised guidance note for the conduct of surveillance activities, which—for the first time—specifically asked that all staff reports “should include assessments of financial market developments and prospects as well as problems and policy issues in the banking and financial sector where they are of macroeconomic significance.”¹⁰¹

Two external developments further spurred the Fund’s work on financial sector soundness. One was the issuance of a set of “core principles” by the Basel Committee on Banking Supervision. The committee—a standing body of central bank officials convened by the Group of Ten (G10) and headquartered at the Bank for International Settlements (BIS) in Basel, Switzerland—issued its recommendations in draft form in April 1997 and in final form in September, after receiving comments from the Fund and other agencies.¹⁰² That document provided guidelines for the staff to follow in assessing the soundness of national banking systems.

The second development was the wave of financial crises that hit Thailand, Indonesia, and Korea in the second half of 1997. That shock laid bare the extent of weaknesses in financial sectors throughout the world’s emerging markets and the depth of the interconnectivity between those weaknesses and macroeconomic vulnerability. It also revealed weaknesses in surveillance. The affected countries were not providing enough information for the Fund to assess the strength of their banking sectors, and the staff did not have sufficient specific expertise to uncover problems independently or to challenge the authorities’ representations. The need for better and more consistent surveillance of financial systems was to be one of the crucial lessons from the Asian crises.¹⁰³

By 1998, almost every staff report on Article IV consultations with emerging-market countries included an assessment of the country’s framework for financial sector oversight. The great majority of those reports included recommendations for reform. In May 1999, the Fund and the World Bank jointly launched an additional program on a pilot basis, known as the Financial Sector Assessment Program (FSAP). On the Fund side, this program involved sending a special FSAP mission to each participating country ahead of the regular Article IV mission to prepare a Financial Sector Stability Assessment. Those assessments, prepared by a separate team of specialists rather than as a side activity by the area department’s country team, provided in-depth analyses of financial soundness and oversight for review by the Fund as an integral part of its bilateral surveillance. Four FSAP missions were sent out in 1999, to Lebanon (in May), Colombia (July), Canada (October), and South Africa (October). The first reports were completed in 2000, and the FSAP program was made permanent later that year.¹⁰⁴

Unproductive Spending

In 1988, Helmut Schmidt (chancellor of Germany, 1974–82) convened a UN-sponsored commission of distinguished officials from around the world to prepare a report on what could be done to increase aid flows to developing countries. The resulting 1989 report made several recommendations, one of the more intriguing being that donor countries should direct their scarce aid resources to countries spending less than 2 percent of GDP on the military. At the time, developing countries were spending an average of 4.7 percent of GDP on military outlays—more than four times what they were receiving in official development assistance. By cutting back on the “excess” and getting more aid as a reward, countries could make large gains in education, health, infrastructure, and other development essentials (see Schmidt, 1989).¹⁰⁵

Camdessus took up this proposal as a cause for the IMF. In February 1990, a few months after the Schmidt Commission report was published, Camdessus used a speech in the Philippines to warn his audience that “countries that take their social responsibilities seriously … are more likely to attract international help than countries that waste money on unproductive prestige projects or excessive military display.”¹⁰⁶ Meanwhile, the crumbling of the Soviet empire and the consequent end of the Cold War were raising the prospect that the major countries could and should sharply reduce their own military spending. The combination of the moral imperative for development and the economic benefit of the presumed “peace dividend” was pushing “unproductive spending” to the forefront of international debates.

In the first half of 1991, the Development Committee, the World Bank’s World Development Report, and the G7 summit meeting all appealed to countries to cut back on military spending.¹⁰⁷ The summit leaders specifically welcomed the efforts being made by Camdessus and by Barber Conable, President of the World Bank, to call attention to “excessive military spending, in the context of reducing unproductive public expenditure.”¹⁰⁸

In preparation for the 1991 IMF/World Bank Annual Meetings in Bangkok, the Fund responded to these various appeals by reviewing its own ability to assess members’ military budgets in the context of its bilateral surveillance under Article IV. Not surprisingly, the Fund’s Executive Directors were not of one mind on this controversial issue. Many countries closely guarded any information on military spending, and they were not likely to give it out to satisfy the IMF. If they expected the Fund to use the data to try to induce them to cut down, they would be even less willing to divulge it. Several African countries were fighting off armed insurrections, a particularly brutal example being the one waged by Liberia-based rebels against the government of Sierra Leone. In such circumstances, at what level could military spending be said to be excessive? Lacking a clear mandate, the Fund’s options seemed limited, and its case seemed weak.

