12 Boats Against the Current: Coping with a Global Tide

James Boughton
Published Date:
March 2012
  • ShareShare
Show Summary Details
F. Scott Fitzgerald

So we beat on, boats against the current, borne back ceaselessly into the past.

F. Scott Fitzgerald

The Great Gatsby


The East Asian crisis of 1997–98 was the apex of a wave of financial breakdowns extending from Mexico in 1994 to Argentina in 2002. The spillover or contagion effects from the “tequila” crisis were significant around the world but had major macroeconomic impacts in only a few countries in Latin America. The East Asian crisis was different, striking in multiple countries thought not to be particularly vulnerable, then spreading quickly throughout the region. Within a few months of its outbreak, analysts and investors began to reassess the risk of placing financial capital in emerging markets, regardless of geographic proximity to the crisis epicenter. The undercurrent from the withdrawal of capital threatened to undo the economic achievements of developing countries everywhere.

East Asia

The Asian crisis was not limited to the three countries—Thailand, Indonesia, and the Republic of Korea—discussed in Chapter 11. The major financial centers in Asia, including Tokyo, Hong Kong SAR, and Singapore, felt the effects but had ample resources to manage the challenges and ride out the downturn. Financial contagion hit two other developing countries—the philippines and Malaysia—particularly hard, and the low-income countries in Indochina suffered from the regional slowdown in aggregate demand. From the broad perspective of the IMF, the key to limiting these effects and keeping the downturn from getting out of control was to ensure stability in China, the largest emerging market of all.


Developments in China had substantial effects, not only on its Asian neighbors but on the global impact of the Asian crisis. By not devaluing the yuan in response to the depreciations of other Asian currencies, the Chinese authorities helped prevent the financial crisis from exploding into a generalized currency meltdown. A decade later, China’s adherence to a tightly managed exchange rate would come to be seen by the U.S. authorities and many external analysts as a destabilizing contributor to global payments imbalances. In the late 1990s, it was rightly viewed as a calming source of stability.

Officially, China had a managed-float regime for its exchange rate. In practice, from the beginning of 1995 the authorities had managed the rate tightly to maintain it close to 8.3 yuan per U.S. dollar.1 Because the country’s economic growth depended heavily on exports, IMF officials and other external analysts feared that China would try to prevent a real effective appreciation by allowing the rate to move in response to the depreciations taking place elsewhere in the region. Whether the Chinese authorities ever seriously contemplated devaluing the renminbi is difficult to know. The renminbi had depreciated substantially in nominal and real terms in the early 1990s until the authorities unified the foreign exchange market and started managing the rate against the dollar at the beginning of 1994. Since that time, the fixed nominal rate had produced a steady appreciation in real effective terms, but the real rate remained well below its early 1990s levels.2

On several occasions in the second half of 1997 and the first few months of 1998, the Fund’s Managing Director, Michel Camdessus, met with the Chinese authorities to urge them to hold the exchange rate steady. On each occasion, they assured him they understood the importance of stability and intended not to alter their policy. During the same period, U.S. officials including President Bill Clinton made similar direct appeals to their Chinese counterparts. They also received reassuring responses.3 Even if the Chinese intended all along to keep the rate where it was, this quiet but public diplomacy helped to preserve a measure of calm in otherwise very nervous financial markets.

The other currency-related issue for China was the strength of the Hong Kong dollar. At the same moment that Thailand was preparing to devalue the baht at the beginning of July 1997, the United Kingdom was formally handing over control of Hong Kong to China, at which time the territory would become Hong Kong Special Administrative Region of China, or Hong Kong SAR. The Hong Kong Monetary Authority (HKMA) had been operating as a currency board since 1983, maintaining the exchange rate at HK$7.8 per U.S. dollar. Adhering to this fixed rate through the market uncertainty associated with the hand over and the tumult of the regional financial crisis was going to be a major challenge. The Hong Kong authorities expressed their determination to persevere, and the Chinese authorities expressed their support and their commitment not to interfere.4

The Hong Kong dollar came under speculative attack, first in July and August 1997 and then again in October. In each case, the authorities drew on their ample stock of foreign exchange reserves to intervene in support of the currency. They also raised interest rates and mopped up excess liquidity in the economy. The defense succeeded, but at the expense of a sharp downturn in economic activity and a rise in unemployment. Real GDP declined by more than 5 percent in 1998 before beginning to recover in 1999; the unemployment rate rose from 2.2 percent in 1997 to 4.7 percent in 1998 and 6.3 percent in 1999.

The IMF strongly endorsed the Hong Kong authorities’ commitment to pursuing tight monetary and fiscal policies while persisting with the fixed exchange rate. The staff noted that the currency board was well run and that it provided a solid anchor for expectations. In the immediate aftermath of the hand over, the currency board provided assurance that Hong Kong SAR would continue to run economic policies independently from Beijing. In the middle of the regional financial crisis, it provided assurance that Hong Kong would preserve its financial strength and stability.5

The real test for the HKMA came in August 1998, when a collapse of confidence severely depressed equity prices just as the Russian financial crisis was exploding. Concluding that the problem was a speculative attack rather than weak economic fundamentals, the HKMA made the unusual decision to buy equities and futures contracts for its own account. That move stabilized market conditions somewhat, but it also highlighted the vulnerability that resulted inevitably from the small size of Hong Kong SAR’s completely open equity and foreign exchange markets. If a few well-capitalized hedge funds could easily attack these markets, would it not be better to adopt a more flexible policy with respect to the exchange rate? That question was hotly debated in Hong Kong SAR after August 1998, but in the end the authorities prevailed and maintained the unchanged exchange rate peg.

The staff unreservedly supported this decision.6 So, for the most part, did the Executive Board, although several Directors suggested that the authorities should be developing a strategy to exit from the exchange rate peg at an appropriate time.7 Soon afterward, however, economic growth resumed in Hong Kong SAR, and the peg survived.

The Philippines

In 1997, the Philippines was reaching the end of a 35-year stretch in which it had been a nearly continuous debtor to the IMF. Since the fall of 1994, the authorities had been successfully reforming the economy through a program supported by the Extended Fund Facility (EFF). Even more impressive, they had been able to treat the EFF arrangement as precautionary and had financed their balance of payments without further financial help from the Fund.8 By June 1997, they had repaid nearly all of their earlier loans (Figure 12.1), and their remaining obligations totaled just 18 percent of quota, the lowest percentage since 1974. The Thai crisis interrupted this march to the exits, but—thanks to timely action by the authorities—it did not bring it to a halt.

In February 1997, a staff mission reached agreement with the authorities on economic policies for the rest of the year. At that time, the staff expected the extended arrangement would remain in effect until it expired on schedule in June and that the authorities would continue to treat it as precautionary and then end their era as a borrowing country. That continued to be the Fund’s view until the Thai baht came under speculative attack in the middle of May. As investors began to reassess the strength of other currencies in the region, the Philippine peso was among the first to be downgraded.

The Philippine authorities reacted swiftly to this contagion by raising overnight interest rates. The IMF’s Executive Board was about to complete the final review of the 1994–97 arrangement, but put it on hold until the extent of this new difficulty could be clarified. When the initial interest rate response proved inadequate, the governor of the central bank, Gabriel Singson, called Camdessus and asked for additional support. The Managing Director immediately sent a mission to Manila, led by John E. Hicklin (Assistant Director, Asia and Pacific Department, or APD), to negotiate terms for an extension of the existing arrangement.9

Figure 12.1.The Philippines: Use of Fund Credit, 1989–2000

(In millions of SDRs, monthly data)

Source: International Financial Statistics.

Note: EFF = Extended Fund Facility; SBA = Stand-by arrangement.

The staff determined that the key to avoiding a financial crisis was exchange rate policy. The government had recently implemented a number of difficult measures to strengthen both the banking system and its own finances. The success of these measures had not yet been tested, and many domestic banks had large short-term liabilities in foreign currencies, making the currency vulnerable to a speculative attack. For a year and a half, the authorities had been managing the exchange rate tightly, in a de facto peg to the U.S. dollar. In the middle of 1997, they believed they were managing well enough to continue pegging, but Hicklin and others in the IMF quickly concluded this was going to be impossible.

For the IMF, preserving the fragile gains the Philippines had made in the mid-1990s was an important goal. It was crucial not only for the future of the Philippine economy but also for stability throughout the Pacific region in the wake of the crisis engulfing Thailand. It was also an important test of whether the Fund really could help a country with long-standing financial and developmental weaknesses to get on its feet and graduate from dependence on multilateral assistance.

Camdessus realized that convincing the authorities to let the value of the peso float relative to the dollar was going to be a tough sell. To reinforce the message, he asked Bijan Aghevli (Deputy Director, APD) to join Hicklin’s mission in Manila and impress upon the authorities the importance of abandoning or at least loosening the peg. He also asked Shigemitsu Sugisaki (Deputy Managing Director) to deliver that message directly to Singson. Sugisaki and the governor were both scheduled to go to Hong Kong at the end of June for the ceremonial handover of sovereignty from the United Kingdom to China. There, they could meet quietly and discuss what to do next.

As expected, Singson was reluctant to let the exchange rate float because he feared that speculation would get out of control and produce a ruinous overshooting. Sugisaki persisted, and after the ceremonies in Hong Kong the two traveled together to Manila for further discussions as part of the mission already in progress.

At 9:00 a.m. on Wednesday, July 2, Sugisaki, Aghevli, and Hicklin were meeting with Singson in the governor’s office in Manila when they received word that the Bank of Thailand had just announced it was no longer defending the value of the baht. Everyone in the room understood immediately that the Philippine central bank would have to follow suit. If they tried to maintain the existing rate against the dollar, speculators would quickly acquire all of the country’s foreign exchange reserves. Singson could not make such a decision on his own. As soon as they could, he and Sugisaki went across town to inform the president, Fidel V. Ramos, of this turn of events.

The market reaction was not immediate, but it did come. After a long weekend of hesitation, the government decided to allow some depreciation of the peso while trying to control the rate of change by continuing to intervene in the foreign exchange market. Predictably, that led to even more intense speculative pressure. On July 9, a full week after the onset of the crisis, Camdessus picked up the telephone himself and called Singson. By then the governor was ready to accept the Fund’s advice. The next day Singson met again with Ramos and got approval to act. On Friday, July 11, the peso was allowed to float.10

The staff mission wrapped up its work on July 11 and returned to Washington. In the end, the Fund agreed not just to extend the EFF arrangement but to augment the amount. The Fund was able to act quickly in this case by invoking—for the first time—the Emergency Financing Mechanism it had adopted in 1995 for coping with financial crises (Chapter 5). In mid-July, the Philippines borrowed $700 million from the IMF (SDR 508.75 million, or 80 percent of quota). This bold move did not completely stop the speculative pressure, and the peso’s value continued to fall, but at least the pace of depreciation now seemed manageable.

The onset of the regional crisis emanating from Indonesia and Korea in November 1997 worsened the outlook considerably. The peso tumbled sharply against the U.S. dollar. By early January, it was selling for less than 60 percent of its precrisis level. That, fortunately, turned out to be its low point. By the time a new program was negotiated with Fund staff in February, the authorities were already tightening monetary policy, confidence was beginning to be restored, and the peso was recovering. In April, the Fund approved a two-year stand-by arrangement to succeed the extended arrangement that had finally expired.11

In contrast to Thailand, the Philippines successfully protected itself from a major financial crisis by abandoning the fixed exchange rate while the central bank still had ample foreign exchange reserves. In contrast to Thailand and Korea, the authorities shared vital information with the IMF at an early stage and thus enabled the Fund to provide informed and timely policy advice. In contrast to Indonesia, early action to strengthen the financial system limited the extent of depreciation after the rate was allowed to find its own level. Contagion from the crises in those countries was a serious shock to the Philippine economy, but the outcome was manageable with only a continuation of routine lending by the IMF. The Philippines made several drawings on the 1998–2000 stand-by arrangement, at the end of which its debit position in the Fund peaked at just over $2 billion (SDR 1.56 billion, or 178 percent of quota). By then, the worst was truly over, and the country was able to repay all of that debt on time, completing the process in 2006.


Malaysia was the only IMF creditor country to be hit by market contagion from the East Asian crises. Even at the height of speculative pressure against the currency, Malaysia had no need to borrow from the Fund. Nonetheless, an ugly and very public dispute flared up over the Fund’s policy advice and Malaysia’s decision to impose capital controls in 1998 as part of its response to the crisis.

From the time it joined the IMF in 1958, the newly independent Federation of Malaya (expanded and renamed Malaysia in 1963) generally maintained a creditor position in the Fund. On a few occasions from 1977 to 1983, it drew on special facilities (the Compensatory Financing Facility and the Buffer Stock Financing Facility) to cope with low export prices. It repaid all of its loans on schedule by 1985. Throughout this period, Malaysia experienced strong economic growth and a remarkable reduction in the incidence of poverty.

Progress accelerated further in the late 1980s, when Prime Minister Mahathir Mohamad set the economy on a path of rapid industrialization. In 1991, he initiated an export-led growth strategy known as the New Development Plan and appointed a new economic team, led by Anwar Ibrahim as minister of finance. Two years later, he elevated Anwar to deputy prime minister. The export-led growth strategy was introduced at a propitious time, when international capital was beginning to flow freely to emerging markets after several years of net outflows. As one of the new “Asian tigers,” Malaysia quickly became a major recipient of these inflows, which fueled an investment boom that turned the country into a major center of production for export and transformed Kuala Lumpur into a showcase for modern Asian architecture. From 1988 through 1997, annual real output growth averaged more than 9 percent.

IMF staff reports broadly supported Malaysia’s aggressive growth strategy, but from 1991 on the staff quietly urged the authorities to be cognizant of the potential for the economy to overheat. By 1994, the inflow of financial capital was becoming difficult to manage. Inflation pressures were building up, and the first signs were emerging of a bubble in the property market. To contain the pressure, the authorities temporarily introduced controls on capital inflows.12 IMF officials expressed skepticism about this move but recognized they had no power, nor even a mandate, to stop it. Accordingly, the Fund’s response was muted and referred only to the desirability of ending the controls as soon as possible. Specifically, the 1994 Article IV mission report noted that “as such measures introduced distortions, they were not desirable in the longer term.” The authorities agreed, and they terminated the controls in August.13

Throughout the early and mid-1990s, the IMF had excellent relations with the Malaysian authorities. Staff reports generally called for a tightening of monetary and fiscal policies to counter inflationary pressures, but the overall message was always based on admiration for skillful economic management and the strong performance of the economy. Camdessus had a particularly high regard for Malaysia’s achievements, reinforced by his respect for, and personal friendship with, Anwar Ibrahim. The two men had similar views on economic policy, a shared sense of the universality of religious experience, and an urbane and philosophical approach to policymaking and diplomacy. In 1996, Camdessus twice visited Malaysia in response to invitations from Anwar, who introduced his friend to Mahathir. Both in Kuala Lumpur and on his return to headquarters, Camdessus was effusive in his praise for Malaysia’s record of harmonious ethnic relations, commitment to reducing poverty, and strong economic growth.14 In his public address, he praised Malaysia as “exceptionally successful” and promised to “support your efforts” to achieve the status of an industrial country “with every means at our disposal.”15

The IMF staff working on Malaysia also were generally upbeat. Their concerns about overheating and the risk of a real estate bubble were tempered by a respect for the high rate of domestic saving and the boom in productive investment fueling the economy’s growth. The banking sector was generally free of the weaknesses plaguing other countries in the region. External debt was largely long term and did not appear to pose any problems.16 Of all the emerging markets in East Asia, Malaysia was one of the least likely to get into financial trouble.

Trouble came, however, in the wake of the crisis in Thailand, beginning when the Thai baht came under speculative attack in May 1997. The central bank (Bank Negara Malaysia) raised interest rates and intervened in the foreign exchange market in a rapid response to limit contagion. A week later, an IMF staff mission led by David J. Robinson (Division Chief, APD) arrived to conduct the already scheduled Article IV discussions. Robinson urged the authorities to take more aggressive action to tighten policies, not because the stance of monetary policy was particularly loose but because of the elevated risk of a catastrophic loss of investor confidence resulting from the situation in Thailand. The Malaysian authorities decided that further action was unnecessary.17

The investment climate worsened further after Thailand devalued the baht at the beginning of July. Domestic banks and other local institutions in Malaysia began shifting assets out of the ringgit, and the resulting pressure on the exchange rate caused foreign investors to speculate against the currency as well. Bank Negara again reacted swiftly, raising interest rates sharply for about two weeks. By then the authorities realized circumstances in the region were deteriorating more fundamentally. They lowered interest rates and allowed the exchange rate to depreciate in value by more than a third, from about 2.5 ringgit to the dollar—the level that had prevailed for more than five years—to nearly 3.9 by the end of 1997.

