NBER Project on Exchange Rate Crises in Emerging Markets

John McHale
February 16, 2001

Directors: Martin Feldstein and Jeffrey Frankel
Program Organizers: Jeffrey Frankel, Dani Rodrik, and Wing Thye Woo

The NBER project on exchange rate crises held the latest in its sequence of country-focused meetings on Malaysia on February 16, 2001. The venue was the Royal Sonesta Hotel in Cambridge, Massachusetts. The program organizers gathered a diverse group for a daylong debate on the Malaysian economic crisis. The participants included key Malaysian economic officials, current and former members of international organizations and the US government, members of the financial community, and distinguished academics from Asia and the United States.

Not surprisingly, the Malaysian government's experiment with capital controls dominated much of the discussion, though the topics addressed ranged widely. The program was divided into four sessions based on the temporal sequence of events: the situation before the outbreak of the Asian crisis in mid-1997; the period of contagion between mid-1997 and the imposition of capital controls in mid-1998; the period immediately following the imposition of capital controls; and finally Malaysia's future prospects in the aftermath of the crisis. This report provides a brief summary of the informal panel presentations in each session and the ensuing general discussions. Background papers distributed at the meeting are available at the NBER web site.

Session 1. Background: Malaysia Before Mid-1997

Chair: Jeffrey Frankel, Harvard University and NBER

Panelists: Jomo Kwame Sundaram, University of Malaya

Dwight Perkins, Harvard University

Homi Kharas, World Bank

Wing Thye Woo, UC Davis

Jomo Kwame Sundaram explains that the crisis followed a decade in which growth had averaged close to 8 percent. Malaysia had gone through a banking crisis in the mid-1980s, which induced the central bank to improve financial sector regulation and supervision. Signs that the financial system was "beginning to fray" appeared around 1996, however. Even so, the system was in a healthier state than the systems of other crisis-affected Asian economies, reflecting, in part, limitations on foreign borrowing. Most importantly, Malaysia had a relatively low exposure to short-term foreign loans.

There were three main areas of vulnerability. First, Malaysia was exposed to international portfolio investment flows, following active promotion of this form of finance from the early 1990s. Total stock market capitalization was more than four times GDP in the lead up to the crisis. Second, bank intermediation of domestic and foreign savings was problematic. Much of the funds went to finance consumption credit, property market investments, and stock purchases. The results were a consumption boom and an asset price bubble. And Third, competitiveness was declining in the face of competition from China, rising real wages, and the central bank's unwillingness to revise its dollar peg exchange rate policy.

Dwight Perkins observed that most of Malaysia's growth came from large imports of capital in the form of foreign direct investment and reinvested rents from its large exports of natural resource products. There was relatively little increase in total factor productivity particularly among the domestic Malaysian producers when compared to economies such as that of Taiwan. Specifically the government's efforts to promote growth had focused first on heavy industries run by the state and financed at first by natural resource rents and later with foreign debt and these proved to be extremely inefficient. The government then privatized these industries but did so as much to promote a Malay/Bumiputera billionaire business class as to increase the efficiency of the economic system.

Because few Bumiputera have much experience with the manufacturing sector, they are not in a position to lead a major industrialization effort, and, in fact, much of the government bail out effort since the onset of the Asian financial crisis has been designed to rescue these large Bumiputera businesses many of which were in serious financial trouble. Chinese-Malaysian businesses, in contrast, do have considerable experience in manufacturing, but they are hampered by the need to meet Bumiputera ownership requirements set by the earlier New Economic Policy but still in force and by the fact that most government support is not available to them. Nor are natural resources a source of growth for the future as they were throughout much of Malaysia's history up to the early 1980s. Malaysia's future export and GDP growth prospects, as a result, depend mainly on the continued large scale inflow of foreign direct investment. As long as these flows continue, growth will continue, but Malaysia is taking unnecessarily serious risks by relying on this one source of growth almost exclusively. A less risky approach would be to continue to promote foreign direct investment, but to also take steps to free up the entrepreneurial talents of its domestic entrepreneurs.