After two days of discussion, the Executive Board agreed on a minimalist policy under which the staff could request data on military spending in the course of Article IV consultation discussions but could not require the authorities to provide it. The staff was expected to respect national sensitivities and sovereignty concerns. The Board explicitly eschewed using such data to establish conditions on loans from the Fund.¹⁰⁹ The policy thus amounted to little more than an expression of concern that “information on [military] expenditures may be necessary to permit a full and internally consistent assessment of the member’s economic position and policies.”¹¹⁰

The Interim Committee, meeting in Bangkok, also took a cautious view, noting only that in the global effort to promote national saving, “an important contribution could be made by reassessing spending on defense and subsidies.”¹¹¹ Camdessus was undeterred, and for the next few years he conducted a personal campaign to persuade countries to curtail military spending. He began by stopping in India and Pakistan on his way back to Washington from Bangkok. In both countries, he pressed the authorities to take advantage of more peaceful global conditions by redirecting spending from military outlays to education and health. In both countries, he made the same point in public speeches: “What a fine example it will be to the rest of the developing world, if these two great nations can each transfer substantial human and financial resources to activities that will more directly contribute to growth and to the reduction of poverty. What a prospect that could create for a better life for everyone in the subcontinent!”¹¹²

Staff in the Fund’s Fiscal Affairs Department (FAD) were assigned to monitor the adequacy of statistics and data reporting and to undertake research on the macroeconomic effects of military spending, and the subject was raised in about 9 percent of Article IV consultation reports (none in industrial countries).¹¹³ The Fund’s involvement remained on a low flame—just bright enough to throw some light onto the most troubling cases—and it was not elevated further in the course of the 1990s. Other types of “unproductive” spending, such as the subsidies to which the 1991 Interim Committee communiqué alluded (see the preceding paragraph), were never systematically treated in a Fund policy statement but were made the subject of staff analysis in some consultations.¹¹⁴

National Governance and “Reform of the State”

As with excessive military spending, staff and management at the IMF were always concerned about corruption and weak governance in member countries but had difficulty figuring out what if anything to do about it. Until the 1990s, they were reluctant to express those concerns openly or to take any action in response. There were exceptions, usually limited to specific advice or requirements in the context of IMF lending. For example, if tax policies were routinely abused, or government spending practices were poorly documented, or banks were induced to lend to official or politically connected borrowers, the Fund might try to find ways to persuade or induce countries to limit such practices. The Fund also offered technical assistance to many countries to help strengthen fiscal and financial sector practices by making them more open and transparent.

These occasional efforts did not amount to a real policy. From the late 1970s through the 1980s, for example, the IMF lent frequently to Zaïre despite knowing that much of the proceeds were being siphoned off by President Mobutu Sese Seko and his cronies for their personal benefit (see Boughton, 2001, pp. 804–10). Protecting the Fund from political pressure and stopping such practices was becoming a necessity and a priority.¹¹⁵

In the 1990s, the Fund broadened its involvement in fighting corruption and in improving fiscal and financial governance and applied its concerns more consistently. It also extended this reach more regularly into its surveillance activities as well as its lending practices. To do so, it had to pull back a bit from its long-standing reluctance to inject political considerations overtly into its lending decisions and policy advice.

The drive to reduce military spending that rose to the fore in 1991 was motivated in part by a conviction that such excesses were often a symptom of official corruption. For the most part, those who spoke out on the issue stuck publicly to the theme of promoting development and increasing economic efficiency. Nearly everyone understood, however, that corruption was the heart of the problem. At the 1991 Annual Meetings in Bangkok, the point was made most clearly by Wim Kok, deputy prime minister of the Netherlands, who told the assembly that successful development policies require a “substantial investment in human resource and physical infrastructure. … The effectiveness of such policies depends on international economic cooperation and support, but mainly on good governance, and could often be enhanced by a cutback of disproportionate military expenditures” (IMF, 1991, p. 35).¹¹⁶ Still, the time had not yet come when the IMF, or even the World Bank, had enough official support from the international community to take on corruption openly as a policy issue.¹¹⁷

In contrast to the Fund’s reluctance and timidity in the 1980s, in the 1990s the institution gradually showed a willingness to address at least the most egregious cases of official corruption. As recounted in Chapter 14, the staff, the Executive Board, and Camdessus strongly criticized the Kenyan government in 1993 for the country’s pervasive culture of bribery, nontransparent relations between the government and economic enterprises, and other forms of corruption. Those criticisms were made in private and did not—at that time—block approval of the financial support Kenya was asking for, but they at least served notice of unease. In the transition economies, especially those that emerged from the former Soviet Union, the Fund routinely included loan conditions aimed at reducing corruption and promoting good governance. Effective enforcement of those conditions, however, proved difficult.¹¹⁸

In January 1994, Camdessus decided to take his broader concerns public. Addressing a meeting of world church leaders at the Inter-American Development Bank headquarters in Washington, he asserted that “high quality growth … the ultimate objective of our work at the IMF … requires good governance. By this I mean publicly accountable and participatory government that serves the whole of society rather than sectional interests, and legal and regulatory frameworks that are transparent, fair, and limited to what is strictly necessary, so that the scope for arbitrary administrative decisions and corruption is minimized.”¹¹⁹ In a 1995 speech to the UN Economic and Social Council, he made a direct link to economic development: “to fight corruption … is a difficult task, requiring courage and perseverance. But progress must continue: it will profoundly strengthen the development process.”¹²⁰

Talking about corruption was a vital first step, but the greater challenge was to broaden support in the Executive Board sufficiently to turn that campaign into action. An opportunity arose in March 1996, when the Board was scheduled to hold an overnight retreat in the Virginia suburbs of Washington. One of the issues that Camdessus put on the table for discussion was “whether and how the Fund, in its relations with members, should address governance issues that have significant macroeconomic implications that are important to the Fund’s surveillance and financial assistance.”¹²¹ As background material, Camdessus circulated a working paper by Vito Tanzi (Director, FAD) on “Corruption, Government Activities, and Markets,” which set out a theory of the way corruption affected economic performance and how it might be controlled.¹²²