As the exchange value of the ringgit fell, Mahathir embarked on a public rampage against “billionaire … currency traders,” whose activities were “unnecessary, unproductive and immoral.” In various forums, including during the IMF/World Bank Annual Meetings in Hong Kong SAR in mid-September, Mahathir singled out George Soros for particular scorn, even though Soros denied that his hedge funds had been selling ringgit. Mahathir criticized the IMF for tolerating such practices. He made a broad attack on Jews, accusing them of deliberately trying to undermine the Malaysian currency.18 The tone of these various remarks doubtless made many investors nervous and more reluctant to keep their money in Malaysia.

In counterpoint to Mahathir’s attack, Anwar made a conciliatory speech at the Annual Meetings in Hong Kong SAR and generally did his best to try to present a calmer and more positive image to international investors. The IMF also tried to restore calm. From November 1997 through January 1998, Camdessus made three more trips to Kuala Lumpur. He met with the prime minister and explained the IMF’s view that hedge funds were not the major culprit in currency markets and that suppressing speculation would do more harm than good.19 These efforts seem to have helped. The 1998 Article IV mission was advanced to January, and those discussions produced agreement on a modest tightening of macroeconomic policies that the Fund was willing to endorse informally.20 On January 16, Camdessus held a press conference in Kuala Lumpur in which he applauded the actions already taken and offered the Fund’s technical assistance and advice in formulating a more comprehensive package.21

Although the IMF and the authorities still enjoyed a productive working relationship in January 1998, the public perception of the Fund in Malaysia was deteriorating almost irretrievably. Fund-supported programs in Thailand, Indonesia, and Korea were viewed as disastrous failures. The Fund was thought to be forcing countries throughout the region into inappropriate and even ruinously restrictive policy changes. “One size fits all” was being bandied around as a pejorative description of the austerity many observers thought the Fund was imposing across East Asia and was now trying to import into Malaysia. When the infamous photograph of Camdessus apparently lording his authority over Indonesian President Suharto (see Chapter 11) splashed across the front pages of newspapers everywhere on January 16, 1998—the same day that Camdessus was arriving in Kuala Lumpur to express the IMF’s support for Malaysian policies—it solidified the perception that the Fund posed a serious threat to Malaysia’s sovereignty as well.

Notwithstanding these setbacks, quiet efforts at cooperation continued throughout the first half of 1998. At no time did the Malaysian authorities ask for financial assistance from the IMF. Nor did the IMF suggest to them that they should. Rather, the authorities wanted a positive assessment from the Fund so they could borrow more readily from other agencies, including the World Bank and the Asian Development Bank, and bilaterally from Japan. Before offering such an assessment, the Fund was asking for more policy actions.

Both Camdessus and the First Deputy Managing Director, Stanley Fischer, made further trips to Malaysia, and Fischer met again with Mahathir during a conference in Tokyo in June. Throughout this time, the Fund’s message to the authorities was that the steps taken so far to tighten policies in response to capital outflows were not going to be sufficient to stabilize the economy. As long as the East Asian region was in financial turmoil, Malaysia was going to suffer contagion effects in its financial markets until it protected itself with a sustainable and comprehensive package of economic policies. Mahathir rejected that message and argued instead that the fundamental problem was external: the power of unregulated financial market players to destabilize emerging-market economies. In July and August he began shifting policies in the opposite direction from that which the IMF was advising.22

Independently from any discussions with the IMF, the government developed a new plan during the summer of 1998 that aimed to impose a wide range of capital controls. Those controls would create some scope for monetary policy by insulating Malaysia’s financial markets from external pressures and thereby enable a shift toward more growth-oriented policies without suffering even further outflows. At the southern tip of the Malaya peninsula, international banks operating in Singapore were offering high interest rates for ringgit deposits. Some Malaysian officials were determined to shut that market down and stop the ringgit from becoming an international currency. Others, including both Anwar and the governor of Bank Negara, Ahmad Mohd Don, worried that doing so would also shut down foreign investment and seriously damage the economy. When it became clear that Mahathir supported the plan, Ahmad Don resigned.

On September 1, 1998, Ahmad Don’s successor at the helm of Bank Negara, Acting Governor Zeti Akhtar Aziz, announced the imposition of controls, but only after she had persuaded the prime minister to scale down the complexity and coverage of the original scheme. The principal new rules comprised a requirement that offshore ringgit accounts had to be repatriated within 30 days and a one-year holding period on ringgit-denominated nonresident accounts. With channels for outflows drying up, the exchange value of the ringgit jumped the next day. That increase enabled Bank Negara to peg the exchange rate at 3.8 ringgit per U.S. dollar, which was close to the market but was thought to undervalue the ringgit somewhat.23

These acts by themselves would not necessarily have been a problem from the IMF’s perspective, although Fund officials did not think them wise or appropriate under the circumstances. The Fund’s general policy on capital controls at this time (see Chapter 4, “Oversight of Capital Controls”) discouraged their use but accepted that they could be helpful in certain circumstances. The suitability of controls for Malaysia in 1998 could be disputed, but that dispute by itself should not have given rise to an acrimonious rupture.24

The situation became more worrying the next day when Mahathir dismissed Anwar from his posts and placed him under a form of house arrest. Anwar had objected to the controls, both because he believed controls in general to be unnecessary and counterproductive and because he concluded that this particular act was designed to benefit firms that were well connected politically.25 Up to this time Anwar had been widely expected to succeed Mahathir eventually as prime minister. His opposition to a major policy initiative could not be tolerated within the government. However, much of the international reaction interpreted his arrest as a signal that conditions and policies in Malaysia were about to deteriorate more fundamentally. Whether that reaction was justified is also a matter for debate, but the tension that arose between the IMF and the Malaysian authorities over the imposition of capital controls is difficult to explain without reference to the treatment of Anwar, which continued to worsen.26

The difficulty for Malaysia was that without an endorsement of the policy regime from the IMF, neither private nor official capital would be easy to attract. World Bank officials were equally concerned and were even more outspoken than the Fund. The Japanese government sided with Malaysia and offered new loans, but the U.S. and other governments either stayed on the sidelines or called openly for change.27 IMF officials avoided speaking in public about the treatment of Anwar and generally limited their comments on capital controls to pointing out the long-run dangers. At the Annual Meetings, for example, Fischer told reporters that “the reimposition of controls—the attempt that seems to be being made to cut the country off, cut the domestic financial system off from the international system—is not one that will do very much for Malaysia … over any sustained period.”28 Both publicly and privately, Fund management and staff persistently urged the authorities to phase out the controls and allow some flexibility in the exchange rate as quickly as possible.

Despite the public contretemps, the authorities heeded this advice. In February 1999, they replaced the ban on repatriation of financial capital with a tax on short-term outflows. They also accelerated banking sector reforms and kept the stimulative monetary and fiscal policies within the generous bounds determined by their relatively favorable initial conditions. Export growth, and output growth more generally, rebounded in the early months of 1999. As confidence returned, the economic recovery in Malaysia proved to be as strong and as durable as that being experienced elsewhere in the region.29

In July 1999, the Executive Board reviewed the Malaysian economy for the first time since the imposition of controls. Most Directors acknowledged that the Fund (along with “many observers”) had overreacted in its initial response. The staff also pulled back a bit and implicitly blamed the first reaction on the views of an amorphous “international community.” The Summing Up of the Board discussion concluded by noting the earlier error and adopting a more balanced view on the way policy should evolve in the future:

Directors broadly agreed that the regime of capital controls—which was intended by the authorities to be temporary—had produced more positive results than many observers had initially expected. They welcomed the pragmatic and flexible way in which Malaysia had implemented and adjusted the controls, notably by replacing the quantitative restrictions on the repatriation of portfolio investments by an exit levy in February 1999. A number of Directors expressed support for the authorities’ intention to maintain the control measures while preparing for an orderly exit from these controls. A number of other Directors, however, were more skeptical about the decision to impose capital controls, as they felt that the costs, in terms of an adverse impact on the prospects for recovery, may become more visible in the future. They therefore recommended that the authorities remove the exit levy applied to profits on portfolio investments. These Directors also considered that, since Malaysia is in a position of strength, an early exit would help to boost investor confidence in Malaysia and attract long-term capital.30

Perhaps this moderation could have ended the dispute, but it did not. For some years afterward, staff and management continued to argue that the controls had not been helpful, and the authorities continued to defend their actions. The perception of the Fund by the Malaysian public and by some officials as an overbearing institution that had offered totally wrong policy advice prevailed for years. Everyone concerned viewed the whole episode as an embarrassing dispute that seriously damaged the earlier positive and productive working relationship between the Fund and Malaysia.


The three low-income neighbors of Thailand known collectively as Indochina—Cambodia, the Lao People’s Democratic Republic (Lao PDR), and Vietnam—had all been borrowing from the IMF through the Enhanced Structural Adjustment Facility (ESAF) in the period leading up to the regional financial crisis. All three suffered from the crisis, mainly because it aggravated the economic effects of preexisting political problems. In its efforts to help these countries, the IMF focused more on the long-term objective of removing structural impediments to growth than on the short-term financial difficulties. Consequently, lending was slow to resume (Figure 12.2).


Cambodia resumed borrowing from the IMF in 1993 after two decades of devastating internal and regional conflicts (see Chapter 16). As with other transition countries, the Fund initiated its lending to Cambodia with a fast-disbursing loan through the Systemic Transformation Facility, in October 1993. It then shifted to concessional lending and approved a three-year ESAF arrangement in May 1994. Midway into the arrangement, however, it suspended disbursements because of a controversy over forestry management. As discussed in Chapter 4, the international community generally regarded Cambodia’s logging practices as environmentally unacceptable, and the IMF also regarded them as undermining long-term fiscal viability. The Executive Board repeatedly extended the arrangement while efforts to resolve the dispute continued. Then in July 1997, just a few days after the devaluation of the Thai baht initiated the regional financial crisis, former Prime Minister Hun Sen led a coup d’etat and resumed power. The ensuing political uncertainty and international isolation compounded the economic set-back. The economy continued to weaken for the next year, until new elections in July 1998 restored a measure of political stability. The Fund then approved a successor ESAF arrangement in October 1999, which eventually was fully drawn.

Figure 12.2.Cambodia, Lao PDR, and Vietnam: Use of Fund Credit, 1990–99

(In millions of SDRs, monthly data)

Source: International Financial Statistics.

Note: ESAF = Enhanced Structural Adjustment Facility; SAF = Structural Adjustment Facility; SBA = Stand-by arrangement; STF = Systemic Transformation Facility.

The Lao PDR

The Lao PDR successfully completed an ESAF arrangement in the spring of 1997, after which the attack on the Thai baht and the subsequent regional slowdown caused a serious reversal of fortune. Thailand was the major market for Lao exports, and when the Lao government tried to counter the spillover effect by pursuing expansionary policies, the currency (the kip) depreciated by even more than the baht. As inflation accelerated, the authorities requested a successor arrangement with the Fund and began formulating a new program as the basis for it, but conditions were too difficult.31 Discussions continued at the staff level, but the IMF did not resume lending to the country until 2001.


Vietnam began borrowing from the ESAF in November 1994, with a three-year arrangement scheduled to expire in November 1997. Throughout the third year of the arrangement, the Fund declined to approve the government’s policies, primarily because of dissatisfaction with the pace of structural reforms to privatize state-owned enterprises, restructure state-owned banks, and eliminate nontariff trade barriers. As the arrangement was about to expire, the regional crisis weakened demand for Vietnam’s exports and ushered in a slowdown of foreign direct investment. Economic growth slowed sharply, prompting the authorities to request a new ESAF arrangement. Despite the external shock, the Fund did not back away from its insistence on an acceleration of structural reforms. Only after more than three years of negotiations did it approve a successor arrangement, in April 2001.


The greatest influence of the Asian crisis on European economies was felt in the Russian Federation. As recounted in Chapter 7, while several countries in East Asia were still embroiled in crisis management, the IMF and others warned the authorities in Russia that they were vulnerable to the same kind of market pressures. In view of Russia’s severe and persistent weaknesses, both in fiscal policy and in oversight of the banking system, throughout the first half of 1998 the economy was clearly at the edge of a precipice. As the collapse of confidence in the economies of Thailand, Indonesia, and Korea gradually infected other emerging markets with similar problems, Russia faced a massive withdrawal of short-term financial capital. When the Russian financial system collapsed in August, many of its neighbors were increasingly at risk as well.

Central and Eastern Europe

The Baltic states and the other European countries of the former Soviet Union were adversely affected in three ways. First, many of them still had sizeable exports to Russia. The drop in demand directly worsened their trade balances and forced cuts in output. Second, investors generally feared that these countries were still at such an early stage of development and transition to a full market economy that they might have to resort to the same drastic measures (default and a large devaluation) as in Russia. Financial contagion thus aggravated the direct effect working through the trade accounts. Third, many of these countries still had fragile political coalitions set against nationalist or nostalgic-socialist oppositions. The risk of reversal of the political momentum for reform only made matters worse.

Chapter 8 discusses the progression of this crisis, country by country throughout the former Soviet region, and the IMF’s response to it. Ukraine suffered most severely from the financial contagion. The country, after Russia the second largest economy in the former Soviet Union, had enjoyed initial success in attracting foreign capital but—unlike the Baltic states—had not yet solidified its reputation as a western-oriented market economy. The rest of the region to the south and east was affected more by trade links to Russia. Even the low-income countries (the “CIS-7”32), which depended on official financing rather than on private capital markets, suffered large current account and output effects owing to the direct decline in exports to Russia, Ukraine, and others. For that group of mostly smaller countries, the IMF responded on a regional level by encouraging donor countries to help cover the estimated losses. Notably, on December 11, 1998, the Fund and the World Bank cohosted a donors’ meeting that raised $200 million for six of the CIS-7 countries, an amount estimated to compensate for half of the financing gap caused by the Russian crisis.33

Table 12.1.IMF Lending to European Countries Other Than the Former Soviet Union, 1990–99
CountryType of

Loan or

DateSize of Loan or


(Millions of

Amount Drawn

through end-1999

(Millions of

Czech RepublicSBA1993–94177.0070.00
Slovak RepublicSTF199364.3564.35
Bosnia andEPCA199530.3030.30
Macedonia, FYRSTF199412.4012.40
Source: International Financial Statistics.Note: CCFF = Compensatory and Contingency Financing Facility; ECM = External Contingency Mechanism; ENDA = Emergency Assistance for Natural Disasters; EPCA = Emergency Postconflict Assistance; ESAF = Enhanced Structural Adjustment Facility; EFF = Extended Fund Facility; SBA = Stand-by arrangement; STF = Systemic Transformation Facility.

Amount includes augmentations through December 31, 1999.

Source: International Financial Statistics.Note: CCFF = Compensatory and Contingency Financing Facility; ECM = External Contingency Mechanism; ENDA = Emergency Assistance for Natural Disasters; EPCA = Emergency Postconflict Assistance; ESAF = Enhanced Structural Adjustment Facility; EFF = Extended Fund Facility; SBA = Stand-by arrangement; STF = Systemic Transformation Facility.

Amount includes augmentations through December 31, 1999.

Elsewhere in central and eastern Europe, ongoing transitions left several countries financially vulnerable (see Chapter 5). In the first half of the 1990s, 10 or so countries in this region had borrowed from the IMF (Table 12.1).34 Some were new members, borrowing for the first time: Bulgaria beginning in 1991, followed by Albania (1992), the Czech and Slovak Republics (1993), the former Yugoslav Republic of Macedonia (1994), Croatia (1994), and Bosnia and Herzegovina (1995). Others were borrowing to help stabilize their economies during the early stages of a transition away from the Soviet system: Hungary (1990–93), Poland (1990–94), and Romania (from 1991 through the rest of the decade). After several years in which the IMF had done very little lending to European countries, Europe had again become a continent with substantial financial needs.35

Several European countries continued to borrow in the late 1990s, but mostly for reasons unrelated to global financial contagion. Two countries—Albania and the former Yugoslav Republic of Macedonia—were low-income members that shifted their borrowing from the Fund’s general resources to the less expensive and longer-term ESAF in the course of the decade. Two others—Bulgaria and Romania—were undertaking very difficult transitions and borrowed almost continuously throughout the 1990s. Bosnia and Herzegovina began borrowing in December 1995 following the signing of the Dayton peace accords that ended the regional war. A first stand-by arrangement was approved in 1998. Croatia ended its borrowing in 1997.