Perkins concluded by noting Malaysia has two choices for securing its future growth: do everything possible to attract FDI or "loosen up" the domestic economy.

Homi Kharas said that Malaysia had many decades of rapid growth, but it had been reinventing itself every ten years or so, shifting its emphasis to different sectors--plantation sector, to state-owned enterprises, to private enterprises. The last transformation created a number of vulnerabilities, however. Turning to more positive factors, Kharas noted that the good performance was based firmly on fundamentals, including a massive investment in education. In international terms, Malaysia's growth and poverty reduction record had been second to none in the region, with the possible exception of Japan. This performance was based on very solid fundamentals as seen in various development and institutional indicators--corporate governance, financial regulation, public administration, etc.--as Malaysia looked to international best practice. The country also had very high household and corporate saving rates. There was a problem, however, in that capital markets were relatively underdeveloped and did not play their role of channeling saving into productive investments. Many major firms were not listed on the stock market. The only way for translating saving into investment was through debt.

Kharas pointed to the unusually close connection between politics and economics in Malaysia and the asset redistribution objectives of the government. To meet these goals, a large privatization program had been implemented but with many questions regarding its transparency. This, combined with access to bank credit, resulted in vulnerabilities as evident in various financial variables--increasing high corporate debt ratios, soaring credit to GDP ratios, and dramatic rises in stock market capitalization. All signs, in Kharas's view, of a "classic bubble economy."

Wing Thye Woo said he wanted to emphasize one key point--among the crisis-affected countries, Malaysia had the lowest ex ante probability of a crisis. This shows up from the logit estimations that he has done and most clearly in the relative default premia (measured in basis points):

Indonesia South Korea Thailand Malaysia

January 1, 1997 114 45 50

June 2, 1997 120 90 93 22

July 2, 1997 121 86 98 48

August 1, 1997 129 120 104 58

September 1, 1997 153 120 437 48

February 2, 1998 613 394 452 65

February 3, 1998 624 396 426 292

Why did the market treat Malaysia differently? Woo emphasized thee reasons: sound public finances; greater acceptability of "crony capitalism" under the social consensus of accepting race-based quotas in return for continued racial harmony; clearly defined mode of political succession.

General Discussion

Anoop Singh asked if the motivation for race-based economic policies was political or was it the need for social stability. Jomo responded by pointing to the importance of distinguishing between different sorts of race-based initiatives, some of which (e.g., investments in education and public health) can be efficiency enhancing.

Simon Johnson commented that economic institutions (such as protection of investor rights) were strong in the 1980s but weakened during the 1990s. Jomo agreed that developments with the judiciary seriously undermined investor confidence, and actually had a bigger impact on the confidence of domestic investors than on foreign investors. Homi Kharas pointed to a survey done by the World Bank on the quality of the judiciary in 1996. On a scale of 1 to 6, with 6 being the worst ranking, Malaysia scored a 2.9, compared to an average score for a set of high-income countries of 3.0. He added that businesses believed that the judiciary was strong.

Nouriel Roubini wondered what an investment rate of 45 percent and a growth rate of 8 percent implied about the profitability of the investment. Homi Kharas replied that as of mid-1990s Malaysian investment appeared to be relatively profitable when compared with countries like Thailand and Indonesia.

On Woo's question of why Malaysia was being treated differently, Angus Armstrong reported that net foreign liabilities were relatively low though (with a large current account deficit) rising. What did stand out was Malaysia's high bank credit to GDP ratio and its overheating property market. While these caused real concerns, there were not enough foreign liabilities to abruptly withdraw. When Thailand devalued many foreign investors asked "who is next," with many pointing to Malaysia and selling. Both Woo and Homi Kharas said that portfolio outflows did not drive the imposition of capital controls: major net portfolio outflows had ceased after 1997.

Jeff Frankel asked if there was a consensus that controls on capital inflows in 1994 and restrictions on banks shifted the composition of flows to the long end. He wondered if there is a lesson about the effectiveness of such controls. He noted work that he has done with Andrew Rose on the factors leading to financial vulnerability. They found that the size of the current account balance is not useful as a crisis predictor, but the composition of liabilities has predictive power.