When Camdessus made the case over dinner that Fund-supported programs were failing frequently because of corruption, Executive Directors were mostly skeptical that this was an issue that the Fund could or should tackle. As the discussion continued in the morning, however, views began to shift. Obviously, corruption could and did undermine the effectiveness of the Fund’s financial support. More generally, if the Fund were to ignore corruption it knew was weakening a country’s macroeconomic performance, the credibility and effectiveness of its surveillance would be damaged. At the end of the day, Directors agreed to try to develop a broad and evenhanded policy for examining corruption and other governance issues in the context of surveillance, lending, and technical assistance.¹²³

The IMF was not alone in ramping up its efforts to combat corruption. In the mid-1990s, a wide variety of international organizations—the EU, the Financial Action Task Force, the Organization of American States, the Organization for Economic Cooperation and Development, the UN Development Program, the World Bank and other multilateral development banks, and the World Trade Organization—were taking actions within their own mandates to promote good governance. Several of those bodies as well as civil society organizations were actively pushing the Fund to join in the battle. In most cases, the IMF had observer status or otherwise participated in relevant discussions.

The legal basis for an IMF role was not spelled out clearly, but with a little ingenuity it could be inferred from Article IV, Section 1, of the Articles of Agreement. That section required each member country

to endeavor to direct its economic and financial policies toward the objectives of fostering orderly economic growth with reasonable price stability, and to seek to promote stability by fostering orderly underlying economic and financial conditions and a monetary system that does not tend to produce erratic disruptions.

On that basis, the Fund determined it had an obligation to evaluate, through its surveillance activities, whether poor governance was impinging seriously on a country’s economic performance and, if so, whether its policies were in compliance with Article IV.¹²⁴

While the debate was continuing over a possible role in addressing governance through surveillance, the Fund moved ahead in dealing with governance problems in borrowing countries. Notably, when Kenya asked for an ESAF arrangement in the fall of 1995, the Fund insisted that the authorities take specific measures to control corruption before it would approve the request. On the basis of initial signs of progress, the Executive Board gave its approval in April 1996 and allowed a first disbursement. When the government failed to follow up, and evidence of pervasive corruption persisted, the Fund cut off further lending for the next four years.¹²⁵

The clarity of the Fund’s mandate to evaluate governance was given a further boost with the adoption of the “eleven commandments” by the Interim Committee in September 1996.¹²⁶ The tenth item on the committee’s list of objectives was “promoting good governance in all its aspects … as essential elements of a framework within which economies can prosper.” The declaration concluded by encouraging “the Fund to continue to cooperate with other international organizations in all relevant areas.”¹²⁷

Within days after that document was issued, Camdessus used his speech to the Governors assembled at the IMF/World Bank Annual Meetings to call for a “reform of the state,” meaning that “governments must demonstrate that they have no tolerance for corruption in any form.” At the same meeting, World Bank President James D. Wolfensohn made an even more explicit statement on the subject, decrying the “cancer of corruption” and promising that “the Bank Group will not tolerate corruption in the programs we support.”¹²⁸ Camdessus was not yet able to make a similar promise, because the Fund still lacked a clear policy on corruption. The time had now come.

The Interim Committee declaration provided the impetus for Camdessus to ask the staff to prepare a guidance note for the Fund’s role in governance issues. When the Executive Board had its first look at a draft in January 1997, the key point that emerged from the discussion was that the Fund should limit its involvement to the “economic aspects of governance” and strictly avoid getting embroiled in political issues. Everyone seemed to agree, though, that the demarcation between the two was extremely fuzzy and ill-defined.¹²⁹ The staff was sent back to refine its proposals, and it finally submitted a guidance note that the Board agreed was appropriate and within the Fund’s mandate.¹³⁰ The guidance note set out a specific, though not necessarily exclusive, list of economic governance issues the staff might reasonably address, as well as any instance of corruption judged to have “significant macroeconomic implications, even if those effects are not precisely measurable” (paragraph 10). In the normal course of Article IV consultations, the staff was expected to raise such issues with the authorities and report on the discussions to the Executive Board.

During the next three years, staff reports for Article IV consultations addressed governance issues in nearly a quarter of all cases, up from 18 percent in the preceding two years. More important, the nature of the references tended to be more pointed.¹³¹ Many of those reports, though not all, were for countries borrowing from the Fund or asking to borrow. A notable exception was the 1999 report on Nigeria, to which the Fund had not lent any money since 1992. President Olusegun Obasanjo had just taken office in the country’s first democratic elections, and he wanted to get the Fund’s seal of approval and restore the country’s credibility in international markets. The staff report noted that in economic terms, Nigeria was no better off in 1999 than it had been in 1970, mainly because of long-standing and pervasive corruption, “financial malfeasance,” and other forms of weak governance. The staff mission, led by Hiroyuki Hino (Senior Advisor, African Department), told the new authorities that the Fund could not approve their policies until they took concrete steps to deal with these problems. Obasanjo was already committed to doing so, and the report served to reinforce the message that the international community was behind the effort.¹³²