One European borrower did not fit this general transitional pattern. Turkey, like Russia, straddles Europe and Asia and is treated by the IMF as European. A member since 1947, Turkey began borrowing from the Fund in 1953. It entered into 10 stand-by arrangements in as many years starting in 1961 and continued borrowing through various facilities through 1984. By that time, Turkey had achieved a good measure of financial stability and independence and was able to manage its external debts on its own. It steadily repaid its debts to the IMF and did not borrow again over the next decade.

In the early 1990s, the Turkish authorities gradually loosened their grip on monetary and fiscal policies. The rate of inflation rose, and the balance of payments progressively worsened. The tipping point came in the first quarter of 1994, when investors fled the Turkish market and the value of the currency fell by half against the U.S. dollar in three months. In April, the government announced a new policy program aimed at stabilizing the economy quickly, and the authorities asked the IMF for financial and program assistance. In July, the Executive Board approved a 14-month stand-by arrangement for $740 million (SDR 509.3 million, or 85 percent of quota). That program and lending arrangement worked well through the following summer, after which political instability in Turkey paralyzed policymaking.36

A succession of short-lived Turkish governments tried repeatedly to get inflation under control and stabilize the economy. Beginning in mid-1997, the authorities and the IMF staff engaged in protracted discussions on a policy program that the staff could monitor while the authorities established a track record of good implementation. That effort finally succeeded in mid-1998, at which time Güneş Taner (minister of state for the economy) and Gazi Erçel (governor of the central bank) signed and published a 10-page memorandum of economic policies as the basis for a staff-monitored program covering the period through the end of 1999.37

Carrying out this ambitious program proved to be impossible. The lingering effects of the government’s lack of a credible track record on economic policy interacted with a series of external and domestic shocks. Just as the stabilization effort was beginning, Russia’s bond default induced a large outflow of capital from Turkey and a sharp slowdown in international trade. As recession took hold, the governing coalition collapsed in November 1998. By the first quarter of 1999, real output was down by more than 8 percent from a year earlier. A new government took office in June and began discussing a modified policy regime with the Fund, but then disaster struck. On August 17, a massive earthquake killed more than 15,000 people, left a half million people homeless, and caused extensive disruption to economic activity. Even by the standards of this highly earthquake-prone region, it was one of the worst natural disasters of the twentieth century.38

The IMF reacted swiftly by providing an emergency loan of $500 million (SDR 361.5 million, or 37.5 percent of quota) in October 1999. That loan was followed by approval of a rare three-year stand-by arrangement for $4 billion (SDR 2,892 million, or 300 percent of quota) in December, on which Turkey immediately drew about $300 million (SDR 221.7 million) (Figure 12.3). That extraordinary level of support—the maximum that the Fund could provide without invoking its “exceptional circumstances” provisions—responded to a strong commitment by the Turkish authorities to tighten financial policies sufficiently to get inflation under control, firm up exchange rate policy to anchor expectations until confidence could be fully restored, and undertake numerous structural reforms affecting almost every sector of the economy. The Fund required many of those reforms to be completed even before it approved the stand-by arrangement, and the government rose to the task.

The explicit goal of the IMF’s large financial commitment to Turkey in 1999 was not to cover an actual shortage of international reserves. Those balances were still at a comfortable level. Rather, the goal was to ensure debt sustainability through a rapid and continuing reduction in inflation and interest rates. More immediately, the goal was to convince financial markets that the international community stood ready to provide whatever financial support the authorities might need in the event of a speculative attack. However, that tactic worked only briefly: full restoration of investor confidence was a long way off, and Turkey would undergo a serious financial crisis in 2000–01 and would depend on large-scale IMF loans for years to come.39

Figure 12.3.Turkey: Use of Fund Credit, 1990–99

(In millions of SDRs, monthly data)

Source: International Financial Statistics.

Note: SBA = Stand-by arrangement.

Latin America

After the Russian bond default of August 1998, IMF officials realized that emerging markets all over the world might be vulnerable to a sudden withdrawal of financial capital. Latin America seemed most clearly ripe for the next speculative attack.

In late August, just a few days after the Russian default, Camdessus invited the finance ministers and central bank governors from all of the major financial markets in the Americas to come to the IMF to discuss how to respond to the market pressures already facing many of them. For the IMF to convene such a regional meeting on this scale was unprecedented, and to succeed the meeting would have to occur extremely quickly. Before the end of the month, the ministers and governors from 11 countries40 had accepted Camdessus’s invitation and were making plans to go to Washington for what was to become the first major effort at “regional surveillance” by the IMF.41

The meeting was held on September 3–4, 1998, in the conference hall at IMF headquarters where the Interim Committee met twice each year. Fund officials made presentations on the likely impacts of ongoing developments in the world economy, and country officials discussed their strategies for coping with those developments. The Presidents of the World Bank and the Inter-American Development Bank (James D. Wolfensohn and Enrique V. Iglesias, respectively) also participated. At its conclusion, Camdessus, Wolfensohn, and Iglesias held a press conference and summarized the outcome of the meeting.42

Formally, the main result was a set of commitments by participating countries to strengthen fiscal and monetary policies in response to the Russian crisis, the subsequent turmoil in world financial markets, and other adverse external shocks. Perhaps the greater benefit, though, was that finance officials at the highest level had seized the opportunity to come together and hear from each other and from the leading multilateral agencies about measures being taken and measures still needed. Although each country had to take its own actions in light of its own circumstances, knowing what one’s neighbors were doing had the potential to improve the outcome considerably.

Of the large countries in Latin America, the two most vulnerable to contagion in 1998 were Brazil and Argentina. Among the smaller countries, Ecuador was especially vulnerable. In each case, the IMF followed up the regional surveillance effort with intensive staff work in the country.


Following the shock from the Mexican crisis of 1995 (see Chapter 10), the Brazilian monetary authorities focused on solidifying the underpinnings of the Plano Real. Although inflation fell quickly and sharply after the introduction of the new currency, the real, in mid-1994, and even though the plan initially allowed the exchange rate to float until it found a competitive level, the rate appreciated slightly in nominal terms and substantially in real effective terms (Figure 12.4). That damaged Brazil’s international competitiveness, with effects that would gradually become more severe. In addition, the loss of “inflation tax” revenues as prices stabilized severely affected both government finances and the banking system.

By 1996, Brazil was facing a potentially calamitous banking sector crisis, but the central bank managed to contain it through a combination of takeovers, bailouts, and privatizations.43 The government also took measures in this period to strengthen the independence of the central bank and increase the transparency of its policy decisions. Fiscal policy proved to be more difficult to control, partly because of the familiar political pressures but also because the revenue losses from the end of high inflation coincided with the costs of rescuing the banking system.

Figure 12.4.Brazil: Inflation and Exchange Rates, 1989–99

Sources: International Financial Statistics and Information Notice System.

a Data prior to July 1994 are the equivalent in reais of then-prevailing currency units.

The Asian financial crisis brought these simmering stresses to a boil. Since March 1995, the central bank had been managing the exchange rate within an adjustable band vis-à-vis the U.S. dollar. During the November 1996 Article IV consultation discussions in Brasilia, the staff and the authorities agreed the currency was overvalued and the band should be adjusted gradually over a two-year period to correct the prob-lem.44 A year later, when speculative pressures on the exchange rate picked up in response to the Asian crisis, the government announced a sizeable package of fiscal measures. Although the staff and management of the Fund were privately urging the authorities to shift the exchange rate more aggressively as well, Camdessus publicly praised the fiscal measures, which calmed financial markets, allowing Brazil to ride out the crisis without a change in exchange rate policy.45 Implementation of fiscal policy flagged in 1998, however, and the underlying problems—a large fiscal deficit and an overvalued currency—remained.

The Russian default in August 1998 dealt an even bigger blow. Among emerging markets, Brazil was particularly vulnerable. It was viewed as having a debt structure and a potential exchange rate problem similar to Russia’s, and it was in the heat of a presidential election campaign. Although the central bank had ample foreign exchange reserves, preserving the adjustable-band exchange arrangement had the potential to become expensive if foreign creditors stopped rolling over their loans or domestic residents shifted liquid assets into dollars in large quantities. In that event, would Brazil be forced into the same default strategy employed by the Russians? International financial conditions were more hostile generally in 1998 than they had been in 1995 or 1997, and skeptical international investors were focusing their sights increasingly on Brazil. Many analysts thought Brazil’s circumstances were so dire, and policies so weak, that international financial support would be futile. The MIT economist Rudiger Dornbusch famously advised the IMF that “when they call 1-800-BAILOUT, just let it ring. Say our operators are busy.”46

Among the first to sound the alarm was David Folkerts-Landau, head of global emerging-markets research at the investment bank Deutsche Morgan Grenfell. The bank had taken large losses in the Russian default, and Folkerts-Landau—who had worked in the IMF’s Research Department from 1979 to 1997—was both highly critical of the IMF’s policy advice to Russia and apprehensive of a repeat performance in other weak economies. In a conference call with clients on August 26, Folkerts-Landau warned that Brazil was the country most vulnerable to contagion and that the risk of investing there was now much higher.47 Financial markets immediately panicked, apparently fearing Brazil would soon be forced to ask the IMF for help and the IMF would advise them to default in much the way Russia had. Stocks on the São Paulo stock market fell sharply the next day in chaotic trading conditions.48 Although the declines were soon partially reversed, confidence was badly shaken.

When the stock market again fell sharply on September 10 as part of a broad sell-off across Latin American markets, the government decided it was time to act. The authorities’ immediate response was to raise short-term interest rates to nearly 50 percent to curb speculative pressures. The finance minister, Pedro Sampaio Malan, then called Fischer the next morning to discuss policy options and to gauge international support. Fischer stressed the importance of reducing the fiscal deficit. He then participated in a conference call among the finance deputies of the Group of Seven (G7) countries. That group reinforced the fiscal message, saying they would not issue a statement of support for Brazil until the government committed to a strong fiscal policy for the next three years.49

By this time, the Fund staff was convinced that the real was so overvalued that the policy of depreciating it gradually was no longer sufficient. Once confidence weakened, the question was whether the central bank could continue to defend the announced band without depleting its reserves. If not, could a devaluation be made to stick long enough to allow the government to tighten monetary and fiscal policies without getting hit by a financial crisis? With the presidential election just three weeks away, the staff advised Fischer to press Brazil to devalue as soon as the election was decided.50

President—and former Finance Minister—Fernando Henrique Cardoso decided not to wait to get fiscal policy under control. On September 22, he made a dramatic televised speech to the nation in which he stated unequivocally that “the state has not been able to live within its own means. … This cannot continue.”51 Brazil could not expect to sustain economic growth unless the government restored fiscal balance quickly, and Cardoso promised to do so. The next day, Camdessus issued an enthusiastic statement of support on behalf of the IMF,52 and officials of major creditor countries—notably U.S. Treasury Secretary Robert Rubin—soon followed suit.

Cardoso won reelection on Sunday, October 4, just as the IMF/World Bank Annual Meetings were getting under way. Malan and the central bank governor, Gustavo Henrique de Barroso Franco, were in Washington for the meetings, and they took advantage of the occasion to ask Fischer for advice over dinner on the Friday before the elections. Fischer repeated his warning about the importance of a fiscal retrenchment, but he also calculated that Brazil was going to need substantial financial assistance from official creditors until the private sector regained confidence and resumed lending. To anchor that support and hasten the country’s return to the financial market, he suggested that Brazil needed a Fund-supported program.53 Malan and Franco took that advice back to Cardoso, and within days the government was preparing to enter into discussions leading to a new stand-by arrangement with the IMF.

To defuse the domestic political ramifications of calling in the IMF for help, Malan declined to invite a staff mission to come to Brasilia. Instead, he sent a deputy, Pedro Parente, to Washington to begin negotiating with the IMF on October 17.54 The talks proceeded quickly and fruitfully, primarily because the two sides agreed completely on the need for a large and rapid change in fiscal policy. Two other issues were more divisive: exchange rate policy and the strategy for regaining access to financial markets.

On exchange rate policy, the Fund staff pressed their view that Brazil should increase the rate of crawl to get real effective depreciation of at least 10 percent a year. The authorities resisted and argued that any change in policy would have a devastatingly destabilizing effect on expectations. Quickly but reluctantly, the Fund relented on this issue.

On private sector involvement, the Fund was more insistent, for three reasons. First, lending official money to Brazil while private creditors pulled their money out would do little to stabilize the economy or lay the groundwork for future growth. Somehow, the participation of financial markets had to be secured, and the only real issue was the best way to achieve that objective. Second, the memory of the initial failure of the Korean program in December 1997 was still fresh in the minds of the IMF management team. Only when major central banks began encouraging their countries’ commercial banks to maintain loan exposure to Korea was the crisis resolved (see Chapter 11). Third, in the case of Brazil, European central banks were insisting that some form of officially organized involvement of private creditors was the sine qua non for their participation in the rescue.

Figure 12.5.Brazil: Use of Fund Credit, 1989–99

(In millions of SDRs, monthly data)

Source: International Financial Statistics.

Note: SBA = Stand-by arrangement; SRF = Supplemental Reserve Facility.

The Brazilian negotiators resisted that approach out of concern that they would come to be regarded as in the same boat as other emerging-market countries suffering from financial crises. In their view, Brazil had a much stronger economy and a much more solid financial footing than Korea, Mexico, or Russia, and they were determined to find a market-friendly solution to their current problems. Faced, however, with continuing capital outflows and the need for an agreement with the IMF, they eventually agreed to try an intermediate approach in which interbank credits would be monitored, and the Fund would coordinate an effort to encourage foreign bank creditors to maintain their exposures. To soften the risk of a negative market reaction, they suggested that the stand-by arrangement with the Fund should be purely precautionary. The Fund would put a large sum of money at Brazil’s disposal, but the government would announce its intention not to draw on it.

Ever since the Mexican rescue of January 1995, the usual strategy in devising official financing packages for emerging-market countries facing capital market crises had been to assemble a large pile of money to convince speculators that the country could defend its exchange rate. What was being defended, however, differed greatly from one case to the next. In Mexico and Thailand, financial support was announced after a sizeable devaluation had already occurred, and the object was to restore confidence. In Korea, the authorities had begun allowing more flexibility in the exchange rate, and part of the objective was to limit the extent of the fall. In Russia, the Fund had made a large financial commitment without a devaluation, with the hope of preventing a collapse of the banking system and gaining time for the government to shore up its finances and stabilize exchange markets.

In Brazil, the objective was similar to the Russian case, but with three major differences that added greatly to the credibility of the program. First, Brazil had not borrowed from the IMF since 1992, and it had repaid all of its outstanding loans (Figure 12.5). Second, the authorities were gradually depreciating the exchange rate along a controlled path. The need for a real effective depreciation was not in dispute. The only question was whether to get there gradually or more quickly. Third, Cardoso had just been reelected on a platform that included a major and sustained tightening of fiscal policy, and his party had a solid majority in congress. All he needed, it seemed, was time to carry out his program.

Negotiations on emergency financial support for Brazil took place on multiple fronts throughout the first two weeks of November 1998. The largest single share would come from the IMF, in the form of a three-year stand-by arrangement totaling more than $18 billion (SDR 13 billion, or six times Brazil’s quota). Although this was the second largest financial commitment in IMF history (after the 1997 Korea arrangement), agreement on the accompanying economic program was reached readily in a series of meetings at Fund headquarters. On the other side of the Atlantic, a meeting of central bankers at the Bank for International Settlements (BIS) produced pledges from industrial countries totaling $14.5 billion for a credit facility for Brazil. A separate bilateral loan from Japan and loan commitments from multilateral development banks brought the total to $42 billion.

The multilateral financial package was announced with great fanfare on November 13, at which time the IMF’s portion was just being submitted to the Executive Board for consideration. Formal approval of the stand-by arrangement did not come until December 2. At the Board meeting, the dominant issue was the wisdom of lending to Brazil without demanding a devaluation. The head of the staff team on Brazil, Teresa Ter-Minassian (Deputy Director, Western Hemisphere Department, or WHD), defended the program on practical grounds. Experience in Asia had shown that changing exchange rate policy in the middle of a financial crisis was likely to lead to overshooting and unpredictability. For four years, Brazil had been using exchange rate stability as an anchor for expectations, and that strategy had succeeded in stabilizing prices in a country that had previously suffered through hyperinflation. In the staff’s view, the fundamental cause of the external deficit and the loss of confidence by the markets was the government’s inability to get the fiscal deficit under control. Eliminating that problem was the goal. If the authorities failed to solve the fiscal imbalance, they would have to devalue later. But not now, not in a crisis.55

The stand-by arrangement, which was approved unanimously, had three special features. First, like the Korean and Russian arrangements made in the preceding 12 months, a major portion (70 percent in this case) was being made available under the terms of the Supplemental Reserve Facility (SRF). Using the SRF meant that disbursements could be heavily front-loaded, but also that the country would be expected to repay them relatively quickly as economic conditions improved. Moreover, the interest rate the Fund charged on outstanding SRF balances was much higher than the standard rate of charge. Second, the IMF was borrowing the funds for the SRF portion of the arrangement by activating the New Arrangements to Borrow (NAB) for the first time. Third, as noted above, the authorities stated they viewed it as precautionary. The whole $42 billion package was a crucially important signal of confidence from the international community, but the government was convinced it could continue to manage the exchange rate primarily with its own reserves.56

Serious problems arose almost immediately, not because of any inherent design flaw in the program or because of a failure by the government to implement it, but because of an unexpected flare-up of domestic political opposition. Despite Cardoso’s personal popularity and his party’s majority, in December the Brazilian Senate rejected key elements of his proposal to strengthen the budget. That action destroyed the authorities’ plan to treat the stand-by arrangement as precautionary. On December 15, two weeks after the arrangement was approved, Brazil made an initial drawing of some $4.8 billion (SDR 3,419 million) to rebuild its foreign exchange reserves.57 On January 6, 1999, Itamar Franco—Cardoso’s predecessor as president and at that time the governor of the state of Minas Gerais—declared that his state was defaulting on more than $15 billion of debts to the federal government. Although this declaration was largely political bluster (because the federal government had access to financial assets it could seize to collect the debts), the effect on investor confidence was immediate and devastating. From that moment, the IMF-supported program was effectively dead.

The next 10 days were chaotic and disastrous. As Ter-Minassian had foretold, Brazil now had no choice but to act aggressively to depreciate the currency to a level that would restore confidence. The central bank governor, Gustavo Franco, adamantly opposed any change in foreign exchange policy, fearing such action would destroy the Plano Real that had underpinned the country’s economic success for the past four years. Cardoso decided he had to replace Franco to restore viability to his program, and he eventually found a plan that looked to be worth trying.

On January 12, Malan called Fischer to inform him that Cardoso was replacing Franco with Francisco Lopes, the director of monetary policy at the central bank. Lopes had developed his own plan for preserving the exchange rate’s stability by restructuring the crawling band regime into a “diagonal endogenous exchange rate band,” and he had personally convinced the president to give it a try. Fischer was appalled because he saw clearly that the scheme had virtually no chance of success. In a series of telephone calls throughout that night, he tried to convince Lopes to abandon the idea, close the exchange market for a day or two to give markets time to calm down, and then allow the exchange rate to float. That effort at persuasion failed, and the next morning the central bank announced the new band.58

Fischer was right. Speculators immediately attacked the real, and the central bank was unable to defend the “endogenous band.” At a meeting early Friday morning, January 15, Malan insisted that Lopes abandon the two-day-old regime and allow the exchange rate to find its own market level. The rate depreciated by about 10 percent by the end of the day, but the worst had ended. That night, knowing the weekend would give them a brief reprieve, Malan and Lopes flew to Washington for an intense (and tense) two days of consultations.

IMF officials were decidedly unhappy, both with the change in regime and the way it had occurred suddenly and without prior consultation (as was required under the terms of the stand-by arrangement).59 Camdessus felt compelled to issue a public statement of support, but he made sure it was as neutral and ambiguous as it could be. The authorities, he wrote, have “informed the IMF of the modifications of the exchange rate system adopted today” and have “reaffirmed to the IMF their strong determination to put in place … the full fiscal adjustment program announced in November 1998.” He concluded simply, “I welcome these assurances.”60 The weekend was not going to be much fun.

At that moment, IMF officials had no clearer an idea of the way to move forward than did the Brazilians. When a large contingent of staff met with Malan and Lopes in the Managing Director’s office on Saturday, January 16, the first suggestion from the IMF came from Camdessus. Why not try a currency board, he asked. It was working for Argentina, and the conditions seemed right for it to work in Brazil. Following the depreciation of the previous day, the currency was no longer obviously overvalued, and—in contrast to Indonesia, where Camdessus had forcefully rejected a proposed currency board the year before—the central bank had adequate reserves to back up the monetary base.61 Neither Malan nor Lopes had any interest in trying it. In their view, the structure of the Brazilian financial system—including a low rate of dollarization and large quantities of very liquid securities outstanding, which served as close substitutes for money—would render a currency board ineffective.

Fischer and many on the staff thought a floating exchange rate would have a better chance of success. The danger was that the value of the real could plummet further, plunging the economy into chaos (as had happened in Indonesia). The fiscal adjustment Brazil was already making would reduce that risk, and a sound, complementary strategy for monetary policy could anchor expectations firmly. The crucial point from the IMF side of the table, though, was not the specific proposal. The key was to move decisively to a “corner solution”: either a resolutely fixed exchange rate or a clear commitment not to intervene at all. Any policy that was less clear, they believed, would quickly be undermined by speculation.62

For the rest of January, Brazil muddled through the crisis as best it could, while continuing to allow the exchange rate to find its own level. The real depreciated further against the dollar, and the central bank responded by raising interest rates, though not by enough to satisfy the IMF. At the end of the month, Ter-Minassian returned to Brasilia to renegotiate the program. She happened to arrive on the same day that Cardoso decided to replace Lopes with a new central bank chief, Arminio Fraga.

Although Fraga had spent the past six years in New York running a hedge fund for George Soros, he was a former director of the Central Bank of Brazil and was well known and respected both in the country and in Washington. He was also a keen advocate of a floating exchange rate, which he proposed to anchor by targeting the inflation rate. Immediately after his appointment, he consulted with Fischer and with Lawrence Summers (deputy secretary of the U.S. Treasury) on how to set up an inflation-targeting regime for monetary policy. He also sought technical assistance from the Fund’s Monetary and Exchange Affairs Department. Using such a regime to bring Brazilian inflation down steadily to internationally comparable levels would not be easy, but it was a strategy he believed would work and the IMF was willing to support fully.63

To solidify the Fund’s support for the revised program, Fischer made a 6,000 mile detour to Brasilia on his way back to Washington from the World Economic Forum in Davos, Switzerland. After meeting with Cardoso, Malan, and Fraga, he signed off on a press release announcing agreement on all the key issues. On this occasion, the Fund was unreserved in its endorsement of Brazil’s program:

The authorities and the IMF team are confident that the decisive implementation of appropriate economic and structural policies, with the support of the international financial community, will be instrumental in promoting in the course of 1999 a progressive rebuilding of confidence, a substantial improvement of the current account of the balance of payments, a gradual reflow of private capital to Brazil, and a strengthening of the exchange rate.64

With the Fund’s approval secured, the last critical element in Brazil’s recovery from the crisis was to stabilize private capital flows and rebuild investor confidence. Folkerts-Landau and his colleagues at Deutsche Bank were still extremely skeptical that the government could carry out its intended policies, and many other bankers were similarly skittish. A temporary intensification and broadening of the effort to encourage creditors to maintain their loan exposure was needed. Although reluctant to do anything that might look like arm-twisting, the authorities agreed to do a series of “road shows” around the world to present the program to bank creditors and ask them to roll over loans as they came due. IMF staff participated in each of those meetings and helped explain the program and the economic outlook. The major creditor countries also helped by hosting the meetings, usually on the premises of their national central banks. As with the similar operation carried out for Korea the year before, Fund staff also coordinated and monitored the rollover process. Each creditor bank thus could be assured that others were maintaining their exposures, and no one had to worry about losing a race for the exits.65

For the next two years, the program succeeded remarkably well.66 Despite the depreciation of the currency, the inflation-targeting regime ushered in a steady reduction in price inflation, and economic growth remained positive. At the end of March 1999, the Executive Board reviewed the program favorably and approved disbursal of a second tranche of the stand-by arrangement, in the form of an SRF loan of nearly $5 billion (SDR 3,636 million). By August, international investors were once again eager to lend to Brazil, a turnaround that inspired the British magazine Euromoney to give its coveted “central banker of the year” award to Brazil for the second straight year (Franco in 1998 and Fraga in 1999). In December, Brazil began repaying the SRF loans, and the crisis was past.67


Before, during, and after the East Asian crisis, Argentina was the darling of international investors. The skepticism that had prevailed in the early 1990s—could Argentina really overcome the failures of the 1980s and achieve sustainable, noninflationary growth?—dissipated after the economy successfully weathered the tequila crisis. The willingness and the political ability of the government to raise taxes in the middle of President Carlos Menem’s campaign for reelection in 1995 greatly impressed the IMF and led to a resumption of lending that was still continuing when the Asian crisis hit two years later. Private capital markets responded even more enthusiastically, and the resulting inflow of capital enabled Argentina to end—temporarily—its reliance on official financing.

The central financial issue for Argentina at this time was the development of exchange rate policy. The rate between the peso and the U.S. dollar had been fixed at parity since the enactment of the Convertibility Law in 1991. This stability and the commitment underpinning it functioned as the very foundation of Argentina’s economic recovery, but almost everyone understood that sooner or later the rate would come under market pressure and the government’s commitment would be severely tested. Before much longer, the government would have to find a way either to introduce more flexibility into its management of the rate or to make its commitment even more irrevocable.

Although the IMF never tried seriously to induce Argentina to abandon the de facto currency board, it understood the problem. When the Executive Board met in April 1995 to review performance under the EFF arrangement, Stefan Schoenberg (Germany) characterized the issue as the “Hotel California dilemma”: “You can check out any time you like, but you can never leave.”68 Once confidence begins to wane, either adhering to the regime or abandoning it is likely to damage confidence even further. At the same meeting, both Karin Lissakers (United States) and Camdessus expressed “skepticism about the long-term viability of a currency board,” because of the absence of a lender of last resort in such a scheme. Marc-Antoine Autheman (France) identified the key issue, which would bring down the Argentine regime in 2001: the program was “very risky” because it was “unusually strong” and thus “very difficult to sustain.” In sum, “the risk of failure … is related to a central issue, the political sustain-ability of the very tight monetary and fiscal policy implied by this program.”69 Nonetheless, the Board approved a fourth year and an augmentation of the arrangement.

The convertibility plan was the brainchild of Domingo Cavallo, whom Menem had installed as economy minister in 1991. Five years later, political tension between Menem and Cavallo led to Cavallo’s resignation in July 1996. His departure left a gap that temporarily frightened investors, but the disturbance quickly passed. Roque Fernandez, his replacement, was a scholarly economist with a Ph.D. from the University of Chicago who had worked at the IMF in the 1970s and had served as president of the central bank while Cavallo was in charge of economic policy. As his principal deputy (secretary of finance), he brought in Pablo Guidotti, an IMF veteran from the 1980s who had served under Fernandez at the central bank. This good team, well known and respected in Washington and in financial markets, would have to try to find a viable exit strategy.

The economic case for floating the Argentine peso was fairly strong in 1997, but neither the authorities nor the IMF staff were pushing in that direction. The U.S. dollar was strengthening in foreign exchange markets, raising the value of the peso against other currencies and weakening Argentina’s international competitiveness. With private capital shifting to Argentina from other emerging markets, the biggest short-term risk was that the rate would appreciate further if left alone to find its own level. On economic grounds, the likelihood of a large movement in either direction seemed small. The benign financial conditions of 1997 offered a golden opportunity to achieve a long-term goal of establishing stable and independent monetary and fiscal policies with a market-determined exchange rate.

The real risk was political. The fixed rate, enshrined in law, was extremely popular across the political spectrum in Argentina. Memories of the disastrous effects of previous governments’ inability to control budgets or inflation were painfully acute. After just six years of stability, six years that had raised the international esteem of Argentina beyond the dreams of those who had lived through the 1980s, would a shift in policy toward flexibility be credible? Would it be undercut by renewed budgetary battles? Would international investors flee Argentina, as they were already fleeing Indonesia and Korea? No one in power in 1997 was prepared to take that test. No one in the IMF was willing to risk destabilizing the country by publicly challenging the convertibility regime.

Throughout Menem’s last three years as president (economically, the post-Cavallo years, 1996–99), the goal for exchange rate policy was to move toward discarding the peso altogether and making the U.S. dollar the official currency of Argentina. Until 1999, that goal was pursued quietly because the public face of policy was preservation of the convertibility plan at the fixed exchange rate. Increasingly, however, Argentine residents were denominating contracts, including long-term home mortgages, in U.S. dollars. Because the dollar became the dominant store of value while the peso remained the dominant means of payment, the potential cost of ever devaluing the peso or allowing it to float was becoming unbearable.

In January 1999, Menem publicly announced that he wanted ultimately to dollarize the economy. To follow up, the finance authorities entered into negotiations with their counterparts at the U.S. Treasury and Federal Reserve Board to develop a cooperative agreement on dollarization that they hoped would lead ultimately to a “monetary association treaty.”70

These negotiations concentrated on the central issues of seigniorage and oversight of the banking system. At the time, the Argentine government was receiving about $750 million a year from the U.S. Treasury in interest on the securities the Treasury held as backing for the peso. Dollarization would involve converting those securities into cash, which would give rise to a seigniorage windfall for the United States at Argentina’s expense. Some means would have to be found to share that windfall more equitably. The deeper problem was that the central bank would surrender its ability to act as a lender of last resort in the event of a banking crisis. U.S. officials made it quite clear, first privately and then publicly, that they would take no heed of any dollarized country’s monetary policy needs, nor any responsibility for the soundness of banks in dollarized economies. That sobering conclusion did not overly discourage the Argentine authorities, who continued discussing the possibility of dollarization until they left office after the presidential elections of October 1999.71

Most IMF staff throughout this period accepted this very strong commitment to the fixed exchange rate as a fact of life, and in any case did not generally disagree with it.

An IMF staff mission, led by Ter-Minassian (who was also covering Brazil), went to Buenos Aires in May 1997 to review the 1996–98 stand-by arrangement. She found the trade deficit worsening, owing to a surge in imports. The authorities were not particularly worried; they were counting on productivity gains and low domestic inflation to preserve international competitiveness and eventually to moderate the imbalance in trade. The mission report warned them to be “vigilant,” not in thinking about exchange rate flexibility, but rather of the need to tighten fiscal policy if the current account were to weaken substantially.72 A few weeks later, Camdessus met with Menem, other officials, and leaders of civil groups in Buenos Aires. While sympathizing with concerns of nongovernmental groups about rising unemployment and poverty, he “expressed confidence that high unemployment could be dealt with effectively through maintaining the present course of economic policy and through deepening the [structural] reforms.”73

The staff issued a favorable review of program implementation, and the Executive Board approved the release of two more tranches of the stand-by arrangement in June and September 1997. Those two drawings turned out to be the last by Argentina in this decade, first because of fiscal slippages and later because the government no longer needed official financing.74

The outbreak of the Asian financial crisis in the second half of 1997 coincided with weakening political support for the government. Menem’s party, the Justicialist Party, lost its parliamentary majority in the October elections, after which the government’s ability to control fiscal policy weakened. A sell-off of equities in the Hong Kong SAR market later that month caused a brief bout of contagion to Argentina, but private capital inflows soon resumed. At that point, the authorities decided to secure the IMF’s seal of approval as a precautionary bulwark against a potential reversal of financial fortune.

Argentina’s request for a precautionary EFF arrangement put the IMF in a quandary. Excessive spending by the central government had violated the terms of the stand-by arrangement that was about to expire. The authorities had not requested a waiver and had decided to forgo the last scheduled drawing. IMF staff and management urged the authorities to tighten fiscal policy, but their leverage was poor as long as Argentina was asking only for a vote of confidence, not a loan. The Fund could have denied the precautionary arrangement, but at the outset of 1998 Argentina’s policies were not so bad as to warrant such a drastic step. Looking forward, the staff concluded that “the authorities’ program is consistent with maintaining financial stability, while promoting growth and advancing the structural reforms needed to provide lasting support to Argentina’s monetary and exchange arrangements.”75 The Executive Board concurred and approved a three-year arrangement for $2.8 billion (SDR 2.08 billion, or 135 percent of quota) in February 1998.