All of the Article IV staff reports covering governance problems in the late 1990s were for developing or transition countries. Despite the repeated calls for evenhandedness, mission chiefs remained reluctant to call attention to such issues in the more advanced economies. A clear example occurred in 1997, when the Fund conducted its annual consultation with Austria. The staff mission, led by Hans M. Flickenschild (Advisor, European I Department, or EU1), held extensive discussions with the authorities in Vienna and with trade union representatives and other nongovernmental organizations on the pressing need for Austria to adopt structural reforms that would make it more competitive in an increasingly globalized marketplace. The approaching date of the introduction of the euro added to this imperative. At the time, the Austrian government still had a heavy hand in the economy, and a wide range of rigidities was hampering overall economic performance. One important factor was the role of political patronage in the appointment of senior managers in enterprises and banks. The staff was aware of and concerned about the problem, but no mention was made of it in the staff report.¹³³

When the Executive Board met to conclude the consultation with Austria, Flickenschild explained the staff’s concerns and noted that the issue had been omitted from the report owing to its “sensitive nature.” That drew rebukes from Martin A. Brooke (Advisor, United Kingdom) and Eva Srejber (Sweden), who reminded staff of the need for evenhanded treatment of all countries. Wolf-Dieter Donecker (Alternate, Germany) defended the staff for its caution and even questioned whether “exerting political influence on major companies and banks” was really “bad governance” or was “simply a custom in many countries.” Rejoining for the staff, David Burton (Senior Advisor, PDR) acknowledged that even treatment was important and that the matter “should probably have been touched upon in writing in the staff report.” Even so, the Summing Up of the Board discussion made no mention of it.¹³⁴

Protection of the Environment

No issue illustrates the delicate balance between focus and breadth better than preservation of the natural environment. Could the IMF assess the sustainability of a country’s macroeconomic policies without examining the effect of economic growth on the environment? Before the 1990s, few would have taken that question seriously. The World Bank, the UN, and civil society were much better equipped than the Fund and had much clearer responsibilities in this domain. By this decade, however, preserving the environment was becoming such a critical and universally shared goal that all able-bodied institutions were being conscripted into the cause. The IMF was no exception, but views differed strongly on how it should respond.

The prime advocate for IMF involvement in this area was the U.S. government. In 1989 and 1990, the U.S. Congress passed two laws calling on the U.S. Executive Director to encourage the IMF to consider the effects of its policies on environmental sustainability and to reduce or eliminate any negative impacts. The Fund responded by preparing a study discussed by the Executive Board in December 1990. The staff study noted that Fund-supported programs were macroeconomic rather than structural and therefore had no direct or generalized impact on the environment. It acknowledged, though, that the indirect effects could be significant and harmful. If a country had to take action to control aggregate demand or reduce imports to fix an unsustainable payments deficit, exporters might take actions such as increasing logging of old-growth forests to make up for the short-term losses. Positive responses also were possible, such as an increase in taxes on polluting activities or cuts in subsidies on chemical fertilizers. The challenge was to understand these various possibilities and possibly to guide the responses in a positive direction.¹³⁵

Camdessus had no doubt that the environment was an appropriate concern for IMF surveillance. The U.S. authorities were asking the Fund to monitor the effect of economic policies on the environment, and he proposed establishing a small unit of two or three staff within FAD for this purpose. The existing staff was already stretched thin, and he pointed out that the Fund could not take on this task without the budget authority to hire additional specialists.¹³⁶ The Executive Board, however, was unconvinced.

Executive Directors raised two objections, one general and one more specific. The general issue was that the IMF lacked a mandate to deal with environmental issues. The Fund’s General Counsel, François Gianviti, disposed of that objection in a Board seminar in March 1991. Article I of the Articles of Agreement, in his explanation, implied that “while the Fund was not concerned with the environment as such, it was concerned with the economic and financial consequences of environmental degradation.”¹³⁷

The second issue was more subtle. If the Fund were to hire environmental specialists, how would it use their findings? Would it deal evenhandedly with industrial and developing countries, or would it become part of a broad campaign to force environmental standards on developing countries that needed to borrow from the Fund? After two Board meetings on the subject, only six Directors, holding less than 40 percent of the voting power, were willing to support Camdessus’s proposal. Of those, only one was from a developing country—Tanya Sirivedhin (Alternate, Thailand). She conditioned her support on the understanding “that Fund environmental scrutiny could be applied under Article IV consultations with any member—with or without a Fund-monitored program—if the circumstances seemed to warrant it.”¹³⁸

As soon as nongovernmental environmental organizations learned of this internal debate, they began lobbying for Fund involvement. One major group, Friends of the Earth, wrote letters to Executive Directors in early February urging a positive response, but that served only to strengthen the opposition.¹³⁹ Shortly afterward, Camdessus toned down his proposal in four ways. First, the Fund would not undertake any original research on links between macroeconomics and the environment. Instead, it would “monitor relevant research [by others] and channel information to staff members in area departments.” Second, he dropped the plan to designate the specialists as a separate unit within FAD and promised that they would instead “be fully integrated within the work of that department.” Third, he proposed that there be no “environmental conditionality” and that the new approach “be applied to all members in an evenhanded way.” Fourth, he agreed that “the decision to devote Fund resources to environmental issues will be given minimal publicity.”¹⁴⁰

With these modifications, a majority of the Executive Board reluctantly agreed to go along. Several chairs that had earlier opposed any move in this direction, including Japan and Saudi Arabia, now were willing to accept it. Others, including Australia and China, remained opposed.¹⁴¹ Enthusiasm, however, was notable in its absence. The combination of Camdessus’s infectious promotion and U.S. pressure for action had simply worn down the opponents. The Board made no formal decision, and management issued no public announcement.