By mid-year, when Ter-Minassian returned to Buenos Aires to review the program, the economy was weakening. The assumption of 5 percent growth on which the program was based was now seen clearly to have been overly optimistic. Lower growth rates meant tax revenues would fall short, and the fiscal deficit would exceed the program target. After the Russian default in August and the collapse of the Long-Term Capital Management hedge fund in September, the problem worsened as the inflow of private capital began to falter. The main stock market index for Argentina fell by about 40 percent.

The IMF was at a crossroads. It could insist, as a condition for continuing its seal of approval, on a tightening of fiscal policy or a deepening of structural reforms to correct the financial problem. A fiscal tightening, though, would worsen the economic downturn and could lead to panic. Alternatively, the Fund could treat the situation as a temporary setback for which automatic fiscal stabilizers were the correct response. Because the policy stance was basically sound, this second option was preferred at that time.

The Executive Board approved the program review on September 23, 1998. To reinforce the message that the Fund (and the United States) strongly supported Argentina’s policy stance, Menem made a well-publicized trip to Washington two weeks later. In a highly unusual maneuver, he was invited to address the opening plenary session of the IMF/World Bank Annual Meetings, immediately following the traditional address by the head of state of the host country (U.S. President Bill Clinton). Defending that decision, Camdessus praised Menem’s economic policies and told the press that “Argentina has a story to tell the world.”76 Many on the staff were less impressed by the authorities’ ability to rein in fiscal pressures as political support for them waned, but the Fund’s public stance was united in support.77

Throughout 1999 and beyond, the IMF continued to support Argentina by approving the EFF-supported economic program. The $2.8 billion loan commitment remained in force, while inflows of private capital enabled the authorities to continue to forgo drawing on it. As the presidential campaign heated up, Fischer met with senior Argentine officials at a conference in Chile, after which he privately told the Fund’s Executive Directors that “we could be in for a fairly difficult time in the months to come.”78 However, the program worked reasonably well for another two years, before pressures finally built up and exploded in the massive crisis of 2001–02.79


Ecuador faced extraordinarily difficult economic circumstances in the late 1990s, brought on by natural disasters (drought followed by floods, then the decimation of the fishing harvest caused by “El Niño” tides), border warfare with Peru, extremely low prices for oil exports, and political instability (four presidents in quick succession). Macroeconomic policy implementation was lax, owing in large measure to the inability of the government to secure parliamentary support. Weaknesses in the banking system erupted into crisis in 1998 after the Asian meltdown aggravated weakening capital inflows. A major financial and economic crisis ensued a year later.80

As long as the border conflict and the domestic political quarrels occupied the government’s attention, the IMF could do little to help. The staff monitored policy implementation in 1995–96, but the government’s performance (in the staff’s cautious terms) “fell well short of expectations.”81 After the short-lived presidency of Abdalá Bucaram ended in his impeachment, the staff entered into program negotiations with the interim president, Fabian Alarcón, and then (from August 1998) with his newly elected successor, Jamil Mahuad. Mahuad’s election, closely followed by the settlement of the border dispute with Peru in October, created an opportunity to stabilize the economy and regain the support of the IMF, but circumstances remained difficult.

In 1999, continued inaction by the authorities in the face of unrelenting negative circumstances turned the domestic banking collapse into a currency crisis that threatened the country’s whole economy. As more and more of its dwindling export revenues were siphoned off to service external debts, defaulting on that debt became a serious option, both for the authorities and for the IMF staff. In April, as a staff mission prepared to go to Quito to negotiate terms for a stand-by arrangement, officials in the Policy Development and Review Department (PDR) who were overseeing the Fund’s policy advice to Ecuador felt the need to issue a warning to the area department staff. “Ecuador should pursue a cooperative approach with its creditors and do all that it can to avoid the need for unilateral action,” they wrote, and urged that the IMF should not take a position on Ecuador’s evident desire for a rescheduling of its external debt.82 That became the institution’s official position, but the tension between preserving market-based relations with creditors and keeping the economy running was only going to get worse.

The May 1999 mission, led by John Thornton (Deputy Division Chief, WHD) and reinforced at the outset by a brief visit by Fischer and Loser, concluded that the government needed to take further action to get its own finances under control, but also that weak domestic political support hampered its ability to do so. In addition to that adjustment effort, Thornton urged the authorities to try to reach three types of agreements with creditors. To regularize relations with external creditors and avoid default, they needed a rescheduling agreement with official creditors through the Paris Club; a similar agreement with private bank creditors through the London Club; and a new syndicated bank loan to cover the interest falling due on the country’s foreign-currency bonds, which were predominantly in the form of nearly $6 billion in Brady bonds. Because many of those bonds were held by large foreign banks, the best prospect for a new loan would be an agreement with a consortium of those creditors. Once those agreements were in place, the Fund could provide additional financing through a stand-by arrangement.83

Fischer repeated that message in a meeting with Mahuad. At this time, Fund management was trying to put in place a plan for “private sector involvement” in debt workouts as advocated by the G7.84 Paris Club creditors were signaling that they would not reschedule debts without assurance that private creditors would maintain their exposure. Private creditor participation thus was essential if a heavily indebted country hoped to finance its balance of payments. In Ecuador, Fischer explained, the key point was that “the private sector [collectively, the holders of Ecuador’s Brady bonds] would have to contribute by maintaining its exposure to the country. We [the IMF] did not mind how they did it, but it had to be done, and it was up to the country to start the negotiations and we would be willing to assist.”85

Figure 12.6.Ecuador: Use of Fund Credit, 1989–2000

(In millions of SDRs, monthly data)

Source: International Financial Statistics.

Note: SBA = Stand-by arrangement.

For each of these targeted agreements, creditors would normally expect the government to reach a prior agreement with the IMF on a stand-by arrangement. That agreement, in turn, would require the Fund to have a solid assurance that Ecuador could finance its external payments. This circle could be squared if all of the main parties could reach tentative agreements, conditional on the others, so that the IMF could coordinate the complex package and bring all the negotiations to a conclusion. Any slippage could be fatal.

Although Ecuador had not borrowed from the IMF since 1994, the Fund was still a creditor to Ecuador as well as a policy advisor and an assessor of its policies and economic prospects. Ecuador’s staying current on those IMF loans was understood to be a prerequisite for an orderly settlement of its other loans. In 1998 and 1999, Ecuador made eight quarterly loan repayments to the IMF totaling about $135 million (SDR 98.8 million). That completed the timely repayment of all of Ecuador’s outstanding obligations to the Fund (Figure 12.6).

Over the summer of 1999, Thornton monitored progress closely from Washington and returned to Quito every few weeks to consult further with the authorities. Gradually, it became clear that private creditors had little interest either in negotiating a reduction in Ecuador’s debt or in offering new loans to cover bond interest. Ecuador had a poor track record for economic management and thus little credibility in financial markets. The options for a market-based solution to the impasse were rapidly dwindling.

Matters came to a head late in the summer. On August 28, the government of Ecuador was scheduled to pay $96 million in interest on its Brady bonds, most of which were held by major international banks. It could not make that payment on time without taking draconian measures that would be politically costly and could be economically ruinous. The authorities were pressing the IMF to approve the lending arrangement in time to enable this payment. Fund management, however, was reluctant to lend to Ecuador for the purpose of repaying private creditors.

On August 18, the minister of finance, Ana Lucia Armijos, traveled to Washington to meet with Camdessus and others at the IMF. In the course of the meeting, Camdessus informed her that he could not propose approval of the arrangement without adequate assurance that the country’s economic program would be fully financed. That meant, in the first instance, that Ecuador could not make the August 28 interest payment.86

When the IMF’s insistence on this point leaked out, the Fund appeared to be forcing Ecuador to default on its Brady bonds. No country, to this point, had ever defaulted on Brady Plan debt. For Ecuador to do so could raise risk premiums for all other countries with such debts. That was not the intended outcome. Ecuador’s bonds allowed for a 30-day grace period. If Ecuador deferred its interest payment, those 30 days could be used to bring both sides to the table for more serious negotiations on rescheduling and reducing Ecuador’s debt burden. Realistic or not, that was the intention and the hope.

This tactic failed. President Mahuad announced the deferment on August 25 and indicated that the IMF had already approved it (Cisternas, 1999). That angered and raised the suspicions of creditors, while it simultaneously emboldened the authorities. The Fund moved quickly to contain the damage, by announcing a successful conclusion to program negotiations on August 31.87 That also backfired. Although the press release cautioned that approval of a stand-by arrangement still depended on completion of a number of policy actions and “the receipt of adequate financing assurances,” the message was widely interpreted as an implicit approval of default.

IMF officials were hoping for one of two favorable outcomes. Either the combination of Ecuador’s unilateral deferral and the Fund’s announcement of its approval of the policy program would lead quickly to a negotiated settlement with creditors, or the authorities would find a way to make the interest payment in September and then enter into negotiations for a more comprehensive debt re-structuring.88 With encouragement from external advisors, however, Mahuad was leaning increasingly toward defaulting as a way to force creditors to be more flexible in negotiations. On September 23, 1999, a reporter asked Camdessus if the Fund was encouraging Ecuador to default on its Brady debt. Camdessus denied it and replied that the Fund was encouraging the authorities to use the 30-day grace period to “narrow the gap” with its creditors.89 Although the staff still believed Ecuador would pay in time to avoid default, the prospects for narrowing the gap were already dim.

Just four days later, as the grace period approached expiration, Camdessus had to issue another news brief, stating that he “regrets that to date Ecuador appears not to have found it possible to enter into negotiations with its creditors, with a view to reaching a comprehensive resolution of these difficulties in a collaborative manner.” His recommendation for approval of the stand-by arrangement was still contingent upon Ecuador being “judged to be making good faith efforts to reach a collaborative agreement with its creditors.”90 In the view of both private creditors and the IMF, those good-faith efforts were not yet evident.

As of September 30, Ecuador was formally in default.91 The IMF nonetheless accepted the authorities’ Letter of Intent setting out the policies it intended to follow to get the economy back on course. The authorities then needed to engage in meaningful and cooperative negotiations with creditors and to begin implementing the fiscal, monetary, and other policies in the program. Negotiations did get under way soon afterward, but the government failed to carry out the policy program. By end-October, the Fund abandoned the plan to approve the stand-by arrangement, and negotiations with creditors suffered a setback. The impasse continued through the rest of 1999, and economic conditions continued to worsen, contributing to the fall of yet another Ecuadoran government but ultimately forcing a solution to the crisis.

In a last-ditch but ultimately successful effort to generate an economic recovery, Mahuad decided to take the bold step of replacing the national currency, the sucre, with the U.S. dollar. Although he was driven from office early in 2000 in a coup d’etat, the new government continued with the dollarization scheme. The staff renegotiated the program accordingly, and in April 2000 the IMF finally approved a stand-by arrangement for just over $300 million (SDR 226.7 million, or 75 percent of quota). The arrangement enabled Ecuador to reach a debt-exchange agreement with creditors in July, and the crisis was brought to an end.

Lessons Learned

The Asian crisis was a transformational event for the IMF. The intensity and severity of the economic, political, and social consequences were so great; the consequences were felt so widely around the world; and the subsequent criticism of the Fund for allegedly failing to foresee or forestall it and for allegedly mismanaging it was so vitriolic that staff, management, and the affected countries all had to examine every aspect of it in search of lessons to apply in the future.

Reactions at the IMF

We have learned many lessons from the Asian experience, and there are many yet to learn. … we must demonstrate to the world that we are responding effectively.

Michel Camdessus92

Managing Director of the IMF

April 1998

A staff study conducted shortly after the Asian crisis (Lane and others, 1999) found that the primary failing was the limited ability of Fund-supported programs to restore confidence among private investors following a major collapse of capital inflows. Several more-detailed studies examined the reasons for this failing in specific cases and suggested a wide variety of explanations. Possible culprits included the inapplicability of standard macroeconomic models to capital account crises; the need for a more country-specific focus in program design; the absence of a true lender of last resort in international finance; recurring policy failings in borrowing countries, including rigid and overvalued exchange rates and weak oversight of financial systems; and excessive speculative behavior in international capital markets, perhaps buoyed by the moral hazard resulting from the possibility of an official rescue.93 Some broad lessons were drawn fairly quickly, but a more complete reassessment was still ongoing a decade later.94

A common thread running through all of the financial crises of the 1990s was an increasing reliance on short-term foreign-currency borrowing in the months before the outbreak. The details were country-specific—tesobonos in Mexico, GKOs in Russia, corporate debt in Korea, finance-company debt in Thailand, bank debts in Indonesia—but the results were always the same. Because the country’s external obligations were denominated in foreign currency, any sizeable depreciation of the home currency would risk bankrupting a substantial part of the economy. Because the obligations were highly liquid, any shift in investor sentiment would also risk driving up interest rates quickly and undermining the foundations of the country’s economic growth and stability. Many emerging-market countries had little choice but to fall prey to these risks, because the cost of borrowing at longer maturities or in domestic currency was simply too high to absorb. Only short-term credit—financed in part through “carry trade” (see Chapter 4, pp. 136–37)—was readily available on affordable terms.

The analytical tools available to the IMF in the mid-1990s took inadequate account of these balance sheet vulnerabilities. Standard macroeconomic models indicated that a temporary resort to “austerity”—a tightening of fiscal and monetary policies—would close a financing gap caused either by an initial fiscal or monetary excess or by a shortage of net private saving. Once balance was restored, confidence would return and a recovery could begin. Those models did not show that a country—especially an emerging market—could maintain an equilibrium by importing capital from abroad only so long as its “debt dynamics” were perceived to be sustainable. The sustainable level of net capital inflows was not an endogenous variable determined by the quality of the country’s macroeconomic policies and net national saving. It was an exogenous variable determined by the psychology of financial markets. Capital could disappear overnight, especially when it was a speculative flow financed by carry trade. Any economy dependent on short-term capital was vulnerable.95

This cartoon from the Economist in 1998 was a typical reaction to the Fund’s handling of the Asian crisis. (The theme was common in the 1990s; see Chapter 7, p. 293.) Copyright 1998 KALtoons

The emergence of a theoretical literature on the links between balance sheet vulnerabilities and financial crises (for example, Kaminsky and Reinhart, 1999; and Krugman, 1999) had a substantial impact on crisis analysis at the IMF. In September 1999, the staff acknowledged that “recent crises have demonstrated that few observers fully understood countries’ vulnerabilities to capital market shocks.”96 Under prodding by the Group of Ten (G10) industrial countries, the staff set out to intensify its analysis of countries’ liquidity management and risk assessments as part of the regular Article IV consultation process. Separately, the staff conducted extensive research into the theoretical and empirical dimensions of the interaction between balance sheet vulnerabilities and financial crises.97

Although these research initiatives deepened the profession’s and the staff’s understanding of the nature and origins of financial crises, the policy implications remained controversial. Was the key to reducing vulnerability an improvement in economic policymaking in the emerging markets, including better management of debt and liquidity, or was it a strengthening of oversight of capital markets and other elements of the international financial system? These ideas were not mutually exclusive; what mattered was emphasis and priority.

In the IMF, a first lesson drawn from the Asian crisis was that the structural and institutional prerequisites for successfully opening up to capital inflows were more stringent and daunting than previously thought. Before the crisis, Fund management and staff had always been careful to note that sound financial institutions and a stable monetary system were essential for a country to take advantage of the opportunities afforded by an open capital market. For the most part, however, they did not view that cautionary message as an impediment for the major emerging markets in Asia. Their financial systems had problems, but few people thought those problems serious enough to undermine the advantages of capital openness.98 By 1999, the balance of concerns had shifted, and the Fund’s advice on the importance of strengthening financial sectors was becoming more pointed.