Once this policy was accepted, the staff began including brief summaries of environmental issues in Article IV consultation reports. To ensure that no one could accuse them of not being evenhanded, one of the early examples was the 1991 consultation with the United States.

In discussions held in Washington in May and June 1991, just two months after the Board agreed to the proposal, the staff suggested to the U.S. authorities that they should consider an increase in gasoline and other energy taxes as an environmentally helpful way to reduce the fiscal deficit. In 1992, the staff raised this issue again and noted that energy taxes aimed at reducing carbon dioxide emissions by 6–10 percent by 2000 would yield up to 1 percent of GDP in additional revenues. The environment then became a routine part of the annual discussions for a few years. In 1993, the new Clinton administration stressed its view that environmental concerns “should play an important role in economic policy considerations,” subject to a cost-benefit analysis.¹⁴² After that, coverage of the topic gradually died down. For the next five years, references to environmental policy in the U.S. reports were limited to brief explanations of administration plans with respect to balancing environmental concerns against its pursuit of bilateral and regional free trade agreements. The 1999 report included no reference at all to the environment.

Consultations with developing countries often reported favorably on governments’ efforts to preserve the environment. The 1991 report on Malaysia, for example, noted that

Malaysia’s environmental policies are focused on the preservation of virgin forestry resources. … Malaysia has begun reducing log exports and encouraging higher value-added downstream processing through the application of quotas and fiscal incentives in the wood industry. The authorities also argued, however, that environmental issues should generally be placed in a more balanced historical and regional perspective, including the role and responsibilities of industrial countries in this respect.¹⁴³

Similarly, in 1995 the staff praised the Dominican Republic for taking actions to deal with long-standing environmental problems:

Regarding environmental issues, the authorities … explained that key … issues are deforestation, degradation of the urban environment, and the impact of certain unplanned population settlements in the tourist areas that seem to be negatively affecting the coastal ecosystems. A National Environmental Action Plan has been elaborated with the assistance of UNDP to address these problems as well as to improve the management and preservation of river basins. A commission headed by the Vice President of the Republic was recently created to coordinate and supervise this plan.¹⁴⁴

Despite the intended prohibition on environmental conditionality, the Fund did follow up on negative assessments in a few cases in the second half of the decade. As was true generally, the Fund’s influence in this field was bound to be more effective when a country was seeking to borrow than when it was merely consulting on the quality of its policies.

IMF advice to Cambodia in the mid-1990s illustrates the Fund’s delicate balancing act between demonstrating concern for the environment and steering clear of overt environmental loan conditions. As part of the 1992 UN-brokered peace agreement in Cambodia, the transitional authorities agreed to ban log exports, partly for environmental reasons (deforestation was causing massive flooding) and partly to restrict the flow of money to military and paramilitary forces including the Khmer Rouge. When the Fund resumed normal relations with Cambodia soon afterward, the staff included this export ban among its policy concerns. It soon transpired that illegal log exports were continuing to occur and that some state-owned enterprises were simply exporting sawn timber instead of logs. The staff called attention to the problem in its 1993 report on the Article IV consultation discussions.¹⁴⁵

In February 1994, the authorities submitted a Letter of Intent and an accompanying Memorandum of Economic and Financial Policies as the basis for the first tranche of a three-year ESAF arrangement. The staff team on Cambodia, led by Owen J. Evans (Advisor, Central Asia Department), was determined to make this arrangement the Fund’s first “green program,” and the team asked the authorities to set out their environmental policies, including on logging, as part of the documentation. Paragraphs 42 and 43 of the memorandum read as follows:

42. Environmental concerns are receiving increased attention from the Cambodian Government. In late 1993, the Secretariat of State for the Environment was established as a permanent government agency, with authority to supervise and develop a National Protected Areas System covering 3.3 million hectares designated for national parks, wildlife sanctuaries, protected areas, and multiple use management areas. During 1994 the Government intends to formulate a National Environmental Action Plan in consultation with the World Bank and other donors.

43. The forestry sector faces particular environmental challenges. Although Cambodia has a relatively large forest area by regional standards, logging for export timber and fuel-wood is a concern for the Government. Thus far logging has been controlled largely through export restrictions, but in 1994 the Government will investigate ways to achieve its environmental goals more directly.¹⁴⁶

The ESAF arrangement took note of the memorandum, but it did not include fulfillment of the plans set out in those two paragraphs in the list of specific conditions that could derail the scheduled disbursements under the arrangement.¹⁴⁷ In December 1995, Deputy Managing Director P.R. Narvekar visited Cambodia and noted that the illegal export of logs was depriving the government of “much-needed revenue,” but he did not focus on the environmental consequences.¹⁴⁸

This low-key approach changed in 1995, when the staff noted that forests were “Cambodia’s largest and most valuable natural resource. … The pace of deforestation has become a prime environmental concern. Uncontrolled logging, mostly for export, has reduced forest cover from around 74 percent of land area in the early 1970s to an estimated 30–35 percent today.”¹⁴⁹ In response, the Fund cooperated with other multilateral agencies to persuade the authorities to take stronger action. The policy memorandum for the second year of the ESAF arrangement included a more specific promise from the government:

An area of especially great concern to the Government is to avoid an environmentally unsustainable rate of cutting of trees. To this end, the Government will begin to prepare a forest management code in 1995. … The Government has agreed to formulate, in consultation with the World Bank, [the International Tropical Timber Organization], and other contributors, a program and a timetable for a National Environmental Action Plan.¹⁵⁰

Early in 1996, the government entered into new agreements to export logs, prompting the Fund to question the authorities’ commitment to implement the agreed-on forestry policies. Specific actions were then required to be taken before the Fund would make the next disbursement under the ESAF arrangement. When the authorities failed to take those prior actions on time, the Fund interrupted and eventually canceled the arrangement. In doing so, it emphasized the fiscal rather than the environmental consequences of the failure.¹⁵¹ Throughout this three-year period, the Fund prodded the authorities—first gently and then more forcefully—to limit the environmental damage from the stressed condition of the economy, all the while trying to avoid basing its advice explicitly or exclusively on this controversial issue.

The Fund adopted similar approaches in other countries where environmental concerns were paramount. Deforestation was an important issue in numerous developing and transition countries all over the world. Altogether, IMF staff reports discussed deforestation in 71 countries at some time in the 1990s. The Fund also raised concerns about a wide variety of other environmental issues such as unsustainable fishing yields (e.g., in Mauritania and Senegal), depletion of water resources owing to underpricing of water usage or insufficient penalties for polluting activities (e.g., in Ethiopia), and environmental degradation from inappropriate industrialization (e.g., from aluminum production in Tajikistan). In most cases, the Fund’s concerns in these areas reflected those of other agencies—principally the World Bank, but also the UN and its specialized agencies—with which the Fund staff consulted in its preparatory work on each country.¹⁵²

Germany

The largest and wealthiest country in the European Communities came under tremendous financial pressure at the beginning of the 1990s. As the physical barriers symbolized and anchored by the Berlin Wall crumbled, hundreds of thousands of people emigrated from the long-isolated and depressed German Democratic Republic (East Germany) into its much freer and richer western sibling, the Federal Republic of Germany. In May 1990, the two countries signed a treaty of economic and monetary union to make the deutsche mark (DM) the common currency of both. East German ostmarks would be exchanged for DM at parity up to a specified level per person and at a 1:2 rate for the rest. In October, the two countries merged, and East Germany ceased to exist as a separate state.

The economic and financial burden of German unification was large enough to induce the authorities to consider asking for financial assistance for East Germany from the IMF. During a March 1990 Executive Board discussion of the need for an increase in the Fund’s financial resources, the German Executive Director, Guenter Grosche, noted that “his authorities had foreseen a need to draw on Fund resources to facilitate adjustment in the German Democratic Republic.” Now that unification was proceeding, “while he expected that Germany would not need to enter into a Fund program in the foreseeable future, that possibility could not be totally ruled out.”¹⁹³ Although Germany did not end up asking for help, it was incumbent on the Fund to respond more generally to these dramatically changed circumstances.¹⁹⁴

The crucial consultation discussions with Germany, led by Manuel Guitián (Deputy Director, EUR), took place in Bonn and Frankfurt in the first part of June 1990, after the German authorities had finalized plans for unification and for the currency conversion. The German authorities were in the middle of preparations for implementing the currency conversion, and they were committed to engineering a full economic, social, and political unification as soon as possible. The staff noted that the conversion of claims, salaries, debts, and so forth into DM was going to be expensive and potentially inflationary, but they did not question the choice of conversion rates. Instead, the staff report counseled more aggressive fiscal tightening, particularly through tax increases and containment of subsidies.¹⁹⁵ If the entire burden of inflation control were to fall on monetary policy, interest rates could rise sharply, putting a strain on Germany’s neighbors. Without price and interest rate stability in Germany, the viability of the ERM would be gravely threatened, with potentially serious adverse consequences for all of Europe. Two weeks after the conclusion of the staff mission, Camdessus went to Germany to reinforce this message in meetings with the finance minister and the president of the Bundesbank.¹⁹⁶

The authorities viewed the Fund’s advocacy for fiscal tightening as excessively strict. Without large subsidies, the rise in unemployment in the eastern states was likely to be devastating and was expected to perpetuate the massive labor migration to the west. Both sides agreed, however, that while a realignment of exchange rates within the ERM (effectively revaluing the DM) would help the stabilization process, that advantage was outweighed by the risk of weakening the European commitment to maintain firm parities and advance toward EMU.

The staff continued to make the case for policy adjustments in 1991 when Patrick de Fontenay (Deputy Director, EUR) led the first staff mission since unification.¹⁹⁷ The fiscal deficit had ballooned, making the Bundesbank’s task of maintaining price stability much more difficult. As the staff report noted, “much of Europe has come to depend on the firm anchor of German monetary policy.” Without greater fiscal discipline, monetary discipline would be continually threatened.¹⁹⁸

The 1992 consultation discussions took place both before and after the September ERM crisis. At the time of the first mission, led by Jacques Artus (de Fontenay’s successor as Deputy Director, EU1)¹⁹⁹ in early September, the previous missions’ fears were being realized. The high interest rates that had resulted from the continuing fiscal deficit were strengthening the DM in exchange markets, and the effort to keep up with it was weakening economic activity in other countries participating in the ERM.²⁰⁰ Speculative pressures against the ERM parities were building up rapidly. Shortly after that mission concluded, the crisis erupted. When the dust settled, the staff returned for follow-up discussions, mainly to assess how the financial landscape had changed. The basic message on macroeconomic policy—tighten fiscal policy to relieve the pressure on monetary policy—remained the same.

The staff and the German authorities shifted roles a bit in 1993. The economy was now in recession because the earlier appreciation of the DM had taken its toll on international competitiveness. Bundesbank officials were nonetheless still focused primarily on price stability, while the staff suggested that the time had come for some easing to restore the momentum of aggregate demand. The mission—again led by Artus—was still calling for a tightening of fiscal policy, but it was more open than before to a preemptive easing on the monetary side without waiting for fiscal action.²⁰¹ By the time of the next mission, in May 1994, the ERM crisis had passed. The widening of the intervention bands had restored calm in exchange markets, and the German economy was recovering from recession.

Had the Fund given the right advice? From a macroeconomic perspective, it had. With a different mix of monetary and fiscal policies, the German authorities could have maintained the pace of aggregate demand without relying so heavily on high interest rates to preserve price stability. That approach would have relieved pressure on exchange rates and might have averted the ERM crises of 1992 and 1993. From a structural perspective, the authorities’ objections to that course also made sense. The costs of unification had to be met in part by borrowing, and that required larger fiscal deficits than would otherwise have been optimal. A less aggressive response to the severe imbalance in initial conditions in the eastern and western Länder would have seriously delayed and could have derailed the successful integration of the two Germanies. The rejection of an early realignment may have been questionable, but the rationale for it was strong at the time. All the good options were hemmed in by real constraints.

The United Kingdom

On October 5, 1990, the government of the United Kingdom announced that the pound sterling was joining the ERM, though with wider bands (± 6 percent around central parities) than most other members were using (± 2.25 percent). This decision ended a multiyear period in which the British authorities had experimented—without much success—with various strategies for stabilizing the exchange rate. The “medium term financial strategy” introduced at the beginning of the 1980s, which aimed to stabilize prices by controlling the growth rate of the money stock, had led to very wide fluctuations in the exchange rate. For a year in 1987–88, the strategy had been to “shadow” the DM to prepare the way for eventual ERM participation. When that strategy led to unstable domestic conditions, the exchange rate was again allowed to float to a market-determined level. By 1990, a coalition of strong supporters of financial integration with Europe and others who were frustrated with the lack of a good alternative finally won the day.²⁰² The triumph was short-lived.

The Fund supported the authorities’ intention to join the ERM. In fact, for several years, the staff had been quietly encouraging the authorities to prepare for such a step. The only questions were when and at what exchange rate. When the 1989 consultations were concluded in March 1990, every Director who addressed the subject urged an early entry, but a few—including Bernd Goos (Germany)—cautioned that Britain should first bring down its high rate of price inflation.²⁰³ When the Board next discussed the United Kingdom, after the entry was a fait accompli, a few Directors were still nervous about the consequences. With British inflation roughly double that on the continent, did it really make sense to hitch the exchange rates together? However, the prevailing view by far was support for the staff’s conclusion that ERM participation would enhance the credibility of the authorities’ determination to reduce inflation, and that it was indeed the only hope for success.²⁰⁴ In February 1992, the Executive Board again agreed with the staff that adherence to the ERM was the right strategy and that the pound should be brought into the narrow band as soon as the inflation differential could be substantially eliminated.

Even with the flexibility afforded by the wide band, the pound was vulnerable to a speculative attack against the ERM parity. That vulnerability increased with every month that prices rose more rapidly in Britain than on the continent. Because of the sheer size of the British financial system and of the foreign exchange reserves held by the Bank of England, neither the authorities nor the staff of the IMF seems to have regarded an attack as a serious threat in 1991 or 1992. By August 1992, however, the pound was trading near the floor of its trading band, while the whole ERM was coming under pressure owing to a rise in German interest rates and growing doubts about the extent of public support for EMU. Speculation turned into an attack and then into a rout in mid-September, when Italy was forced out of the ERM on September 14. After the Bank of England tried desperately to defend the pound by raising short-term interest rates sharply, it gave up on Black Wednesday, September 16. The investor and hedge fund manager George Soros soon became known as “the man who broke the Bank of England” when he revealed that he had led the attack by betting $10 billion against the pound and that he had made nearly $1 billion in profits when the Bank was forced to abandon the fight (Kaletsky, 1992, p. 1).

Following withdrawal from the ERM, the British authorities knew they needed new rules to replace the discipline of the European system. They therefore adopted an inflation-targeting regime. They remained committed in principle to an eventual resumption of ERM membership, but only after a convergence of cyclical economic conditions vis-à-vis Germany. The staff mission that arrived in London in early December, led by Artus, endorsed the inflation-targeting regime but cautioned that fiscal policy would have to be tightened substantially for it to succeed.²⁰⁵

For the next few years, a crucial question was how to gain greater credibility for British macroeconomic policies so that robust and noninflationary growth could take hold. Inflation targeting was working reasonably well but seemed fragile. Beginning in 1994, the staff urged the authorities to enable the Bank of England to implement monetary policy independently of the Treasury, as one way to establish a more credible anti-inflationary policy regime.²⁰⁶ The authorities resisted that advice until after the May 1997 parliamentary elections, which ended 18 years of rule by the Conservative Party and elevated the Labor Party’s Tony Blair to the premiership. Blair’s chancellor of the Exchequer, Gordon Brown, was a strong advocate of central bank independence, and he granted it to the Bank of England within a few days after taking office.