A second lesson for the IMF was that crisis management requires careful, case-by-case attention to the behavior of private capital markets. The debt crisis in Latin America in the 1980s had been handled with the aid of several ad hoc tactics to ensure that private sector creditors—particularly international commercial banks—maintained their loan exposures until the indebted countries could regain normal access to capital markets. By the time the Mexican peso crisis hit at the end of 1994, those tactics were no longer applicable because bank creditors were less heavily exposed, bond financing had become a major supplement to bank loans, and a flourishing secondary market in emerging-market debt provided an additional safety valve for creditors wishing to reduce their exposures. Consequently, the Fund’s handling of the peso crisis raised serious concerns about moral hazard for creditors, given that the holders of Mexican tesobonos were mostly repaid in full despite the high risk inherent in holding those securities.

The management of the Asian crisis—and even more so, the Russian default that followed it—alleviated those moral-hazard concerns because most investors incurred large losses. In turn, those losses aggravated the difficulty of restoring creditor and investor confidence. Korea’s continuing reserve outflow throughout December 1997 and the almost total collapse of the Indonesian rupiah in the first few months of 1998 were shocking setbacks that forced the authorities, the IMF, and the major creditor countries to reconsider their passive stances regarding private capital flows to and from those markets. Improving domestic macroeconomic policies was necessary, but it would never be sufficient without some limitation on speculative international financial flows.

The remaining challenge was to find a more productive strategy. Was the solution for the country to impose new regulations on capital movements; for creditor-country monetary authorities to use regulations or persuasion to limit speculative investments and induce “private sector involvement” in any workout; for the international community more broadly to develop something akin to a sovereign bankruptcy procedure to force a collective response from creditors; or simply for the IMF and the authorities to redesign the policy reform program so as to try to “bail in” private creditors voluntarily? In the years that followed the Asian crisis, that question remained controversial and was never fully resolved. Plainly, though, the right answer would change over time and would depend on the specific conditions in each case.

A third lesson, even more difficult to apply, concerned the central role of political leadership in the successful management of a major financial crisis. All three of the main crises in East Asia occurred in countries in which government oversight of the economy had once been strong but had dissipated in the precrisis period. Once the economy collapsed, the policy regime could not be corrected without a change of government at the highest level. In Thailand, a new constitution was adopted in October 1997. A new government took office the following month, after which a real recovery began. In Korea, the December 1997 election of Kim Dae-Jung was the critical turning point for domestic support for a new beginning. In Indonesia, President Suharto’s resignation in May 1998 made room for a new team that could deal decisively with the corruption that had undermined the economy. Any effort to manage a crisis by correcting macroeconomic or structural imbalances was likely to run aground until the underlying political weaknesses could be overcome.99

A fourth lesson was that the international financial “architecture” was riddled with structural deficiencies. Regulation of market behavior was essentially a national undertaking; international cooperation on structural policies was piecemeal and sporadic; and compliance with international standards and codes was mostly voluntary. Recognition of these weaknesses led to a diversified and only partially successful effort to shore up the system by institutionalizing standards, monitoring compliance, broadening the major country caucuses to include emerging markets in new groups such as the Asia-Pacific Economic Cooperation (APEC) forum and the Group of 20, and attempting (unsuccessfully) to replace the Fund’s Interim Committee with a more formal Council.

Most directly relevant for the work of the IMF, the fifth lesson was that policy conditions on Fund lending had to focus on correcting the problems most critically important for resolving the issues that had generated the crisis. The temptation to try to fix other, perhaps insidious and entrenched, problems at the same time had to be resisted. That conviction led eventually to the adoption of new conditionality guidelines in 2002 that required policy conditions to be applied “parsimoniously” and only when they were “of critical importance” for specified purposes.100

Reactions Elsewhere

The Asian crisis … has left a scar on the minds of Asian policymakers. But the recovery … has … been robust.

Tharman Shanmugaratnam101

Minister for Education,

Second Minister for Finance,


September 17, 2006

In contrast to the responses by the IMF, in Asia the view took hold that the primary challenge was to strengthen oversight and regulation of the international financial system. Because macroeconomic policies had not been especially worse in this region than elsewhere, the IMF’s focus on austerity, even as a temporary response to crisis, did not seem appropriate. Within the Fund, M.R. Sivaraman (India) gave voice to this view during a March 1998 review of the crisis. The financial institutions that displayed rent-seeking behavior and that underpriced the risks of lending to Asian emerging markets were, in his view, “to be equally blamed for the crisis … as much as the lax policies and weak financial sectors in these countries.”102

One proposed alternative, exemplified by Malaysian Prime Minister Mahathir’s speeches, was simply to prohibit speculation in foreign exchange markets. More practically, limited controls on capital flows were accepted fairly generally as a possible element in a broad strategy to minimize the likelihood of a sudden pullout. The IMF’s general skepticism and sometimes hostility toward such controls created tensions between the agency and its Asian members.

The Asian Monetary Fund

A second arrow in the Asian quiver—in addition to controls—was to build up and pool ample foreign exchange reserves to enable participants to defend themselves against a speculative attack or a temporary payments deficit without having to borrow from the IMF and thus subject themselves to what they saw as inappropriately austere policy conditions. That desire led to a 1997 proposal known as the Asian Monetary Fund (AMF). Although the AMF was not adopted at that time, the principle resurfaced in various forms over the next few years.

The idea of a regional financing mechanism for Asia first arose in this context at a November 1996 meeting of the finance ministers and central bank governors of the Association of South-East Asian Nations (ASEAN) in Jakarta, Indonesia. The G10 had served as a peer group for reviewing industrial countries’ economic policies and conditions since the early 1960s. The Monetary Committee of the European Union and its predecessor committees had served a similar purpose since the 1950s. Why not set up a similar peer mechanism in Asia?

Camdessus was participating in the Jakarta meeting, and he offered a positive assessment at a private dinner he hosted for the ministers and governors.103 If they were to meet regularly to share information and establish standards for good policies, they could help each other avoid getting into a financial crisis. As director of the French treasury, Camdessus had chaired the European Monetary Committee in the early 1980s, and he was enthusiastic about the benefits of a peer review process. The goal, as he saw it, was for countries to act early and together in advance of a crisis.

Some of the Managing Director’s dinner guests interpreted his remarks a little differently. An integral feature of both the G10 and the European Union was a short-term financing mechanism in the form of swap lines or pooled reserves. In both systems, the availability of financing from within the group had helped take those countries out of the IMF sphere of influence by obviating the need to borrow. Peer review might help a little, but peer financing would do much more. Regardless of whether Camdessus intended the one to encompass the other, the extension seemed logical to at least some of the officials around the table.

ASEAN officials resumed discussion of the idea at their next meeting, in Phuket, Thailand, in March 1997. Again, Camdessus was invited to participate. Presciently—the crisis was still a few months in the future—the finance ministers engaged the Managing Director in a dialogue about the IMF’s potential response in the event of a “Mexican-style” financial crisis. Some among them worried that the Fund might be less engaged in their region than it had been in rescuing a large country next door to the United States. Camdessus assured them that the Fund’s assistance was universally available. He then went on to say, “Just as the Fund’s role in the case of Mexico had been decisive because of the support of other creditors, so the Fund’s role could be stronger in the event of an Asian crisis if Asian nations provided complementary financial support to each other.”104

To Camdessus, the critical word in that advice was “complementary,” meaning that intraregional financial assistance should help support a program that was primarily supported by the IMF and subject to policy conditionality. If, for example, Thailand were to request a stand-by arrangement from the IMF, the program would be more likely to succeed if Japan, Korea, and other developed countries could provide supplementary financing, much as the United States and other countries had done through joint participation in the workout of the Mexican crisis in 1995. The implicit primacy of the IMF in such a process no doubt came more naturally to the Managing Director than it did to the Asian officials in Phuket.

The urgency of a regional process intensified dramatically after the outbreak of the crisis in Thailand in July 1997. As described in Chapter 11, p. 509, the IMF and the Japanese finance ministry cohosted a “Friends of Thailand” meeting in Tokyo on August 11, at which eight central banks or national government agencies and three multilateral institutions assembled a support package totaling $16 billion. Although the IMF provided a quarter of that total, it was now abundantly clear to all that if any East Asian country needed financial assistance, they would have to band together to provide the bulk of the money jointly.

From that moment, Eisuke Sakakibara, Japan’s vice minister of finance, embarked on a quest to establish the AMF, not necessarily as a supplement to IMF money but as a possible substitute for it. That twist would mean that the regional fund would not just ensure adequate financial support but also that the IMF’s policy conditions might be avoided.105 The staff at the finance ministry hastily drafted a proposal, which Sakakibara then shopped to his counterparts around Asia, to the U.S. Treasury, and to the IMF during the rest of August and September. This effort culminated in meetings of Asian ministers and governors during the IMF/World Bank Annual Meetings in Hong Kong SAR in the third week of September. At each stop along the way, Sakakibara received a polite reception, encouraging him that the proposal would eventually carry the day. Initially, he believed—incorrectly—that both Camdessus and Lawrence Summers were on board. In fact, both were strongly opposed. Even among Asian officials, support was both shallow and thin. By the time the Hong Kong SAR meetings ended, the effort to establish the AMF was effectively dead.106

The Manila Framework Group

Except for a few potential borrowers, most national officials saw independence from the IMF as the poison pill in the AMF proposal. They were attracted to the idea by the goal of establishing an ongoing peer review group, with or without financing. U.S. officials warmed to that design, though only on the understanding that they would be invited to participate. After the failure of the AMF, most everyone thought strongly that the IMF should take the initiative to bring potential creditors and borrowers together to work out a specific alternative plan.

Shortly after Stanley Fischer returned home from the Annual Meetings in Hong Kong SAR, he wrote to Roberto F. de Ocampo, the secretary of finance of the Philippines, to propose establishment of a “regional surveillance group.” In this scheme, the IMF would provide a secretariat through its Tokyo office and would participate at a high level. Moreover, Fischer suggested, “it would be desirable” for this new group “to adopt a cooperative financing initiative” on which a participating country with an IMF-supported program could draw to supplement the Fund’s own resources when needed.107

De Ocampo responded by convening a meeting in Manila, to be attended by finance deputies from the East Asia and Pacific region plus representatives from North America, major European countries, the IMF, the Asian Development Bank, and the World Bank. By the time the meeting was held, on November 18–19, 1997, the deputies from Japan and the United States had already agreed bilaterally on a draft communique. After two days of talks, the delegates endorsed that draft with minor changes. The outcome was the establishment of the Manila Framework Group (MFG) with 14 member countries from both sides of the Pacific Ocean but no permanent secretariat. The communique endorsed the idea of a supplemental financing mechanism, though without any specific proposals for creating one.108

The MFG met semiannually for the next seven years to discuss regional financial issues. The need for regional financing did not arise during this period. Eventually, the group proved to be too unwieldy to have much relevance, and it disbanded in 2004.

The stillborn AMF and the ill-fated MFG were but two of several efforts by East Asian countries to establish an effective mechanism for regional economic and financial cooperation. ASEAN, the long-standing forum for finance officials of Southeast Asian countries, expanded its membership and included other Asian countries by meeting as “ASEAN Plus.” By the late 1990s, senior officials of 11 central banks were also meeting regularly as the Executives’ Meeting of East Asia-Pacific Central Banks. The APEC forum provided a summit framework at a broader regional level. After the end of the 1990s, these efforts intensified, notably with the establishment of regional swap lines through the Chiang Mai Initiative (Henning, 2002).

Despite all these initiatives, and despite many East Asian leaders’ continuing disaffection with the IMF as a crisis manager and general policy advisor, no viable alternative emerged to replace the IMF in these roles. Instead, vulnerable countries set out to accumulate foreign exchange reserves in unprecedented and costly amounts in an effort to avoid ever having to borrow from the Fund again. More than a decade after the Asian crisis ended, the legacy of fear and uncertainty it engendered was scarcely diminished.


    AthukoralaPrema-chandra2001Crisis and Recovery in Malaysia: The Role of Capital Controls (CheltenhamUnited Kingdom: Edward Elgar).

    • Search Google Scholar
    • Export Citation

    BalisacanArsenio M. and HalHill2003The Philippine Economy: Development Policies and Challenges (New York: Oxford University Press).

    • Search Google Scholar
    • Export Citation

    BlejerMario I.Alfredo M.LeonePauRabanal and GerdSchwartz2001“Inflation Targeting in the Context of IMF-Supported Adjustment Programs,”Staff Papers International Monetary Fund Vol. 49 No. 3 pp. 31338.

    • Search Google Scholar
    • Export Citation

    BlusteinPaul2001The Chastening: Inside the Crisis That Rocked the Global Financial System and Humbled the IMF (New York: PublicAffairs).

    • Search Google Scholar
    • Export Citation

    BoormanJack1999“Lessons from the Asian Crisis: The View from 1999,”revised text of remarks made on February 5 1998 at a seminar at the Capital Markets Research Center of Georgetown University. Accessed at

    • Search Google Scholar
    • Export Citation

    BoormanJackTimothy D.LaneMarianneSchulze-GhattasAlešBuliřAtish R.GhoshA. JavierHamannAlexMourmouras and StevenPhillips2000“Managing Financial Crises: The Experience in East Asia,”Carnegie-Rochester Conference Series on Public Policy Vol. 53 pp. 167.

    • Search Google Scholar
    • Export Citation

    CalvoGuillermo A.ed.2005Emerging Capital Markets in Turmoil: Bad Luck or Bad Policy? (CambridgeMassachusetts: MIT Press).

    ChowPeter C.Y. and BatesGilleds.2000Weathering the Storm: Taiwan Its Neighbors and the Asian Financial Crisis (Washington: Brookings Institution Press).

    • Search Google Scholar
    • Export Citation

    CisternasCarlos1999“Ecuador Leader Defers Payment” Associated PressAugust26. Accessed at

    DasekingChristinaAtishGhoshTimothyLane and AlunThomas2004Lessons from the Crisis in Argentina IMF Occasional Paper No. 236 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    EichengreenBarry and RicardoHausmanneds.2005Other People’s Money: Debt Denomination and Financial Instability in Emerging Market Economies (Chicago: University of Chicago Press).

    • Search Google Scholar
    • Export Citation

    EichengreenBarry and DonaldMathieson1998Hedge Funds and Financial Market Dynamics IMF Occasional Paper No. 166 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    FeldsteinMartined.2003Economic and Financial Crises in Emerging Market Economies (Chicago: University of Chicago Press).

    FragaArminio1999“Monetary Policy in a Transition to a Floating Exchange Rate: Remarks on the Recent Brazilian Experience,”New Challenges for Monetary Policy: A Symposium (Kansas CityMissouri: Federal Reserve Bank of Kansas City) pp. 14954.

    • Search Google Scholar
    • Export Citation

    GhoshAtish R.Timothy D.LaneMarianneSchulze-GhattasAlešBuliřJavierHamann and AlexMourmouras2002IMF-Supported Programs in Capital Account Crises IMF Occasional Paper No. 210 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    HenningRandall2002East Asian Financial Cooperation (Washington: Institute for International Economics).

    HicklinJohnDavid J.Robinson and AnoopSingh1997Macroeconomic Issues Facing ASEAN Countries (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Independent Evaluation Office2003The IMF and Recent Capital Account Crises: Indonesia Korea Brazil (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Independent Evaluation Office2004The IMF and Argentina 1991–2001 (Washington: International Monetary Fund).

    International Monetary Fund (IMF)1998World Economic Outlook: Financial Turbulence and the World Economy; A Survey by the Staff of the International Monetary Fund (Washington: International Monetary Fund). Accessed at

    • Search Google Scholar
    • Export Citation

    JácomeH.LuisI.2004“The Late 1990s Financial Crisis in Ecuador: Institutional Weaknesses, Fiscal Rigidities, and Financial Dollarization at Work,”IMF Working Paper 04/12 (Washington: International Monetary Fund). Accessed at

    • Search Google Scholar
    • Export Citation

    JohnsonSimon and ToddMitton2003“Cronyism and Capital Controls: Evidence from Malaysia,”Journal of Financial Economics Vol. 67No. 2 (February) pp. 35182.

    • Search Google Scholar
    • Export Citation

    KaminskyGraciela L. and Carmen M. Reinhart1999“The Twin Crises: The Causes of Banking and Balance-of-Payments Problems,”American Economic Review Vol. 89 No. 3 pp. 473500.

    • Search Google Scholar
    • Export Citation

    KrugmanPaul1999“Balance Sheets, the Transfer Problem, and Financial Crises,”International Tax and Public Finance Vol. 6 No. 4 pp. 45972.

    • Search Google Scholar
    • Export Citation

    LaneTimothyAtishGhoshJavierHamannStevenPhillipsMarianneSchulze-Ghattas and TsidiTsikata1999IMF-Supported Programs in Indonesia, Korea, and Thailand: A Preliminary AssessmentIMF Occasional Paper No. 178 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Lee YongWook2006“Japan and the Asian Monetary Fund: An Identity-Intention Approach,”International Studies Quarterly Vol. 50 No. 2 pp. 33966.

    • Search Google Scholar
    • Export Citation

    LindgrenCarl-JohanTomás J.T.BaliñoCharlesEnochAnne MarieGuldeMarcQuintyn and Leslie EngSipp Teo2000Financial Sector Crisis and Restructuring: Lessons from AsiaIMF Occasional Paper No. 188 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    LopesFrancisco L.2003“Notes on the Brazilian Crisis of 1997–99,”Revista de Economia Politica Vol. 23 No. 3 pp. 3562.

    MahathirMohamad2000The Malaysian Currency Crisis: How and Why It Happened (Subang Jaya Selangor Darul EhsanMalaysia: Pelanduk Publications).

    • Search Google Scholar
    • Export Citation

    MalanPedro2004“Crises Prevention and Crises Resolution: The Roles of the IMF Governments and Markets” in the proceedings of “60 Years of Bretton Woods—The Governance of the International Financial System—Looking Ahead June 20 to 22 2004” Proceedings of OeNB Workshops No. 3 (Vienna: Oesterreichische Nationalbank).

    • Search Google Scholar
    • Export Citation

    Milesi-FerrettiGian Maria and AssafRazin1999“Current Account Deficits and Capital Flows in East Asia and Latin America: Are the Early Nineties Different from the Early Eighties?” in Changes in Exchange Rates in Rapidly Developing Countries: Theory, Practice and Policy Issuesed. by TakatoshiIto and AnneO. Krueger (Chicago: University of Chicago Press).

    • Search Google Scholar
    • Export Citation

    MoghadamRezaed.2005Turkey at the Crossroads: From Crisis Resolution to EU AccessionIMF Occasional Paper No. 242 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    MulderChristianRobertoPerrelli and ManuelRocha2011“External Vulnerability, Balance Sheet Effects, and the Institutional Framework - Lessons from the Asian Crisis,”International Review of Economics and Finance Vol. 21 No. 1 pp. 1628.

    • Search Google Scholar
    • Export Citation

    MussaMichael2002Argentina and the Fund: From Triumph to Tragedy (Washington: Peterson Institute for International Economics).

    MydansSeth1997“Malaysian Premier Sees Jews Behind Nation’s Money Crisis,”New York TimesOctober16 p. 8.

    ÖnişZiya and BarryRubineds.2003The Turkish Economy in Crisis (London: Frank Cass).

    OstryJonathan D.1997“Are Current Account Imbalances in ASEAN Countries a Problem?” in Macroeconomic Issues Facing ASEAN Countriesed. byJohnHicklinDavidRobinson and AnoopSingh (Washington: International Monetary Fund) pp. 2957.

    • Search Google Scholar
    • Export Citation

    PérezLorenzo L. and Philip R.Gerson2009“Brazil: Anchoring Policy Credibility in the Midst of Financial Crisis” in Successes of the International Monetary Fund: Untold Stories of Cooperation at Worked. byEduardBrau and IanMcDonald (Houndmills and New York: Pal-grave Macmillan) pp. 10620.

    • Search Google Scholar
    • Export Citation

    PuraRaphael and Michael M.Phillips1998“Concern Rises for Dissident in Malaysia,”Wall Street JournalOctober2 p. A10.

    Research Department Central Bank of Brazil2000“Issues in the Adoption of an Inflation Targeting Framework in Brazil,” in Inflation Targeting in Practice: Strategic and Operational Issues and Application to Emerging Market Economies, Proceedings of a Seminar held in Rio de Janeiro, Brazil, May 3–5, 1999ed. byMario I.BlejerAlainIzeAlfredoM. Leone and SérgioWerlang (Washington: International Monetary Fund) pp. 7079.

    • Search Google Scholar
    • Export Citation

    RieffelLex2003Restructuring Sovereign Debt: The Case for Ad Hoc Machinery (Washington: Brookings Institution Press). la PiedraP.KongsamutK.KostialV.Summers and A.Vamvakidis2000Philippines: Toward Sustainable and Rapid GrowthIMF Occasional Paper No. 187 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    RubinRobert E. and JacobWeisberg2003In an Uncertain World: Tough Choices from Wall Street to Washington (New York: Random House).

    SakakibaraEisuke2000The Day Japan and the World Shook: The Birth of Cyber Capitalism (in Japanese) (Tokyo: Chuo-Koron Shin Sha).

    SakakibaraEisuke2001“The Asian Monetary Fund: Where Do We Go From Here” address given at the Institute of Strategic and International Studies International Conference on Globalization Kuala LumpurFebruary26. Accessed at

    • Search Google Scholar
    • Export Citation

    ShengAndrew2009From Asian to Global Financial Crisis: An Asian Regulator’s View of Unfettered Finance in the 1990s and 2000s (Cambridge, England: Cambridge University Press).

    • Search Google Scholar
    • Export Citation

    StiglitzJoseph E.2002Globalization and Its Discontents (New York: W.W. Norton).

    SturzeneggerFederico and JerominZettelmeyer2006Debt Defaults and Lessons from a Decade of Crises (Cambridge, Massachusetts: MIT Press).

    • Search Google Scholar
    • Export Citation

    TamirisaNatalia2001“Capital Controls in Response to the Asian Crisis,” in Malaysia: From Crisis to Recoveryed. byKanitta Meesook Il Houng Lee Olin Liu Yougesh Khatri Natalia Tamirisa Michael Moore and Mark H. Krysl IMF Occasional Paper No. 207 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    TanziVito2007Argentina: An Economic Chronicle (New York: Jorge Pinto Books).

    TharmanShanmugaratnam2006Asian Monetary Integration: Will It Ever Happen? Per Jacobsson Lecture delivered September 17 (Washington: Per Jacobsson Foundation).

    • Search Google Scholar
    • Export Citation

    WilliamsonJohn2004“The Years of Emerging Market Crises: A Review of Feldstein,”Journal of Economic Literature Vol. 52 (September) pp. 82237.