Another of Brown’s first acts as chancellor was to announce a set of five tests that would have to be met before it would make sense for Britain to adhere to EMU and adopt the euro as the national currency. Because none of the tests was met at the time and the prospects were somewhat remote, this tactic enabled the government to maintain a formal commitment to EMU while postponing it indefinitely. This policy, too, was endorsed by the Fund, and the rest of the decade’s consultations proceeded without major controversy.²⁰⁷

France

Meanwhile, France—the other major power in the EMS, along with Germany—was firmly committed to a policy of maintaining parity with the DM, even if that meant pushing interest rates higher than might seem desirable for domestic policy purposes.²⁰⁸ Throughout the early 1990s, Article IV missions endorsed that policy and marveled at the sight of France achieving inflation rates below those of Germany. Consequently, no major policy issues arose in the discussions. Instead, the crucial meetings came after the ERM crisis, which France survived without a devaluation but with a legacy of macroeconomic imbalances that were difficult to ignore.

The Maastricht Treaty specified several “convergence criteria” for countries intending to enter into EMU in 1999, including that they must make steady progress toward a sustainable fiscal deficit of no more than 3 percent of GDP by 1997. France’s deficit was close to 6 percent in 1993, only partly because the economy was in recession. The following year, the Fund began to question whether the authorities’ plans to converge quickly enough to the target were based on credible assumptions. With high unemployment (judged by the Fund to result from structural rigidities in labor markets, not from inadequate demand) and the prospect of presidential elections less than a year away, the government had not yet announced the specific spending cuts that would be needed to stay on course. The staff pronounced the required cuts to be “severe,” and it projected that the deficit would remain well above the convergence path. With characteristic diplomacy, however, the staff report concluded that it “welcomes the strong assurances” by the authorities that France would meet the targets.²⁰⁹

The fiscal debate intensified in 1995, when a staff mission led by Artus arrived a few weeks after the election of Jacques Chirac as president. Chirac’s succession to the post held for 14 years by François Mitterrand completed a shift to the right that had begun with the parliamentary elections of 1993. The election also removed the political necessity of promoting short-term over longer-term goals for economic policy. With the fiscal deficit still clinging stubbornly to 6 percent, reducing it was becoming imperative if France hoped to meet the Maastricht target and avoid a possibly debilitating panic in foreign exchange markets. Accordingly, Artus and his team set out a detailed “adjustment scenario” for fiscal policy and urged that it be incorporated into the government’s forthcoming budget.²¹⁰

On this occasion, the staff’s analysis appears to have made a difference. At the outset of the Executive Board meeting concluding the 1995 consultations, Marc-Antoine Autheman (France) averred that the points made by the staff, “which echoed the opinion of the Bank of France, were convincing.” The government had planned to present a budget that was projected to bring the deficit down just to 3 percent of GDP by 1997, but now it would be more aggressive in cutting spending and more ambitious in its deficit reduction. He acknowledged that the “credibility of our fiscal policy has suffered from the lack of strict implementation of past commitments.” This time would be different, “because my authorities do not consider nonconvergence an option.”²¹¹

That lack of credibility was evident in the staff’s response to the budget announcement and in the responses of other Directors to Autheman’s assurances. After reviewing the budget for 1996, the staff concluded, “Much more is needed and time is running short.”²¹² At the Board meeting, Directors piled on. J. Onno de Beaufort Wijnholds (Netherlands) suggested that meeting the government’s deficit target by 1997 would “require a ‘tour de force’.” Stefan Schoenberg (Germany) suggested that the French social security system was in need of “radical reform.” Huw Evans (United Kingdom) eruditely reminded his colleagues that the “immediate cause of the French revolution was the fiscal crisis of 1788,” which he thought was not too different from that of 1995. Karin Lissakers (United States) picked up that theme by quoting the always quotable American baseball legend Yogi Berra—reading about France’s fiscal problems was “déjà vu all over again.” At the end of the day, Fischer summed up the Board’s sense of urgency by noting that “France’s task was now truly historic … [as] the future of European monetary integration lies largely with France. For the authorities and for the French people, fiscal convergence is a challenge that must be met.”²¹³

It was met. Proving the skeptics wrong, Chirac and Prime Minister Alain Juppé tackled the social security problem and took other measures to strengthen the fiscal accounts. The staff mission for the 1996 consultations—again led by Artus—found that the deficit was likely to be held to 4 percent and then keep falling. That defused much of the worry about the convergence target for 1997, and from that point on the debates were more muted. In the end, the deficit for 1997 was 3 percent, although the target was met only by a one-time transfer of privatization receipts. That year, the EU adopted a new set of more flexible and dynamic targets for EMU convergence, in the Stability and Growth Pact. France continued to meet those targets through the end of the decade.