    • Search Google Scholar
    • Export Citation
1The Chinese currency is the renminbi. The basic unit of value is the yuan.
2See “People’s Republic of China—Staff Report for the 1997 Article IV Consultation,” SM/97/137 (June 3, 1997).
3After meetings in Beijing in January 1998, Camdessus held a press conference in Kuala Lumpur, Malaysia, during which he reported that the Chinese authorities “rightly intend to maintain the present value of the renminbi.” For the transcript, see Later, U.S. Treasury Secretary Robert E. Rubin took pride in having helped to avoid trouble. “Several times, in meetings with President Clinton, with others in the administration, or with me, President Jiang Zemin and Premier Zhu Rongji underscored the firmness of their commitment not to devalue the Chinese currency. And they never did” (Rubin and Weisberg, 2003, p. 227).
4See, for example, “Monetary Relations between the Mainland of China and Hong Kong,” speech by Chen Yuan (deputy governor of the People’s Bank of China) at the Federal Reserve Bank of New York, February 14, 1997; circulated in the IMF as EBD/97/17 (February 20, 1997).
5“People’s Republic of China—Staff Report for the Article IV Consultation Discussions Held in 1997 in Respect of the Hong Kong Special Administrative Region,” SM/97/295 (December 29, 1997), pp. 20–21.
6The December 1997 staff report (cited in the preceding footnote) noted that the “staff continues to endorse the authorities’ commitment to the linked exchange rate system” (p. 20). A year later, while noting that “some have argued … [for a shift] to a more flexible exchange rate regime, the staff … remains strongly of the view that the linked exchange rate system should be maintained”; “People’s Republic of China—Hong Kong Special Administrative Region—Staff Report for the 1998 Article IV Consultation Discussions,” SM/99/4 (January 8, 1999), p. 36.
7See minutes of EBM/99/12 (January 29, 1999), pp. 7–46.
8For the earlier period, see Chapter 9. For a more detailed discussion of the reform of the economy, see Rodlauer and others (2000) and Balisacan and Hill (2003).
9As an interim step, the Fund extended the arrangement for one month while talks proceeded on a longer extension and corresponding revisions to economic policies; see “Philippines—Extended Arrangement—Request for Extension of Period,” EBS/97/111 (June 18, 1997). That extension was approved and took effect on June 23, the date that the three-year arrangement was scheduled to expire.
10This sequence of events was conveyed to the Executive Board by Aghevli at EBM/97/74 (July 18, 1997); see minutes, p. 11. Also see “Philippines—Fourth Review Under the Extended Arrangement and Request for Extension, Rephasing, and Augmentation,” EBS/97/70, Suppl. 1 (July 14, 1997). Additional detail is from interviews with participants.
11In all, the Fund extended the EFF arrangement four times from the original expiration date in June 1997, ultimately to March 1998. After augmentation in July 1997, the IMF’s total commitment under the arrangement was SDR 791.2 million. The authorities borrowed the full available amount in three tranches: the initial drawing in June 1994, the large one on augmentation, and a final one just before expiration in April 1998.
12The 1994 control was a ceiling on the external liabilities of domestic banks, excluding borrowing to finance current account transactions or foreign direct investment. For analyses of Malaysia’s capital controls, see Athukorala (2001) and Tamirisa (2001). Table 5.1 of the latter provides a detailed chronology of controls from 1994 to 2001. The Fund staff later noted that the upward speculative pressure on the ringgit in 1991–92 resulted in large part from the advent of large-scale carry trade at that time (Eichengreen and Mathieson, 1998, p. 17).
13“Malaysia—Staff Report for the 1994 Article IV Consultation,” SM/94/239 (September 1, 1994), p. 11. The general agreement between the Malaysian authorities and the IMF staff was expressed directly in the following year’s report: “The staff agrees with the authorities that restrictions on capital flows should be used only as a last resort, as they can lead to market distortions and an inefficient allocation of resources”; “Malaysia—Staff Report for the 1995 Article IV Consultation,” SM/95/200 (August 16, 1995), p. 25.
14Minutes of EBM/96/69 (July 19, 1996), pp. 47–48.
15“Challenges Facing the IMF and Malaysia,” address by Michel Camdessus at a meeting of financial and business leaders, Kuala Lumpur, July 15, 1996; accessed at Mahathir had announced a goal of completing the transformation of Malaysia into an industrial country by 2020.
16For the official staff view, see “Malaysia—Staff Report for the 1996 Article IV Consultation,” SM/96/217 (August 15, 1996). Also see Milesi-Ferretti and Razin (1999) and Ostry (1997), both of which found that Malaysia’s external deficits were relatively manageable because of strong investment inflows.
17See “Malaysia—Staff Report for the 1997 Article IV Consultation,” SM/97/197 (July 31, 1997).
18The quoted phrases are from remarks made by Mahathir at a forum during the IMF/World Bank Annual Meetings in Hong Kong SAR on September 20, 1997. The text was published in ExecutiveIntelligence Review (LaRouche Publications) of October 3, 1997. Also see “Mahathir’s Roasting,” Economist (London), September 27, 1997, p. 39; and Mydans (1997).
19Letter from Camdessus to Mahathir, February 12, 1998; IMF archives, Accession No. 2001-0284-0009, OMD-DMD, B28266, “Malaysia 1998.”
20“Malaysia—Staff Report for the 1998 Article IV Consultation,” SM/98/79 (March 27, 1998).
22For the prime minister’s own account of the crisis and his reactions to it, see Mahathir (2000).
23For an overview of these developments, see “Malaysia—Staff Report for the 1999 Article IV Consultation,” SM/99/141 (June 16, 1999).
24An internal Fund report prepared immediately after the imposition of controls took note of the potential for long-term harm but did not propose objecting to them; memorandum from Bijan Aghevli and Anoop Singh (both Deputy Directors, APD) to the Managing Director, “Malaysia: Update and Preliminary Assessment,” September 9, 1998; IMF archives, Accession No. 2001-0284-0009, OMDDMD, B28266, “Malaysia 1998.” A staff team, led by Kalpana Kochhar (Deputy Division Chief, APD), then visited Kuala Lumpur to assess the specific impacts. Kochhar thought the authorities were being secretive about the rationale for imposing controls and were unwilling to share information fully with the Fund. Nonetheless, she reported that the controls were not inconsistent with Malaysia’s obligations under the Articles of Agreement and that there was no basis for the Fund to object to them; see memorandum from Hubert Neiss (Director, APD) to the Managing Director, “Malaysia—Staff Visit, September 21–25,” September 30, 1998; IMF archives, Accession No. 2001-0284-0009, OMD-DMD, B28266, “Malaysia 1998”; and “Malaysia—Inquiry Under Article VIII, Section 2(b),” EBD/99/33 (February 23, 1999).
25Johnson and Mitton (2003) found empirical evidence that the controls benefited politically connected corporations.
26After being sacked from his post, Anwar spoke out publicly against the policy changes of September 1. He was then jailed on corruption and other charges widely thought to be unjustified and imposed for political reasons. Anwar would spend the next six years in prison, after which the charges against him were dismissed.
27In the course of a press conference preceding the Annual Meetings, World Bank President James D. Wolfensohn remarked that Anwar was “a friend of mine. When I see him with a black eye and bruises [as a result of abuse while in prison], it troubles me.” U.S. Treasury Secretary Robert Rubin told a public conference that the reported abusive treatment of Anwar was “deeply, deeply, deeply troubling”; see Pura and Phillips (1998). Later, at a summit meeting of the Asia-Pacific Economic Cooperation (APEC) forum in Kuala Lumpur on November 16, 1998, U.S. Vice President Al Gore made a veiled but undiplomatic attack on Mahathir by crediting the “brave people of Malaysia” who were calling for “reformasi” (the rallying cry of Anwar’s supporters). For the transcript, see Japan, in contrast, launched the New Miyazawa Initiative in early October. The initiative offered loans to countries throughout the East Asia region, but the timing of its launch aimed to provide support to Malaysia when few others were willing to do so. For the text of the announcement, see
28For the full transcript, including Fischer’s elaboration of the long-run costs of maintaining controls on capital flows, see The Fund’s World EconomicOutlook report published in October 1998 included a similar caution (IMF, 1998, p. 4).
29The staff’s first detailed postcrisis analysis—following a staff mission to Kuala Lumpur in April 1999—was set out in “Malaysia—Staff Report for the 1999 Article IV Consultation,” SM/99/141 (June 16, 1999).
30Minutes of EBM/99/74 (July 7, 1999), pp. 128–29.
31For a review of the effects of the Thai crisis on the Lao economy, see “Lao People’s Democratic Republic—Staff Report of the 1998 Article IV Consultation,” SM/98/118 (May 28, 1998).
32The CIS-7 comprised the seven low-income countries in the Commonwealth of Independent States (CIS): Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan.
33See “New Financial Support for Poorest Countries Neighboring Russia,” IMF NB/98/51, December 11, 1998. Uzbekistan, where the reform process was less advanced than in the other CIS-7 countries, and which had not been approved as eligible for concessional loans, was excluded from the support effort.
34Counting countries is somewhat arbitrary because of the breakup of both Czechoslovakia and Yugoslavia in 1992. Those two members borrowed from the Fund before their dissolution, and afterward most of their successor states did so, too. Of the immediate successors to Yugoslavia, neither Slovenia nor Serbia borrowed in the 1990s.
35In the second half of the 1980s, the only European borrowers were Hungary and Yugoslavia.
36In the course of 1995, the Fund augmented and extended the stand-by arrangement, but policy disputes prevented the authorities from drawing on it after September. The arrangement expired in March 1996 with SDR 150 million undrawn.
37“Turkey—Memorandum of Economic Policies,” EBD/98/72 (July 2, 1998).
38See “Turkey—Use of Fund Resources—Request for Emergency Assistance,” EBS/99/191 (October 5, 1999). A staff mission had been in Ankara and Istanbul throughout the second half of June, but shortly after the mission left the economy minister who had been leading the discussions for the authorities (Hikmet Uluğbay) attempted suicide and then resigned from office. The earthquake hit before discussions with the new Turkish team could begin.
39The crisis and the role of the IMF are analyzed in Öniş and Rubin (2003). For the aftermath and consequences, see Moghadam (2005).
40Argentina, Brazil, Canada, Chile, Colombia, Ecuador, Mexico, Peru, the United States, Uruguay, and Venezuela participated in the meeting.
41In opening remarks at the meeting, Camdessus explicitly noted that it was intended to be a prototype for regional surveillance; see memorandum from Claudio Loser (Director, Western Hemisphere Department, or WHD) to the Managing Director, “MD’s Speech for Western Hemisphere Countries Conference,” September 3, 1998; IMF archives, “Meeting of Economic Policy Makers August-September 1998,” OMD-AI, Accession No. 2007-0131-01. The closest precedents had involved IMF participation in regional or other affinity-group meetings organized and run by the country groups (e.g., the G7 or ASEAN countries) rather than by the Fund.
42At the end of the first day, the Fund issued a statement to the press; see “Communiqué of Meeting of Economic Policy Makers in the Western Hemisphere Region,” PR/98/37 (September 3, 1998); accessed at
43“Brazil—Recent Economic Developments,” SM/97/44 (February 13, 1997), pp. 70–81.
44“Brazil—Staff Report for the 1996 Article IV Consultation,” EBS/97/11 (January 30, 1997), p. 23.
45In Brazil’s adjustable-peg regime, the central bank announced a wide band, or “maxiband,” on an annual basis and then intervened to keep the rate within a narrow “miniband” that they changed much more frequently. The policy in effect at that time, consistent with the outcome of the 1996 consultation, was to adjust the maxiband each year to effect a depreciation of about 7.5 percent over the year. In January 1998, the central bank announced an adjustment of that magnitude. For the Managing Director’s reaction to the fiscal adjustment, see NB/97/24, “IMF’s Camdessus Welcomes Brazilian Fiscal Policy Package,” November 10, 1997; accessed at
46Quoted in Dornbusch’s obituary in MIT News (July 26, 2002); accessed at
47“Verärgerung über den IMF” [“Anger over the IMF”], Neue Zürcher Zeitung (August 27, 1998); accessed on For a more detailed account based on a transcript of the conference call and an interview with Folkerts-Landau, see Blustein (2001), pp. 274–77.
48See memorandum from Teresa Ter-Minassian (Deputy Director, WHD) to Fischer, “Possible Talking Points for a Telephone Call to Minister Malan” (August 27, 1998); IMF archives, OMD-DMD (Mr. Fischer’s files), Accession 2001-0284, “Brazil-1998.”
49See memorandum from Fischer to the Managing Director, “Phone calls this morning,” September 11, 1998; IMF archives, OMD-DMD (Mr. Fischer’s files), Accession 2001-0284, “Brazil-1998.”
50Memorandum from Claudio Loser (Director, WHD) to Fischer, “Brazil—Exchange Rate Policy” (September 14, 1998); IMF archives, OMD-DMD (Mr. Fischer’s files), Accession 2001-0284, “Brazil-1998”
51Translation of the speech, forwarded to the Managing Director by Claudio Loser (September 23, 1998); IMF archives, OMD-DMD (Mr. Fischer’s files), Accession 2001-0284, “Brazil-1998.”
52See “Statement by IMF Managing Director on Brazil,” NB/98/34 (September 23, 1998); accessed at
53See memorandum from Ter-Minassian to the Managing Director, “Brazil—State of Discussions” (October 3, 1998); IMF archives, OMD-DMD (Mr. Fischer’s files), Accession 2001-0284, “Brazil-1998.”
54Fischer made a quick trip to Brazil on October 23 to take up the key issues directly with Malan and other senior officials; see minutes of EBM/98/108 (October 26, 1998), pp. 3–6.
55Minutes of EBM/98/189 (December 2, 1998), pp. 76–77. For an analysis of the program, including the Fund’s willingness to lend to Brazil without a devaluation, see Independent Evaluation Office (2003), Annex 3.
56See statement by Murilo Portugal (Executive Director for Brazil) at EBM/98/122 (December 2, 1998), p. 9. For the background to the NAB and the SRF, see Chapters 15 and 5, respectively.
57Of that amount, SDR 2,876 million was in the form of an SRF loan with a higher interest rate and an expectation of early repayment, and was financed by activating the NAB. The rest (SDR 543 million) was a conventional credit-tranche purchase. The total was equivalent to 157 percent of Brazil’s quota, or 113 percent of the enlarged quota that was to take effect in January 1999.
58Blustein (2001) pp. 355–69, gives a detailed account of this episode. Lopes (2003) describes the specific technical strategy for trying to avoid floating the currency. Fischer’s advice to float seems to have become muddled, as Lopes interpreted the message quite differently.
59The Letter of Intent signed by Malan and Gustavo Franco in November 1998 concluded with the standard boilerplate commitment: “During the period of the arrangement, the authorities of Brazil will maintain close relations with the Fund, and will consult on the adoption of policy measures that may be needed, in accordance with existing practices.” The text of the stand-by arrangement approved by the Fund in December specified that in accordance with that letter, “Brazil will consult with the Fund on the adoption of any measures that may be appropriate at the initiative of the government or whenever the Managing Director requests consultation because any of the [performance] criteria … have not been observed or because the Managing Director considers that consultation on the program is desirable”; “Brazil—Request for Stand-By Arrangement—Letter of Intent,” EBS/98/189 (November 12, 1998); and “Brazil—Stand-By Arrangement,” EBS/98/189, Suppl. 2 (December 4, 1998), paragraph 10.
60“IMF’s Managing Director’s Statement on Brazil,” NB/99/2 (January 13, 1999); accessed at
61For the IMF staff view in favor of a currency board, see memorandum from Adam Bennett (Chief of the Stand-By Operations Division in the Policy Development and Review Department) to Fischer, “Brazil—Considerations towards a Currency Board Arrangement,” January 22, 1999; IMF archives, DMD-AI, Accession 2002-0149, box B30552. (The staff team in the WHD were less convinced that a currency board was appropriate for Brazil.)
62Fischer’s advocacy of corner solutions to exchange rate policy is discussed in Chapter 1, pp. 23–24.
63See Fraga (1999) and Research Department, Central Bank of Brazil (2000). In May 1999, the IMF organized a seminar on inflation targeting in Rio de Janeiro, at which central bankers from around the world made presentations on their experience with inflation-targeting regimes; see A key issue for the IMF was how to adapt the policy conditions in the stand-by arrangement, which had been designed in the context of a fixed exchange rate, to the new regime. The technical requirements are discussed in “IMF Conditionality in the Context of Inflation Targeting—The Case of Brazil,” SM/99/296, Suppl. 1 (December 16, 1999), and in Blejer and others (2001).
64“Joint Statement of the Ministry of Finance of Brazil and the IMF Team,” NB/99/5 (February 4, 1999); accessed at For Fischer, this detour to Brazil was one of the most memorable experiences of his seven years at the IMF. On leaving the Fund in 2001, he recalled “leaving for Brazil from Davos at 4 a.m., driving through the beautiful snowcovered moonlit mountains, worrying that this could become a disaster. But thanks to the steadfastness of President Cardoso, and the skill of Pedro Malan and Arminio Fraga, helped by the outstanding work of the Fund team led by Teresa Ter-Minassian, disaster was avoided”; farewell dinner speech, August 29, 2001; accessed at
65“Brazil—First and Second Reviews Under the Stand-By Arrangement,” EBS/99/30, Suppl. 2 (March 25, 1999), p. 15. These meetings took place in mid-March 1999, in New York, Tokyo, Frankfurt, Lisbon, London, Madrid, Paris, and Rome. Malan (2004) p. 167, includes a memoir of the road show.
66For a review, see Pérez and Gerson (2009).
67Brazil did not draw on the stand-by arrangement in 2000, and it made just one small drawing in 2001 before the onset of a new crisis led to negotiation of more loans. In all, Brazil borrowed about two-thirds of the $18.3 billion committed by the Fund for the 1998–2001 arrangement. All of the SRF portion was repaid by April 2000. Even after large borrowings in 2001–05, Brazil repaid all of its outstanding obligations to the IMF by February 2006.
68Minutes of EBM/95/35 (April 6, 1995), p. 9. The reference was to the 1977 popular song “Hotel California,” written and recorded by the Eagles.
69Minutes of EBM/95/35 (April 6, 1995), pp. 28 (Lissakers), 29 (Camdessus), and 37 (Autheman). The “skepticism” quotation is from Camdessus.
70Menem’s announcement was made in an interview with a television reporter in Buenos Aires; see “Argentine President Wants Study ‘Quickly’ on Switch to Dollars,” Dow Jones International News, January 15, 1999. The announcement that Argentina was seeking a treaty with the United States was made by the central bank governor, Pedro Pou, the following week; see “Argentine Central Bank Proposes Money Link with U.S.,” Reuters News, January 21, 1999. Both stories were accessed at
71This account is based primarily on interviews with participants. In April 1999, U.S. Treasury Secretary Robert Rubin addressed the issue in remarks at the Johns Hopkins University campus in Washington. Without ever mentioning Argentina, he observed that “some countries” were considering dollarizing. He then noted, “We do not have an a priori view as to our reaction to the concept of dollarization. We would also observe that there are a variety of possible ways for a country to dollarize. But it would not, in our judgment, be appropriate for United States authorities to extend the net of bank supervision, to provide access to the Federal Reserve discount window, or to adjust bank supervisory responsibilities or the procedures or orientation of U.S. monetary policy in light of another country’s decision to dollarize its monetary system”; see “Treasury Secretary Robert E. Rubin Remarks on Reform of the International Financial Architecture to the School of Advanced International Studies,” U.S. Treasury press release RR-3093 (April 21, 1999); accessed at
72“Argentina—Third Review under the Stand-By Arrangement,” EBS/97/133 (July 16, 1997), p. 13.
73Minutes of EBM/97/54 (May 28, 1997), p. 3.
74The pattern of Argentina’s borrowing from the IMF in the 1990s is shown in Chapter 10 (Figure 10.3).
75“Argentina—Staff Report for the 1997 Article IV Consultation and Request for Extended Arrangement,” EBS/98/6 (January 14, 1998), p. 22.
76Transcript of press conference, October 1, 1998; accessed at The text of Menem’s October 6 speech is available at
77For a strikingly frank criticism of Camdessus’s position by a senior official of the Fund, see Tanzi (2007), pp. 117–18.
78Minutes of EBM/99/53 (May 12, 1999), p. 4.
79For critiques of the IMF’s relations with Argentina in the lead-up to the 2002 crisis, see Internal Evaluation Office (2004) and Mussa (2002). For a detailed analysis of the conditions that led to the crisis, see Daseking and others (2004).
80For an analysis of the crisis, see Jácome (2004).
81“Ecuador—Staff Report for the 1997 Article IV Consultations,” EBS/97/212 (August 19, 1997), p. 6. Relations with Ecuador before the staff-monitored program are covered in Chapter 9.
82The emphasis is in the original. Memorandum from Jesus Seade (Assistant Director, PDR) to Claudio Loser (Director, WHD), “Ecuador—Strategy for Dealing with External Debt Problems,” April 6, 1999; IMF archives, Historian’s files.
83Memorandum for files by Thornton and Robert Kahn (Deputy Division Chief, PDR), “Meeting with Ecuadoran Debt Policy Team and Their Legal/Financial Advisors” (May 18, 1999); IMF archives, Historian’s files.
84See Chapter 1, pp. 31–32. On the application of the G7 strategy to the Ecuador case, see Rieffel (2003), pp. 239–42.
85Report by Fischer to the Executive Board; minutes of EBM/99/53 (May 12, 1999), p. 4.
86Memorandum for files by Mariano Cortés (Economist, WHD), “Ecuador—Authorities’ Meeting with Management,” August 24, 1999; IMF archives, DMD-AI (Mr. Fischer’s files), “Ecuador – 1999 (1),” Accession No. 2002-0149.
87“IMF Announces Conclusion of Staff Negotiations with Ecuador,” NB/99/53 (August 31, 1999); accessed at This announcement was issued only after Mahuad assured the IMF that he was personally committed to implementing the agreed-on terms of the program; see memorandum from Loser to the Managing Director, “Ecuador—Letter from President Mahuad,” August 31, 1999; IMF archives, DMD-AI (Mr. Fischer’s files), “Ecuador – 1999 (1),” Accession No. 2002-0149.
88For a more detailed description of the available options, see memorandum from Jack Boorman (Director, PDR) and François Gianviti (General Counsel) to the Managing Director, “Ecuador: Debt Strategy Options” (September 21, 1999); IMF archives, DMD-AI (Mr. Fischer’s files), “Ecuador – 1999 (1),” Accession No. 2002-0149.
89See transcript of press conference at
90“IMF Urges Collaboration Between Ecuador and Its Creditors,” NB/99/61 (September 27, 1999); accessed at
91For an analysis of the default and its implications, see Sturzenegger and Zettelmeyer (2006), pp. 147–64.
92Minutes of EBM/98/38 (April 2, 1998), p. 6.
93Major cross-country studies by the staff included—in addition to the seminal paper by Lane and others (1999)Boorman and others (2000), Lindgren and others (2000), and Ghosh and others (2002). Also see Independent Evaluation Office (2003). For an introduction to the voluminous external analysis of the Fund’s role in the management of late-1990s financial crises, see Stiglitz (2002), Feldstein (2003), and Williamson (2004).
94This review does not attempt to cover all the issues raised regarding how to anticipate and manage financial crises. Instead, it summarizes the main themes that were motivated by the Fund’s handling of the crises of 1997–98. For a discussion of the effort to improve crisis prediction, see “Strengthening Crisis Prevention and Resolution” in Chapter 10.
95For overviews of the balance sheet and microeconomic analyses that developed in the wake of the 1990s financial crises, see Calvo (2005) and Eichengreen and Hausmann (2005). The former spotlights the role of “sudden stops” in capital inflows; the latter, the “original sin” of borrowing short-term in foreign currencies. Under both approaches, the possibility of multiple equilibria makes it impossible for a conventional model to predict whether a good or a bad outcome will prevail.
96“The Management of External Debt and Reserves in Emerging Markets,” FO/DIS/99/124 (September 9, 1999), p. 1.
97For a review and overview, see Mulder, Perrelli, and Rocha (2011).
98An academic literature on the implications of financial sector weakness for macroeconomic vulnerabilities existed before the Asian crisis, but it focused primarily on Latin America. Only a few prescient analysts foresaw the similarities in East Asia; see, for example, Kaminsky and Reinhart (1999), which is a revised version of a 1996 working paper. For a thoughtful postcrisis analysis of the financial sector weaknesses in East Asia, see the papers in Chow and Gill (2000).
99Boorman (Director, PDR) characterized this problem as the “two-try thesis.” Typically, the authorities would initially deny that they had a problem. Only when their first attempts to muddle through proved insufficient would someone take charge and deal with the problems more seriously (Boorman, 1999).
100“Guidelines on Conditionality,” Executive Board Decision No. 12864-(02/102), adopted September 25, 2002; available at
102Minutes of EBM/98/44 (March 26, 1998), p. 3.
103This account of the conference is based primarily on interviews with participants. The official proceedings were published as Hicklin, Robinson, and Singh (1997), but the discussions of interest here took place outside the formal structure.
104Report by Camdessus to the Executive Board; minutes of EBM/97/23 (March 12, 1997), p. 62.
105Sakakibara later made this point explicitly: “I well understand the criticism that IMF conditionality is too heavily involved. After all, it was the desire to create a policy alternative to the IMF prescription that motivated the proposal to create the AMF” (Sakakibara, 2001). For his complete memoir of the episode, see Sakakibara (2000). For a detailed outside account, see Lee (2006). Lee notes that Japanese officials tended to view excluding the IMF from the arrangement as equivalent to excluding the United States.
106For an account of the reaction of Asian creditor countries, see Sheng (2009), Chapter 1. Sheng notes that Sakakibara continued to try to generate support for his proposal until the Manila meeting, which this chapter addresses next.
107“Philippines—Communication from the First Deputy Managing Director,” EBD/97/126 (November 5, 1997).
108“A New Framework for Enhanced Asian Regional Cooperation to Promote Financial Stability,” Meeting of Asian Finance and Central Bank Deputies: Agreed Summary of Discussions, Manila, Philippines, November 18–19, 1997; accessed at An abridged version of the communiqué was published in IMF Survey, Vol. 26 (December 1, 1997), p. 371. The members of the MFG were Australia, Brunei Darussalam, Canada, China, Hong Kong SAR, Indonesia, Japan, Korea, Malaysia, New Zealand, the Philippines, Singapore, Thailand, and the United States.

    Other Resources Citing This Publication