Nor Mohamed Yakop reported that foreign direct investors are able to borrow up to 75 percent of the funds for their investments in Malaysia. He added that this was a sensible policy given the high domestic savings rate. Responding to Frankel, he said that the problematic capital flows in 1994 were not portfolio or FDI flows, but rather speculative currency flows. These were stopped, but with no lasting impact on the composition of flows.

See-Yan Lin asked why you should worry about the current account deficit if it is mainly financed by FDI. He also reported that in earlier times it was easier for the central bank to deal with speculative attacks because it knew who was speculating. The central bank was also in a strong position because of its substantial foreign exchange reserves. The ability of the central bank to control things had become more limited by 1998. Finally, Lin stressed how important the goal of "growth with equity" had been in guiding Malaysian policy.

Dwight Perkins agreed that the goal of "growth with equity" is central to understanding Malaysian economic policy. The race riots at the end of the 1960s made it clear that something had to be done to break the links between race and occupation. What was not obvious, he added, was how the securing of Malay ownership--e.g., through the 30 percent of ownership criterion--would affect economic structures over the years. Homi Kharas noted that the turning point in the "growth with equity" strategy came around 1989/90 when growth began to be associated with a deterioration in equity.

Jomo Kwame Sundaram said that banks became much more beholden to the central bank after getting into trouble in the 1980s. He added that financial liberalization sharpened the division of responsibilities between the central bank and financial supervisors, and caused a weakening of the banking system. On the issue of cronyism, he said that many of the beneficiaries were not Malay, adding that the business community is quite divided across ethnic lines. Finally, on the "growth with equity" question he observed that it is more important to think in terms of ethnic parity that to look at distributional statistics. For example, "are business people treated the same?"

Wing Thye Woo stressed that "growth with equity" need not be growth with "crony capitalism." In Malaysia, however, in the name of political stability "growth with equity" came to be more and more associated with bailing out particular individuals. He noted that in Malaysia listing on the stock market means handing over 30 percent of your shares to the government at a heavy discount. This, he said, is a very high marginal tax rate that limits a business's desire to grow. It has limited export growth by domestic firms, as evidenced by the handful of import-substituting Malaysia firms that matured into export-competing firms, a very different outcome from the Taiwanese experience. Malaysian exports have grown strongly only because of through strong FDI flows. Finally, he said that under the 30 percent rule individuals were handpicked for the opportunity to buy the discounted shares. The banks have frequently been "persuaded" by the government to extend loans to these handpicked individuals to finance their purchases. Hence, the race-based policies might have increased the financial vulnerability of the system through the lowered export potential and the high proportion of bank loans to finance equity purchases,

Session 2. Contagion: July 1997 to August 1998

Chair: Wing Thye Woo, UC, Davis

Panelists: Lin See Yan, Lin Associates

Anoop Singh, International Monetary Fund

Edwin Truman, Institute for International Economics; former U.S. Treasury & Federal Reserve

Takatoshi Ito, Ministry of Finance, Japan and NBER

Lin See Yan began with the observation that growth remained strong until the end of 1997 despite the Thai crisis in the middle of the year. Growth turned negative in 1998--significantly so in the third quarter. He believes that the fundamentals of the economy were still good at this stage--there were "strong vital signs"--so why, he asked did a crisis erupt. The answer is a combination of euphoria, panic, and contagion. He believes that speculative flows played a very large role in the crisis. This was made more destabilizing by the large offshore ringgit market, which gave speculators easy access to domestic currency.

The initial policy response was restrictive, combining tight monetary and fiscal policies. By the end of 1997 there was tremendous uneasiness that this policy was hurting growth. In January of 1998 the government set up a "new machinery" to handle crisis management. Key cabinet ministers were now focused on managing the crisis, and came up with a long list of policy responses. In Lin's view, this single-minded approach helped a lot.

Anoop Singh said that Malaysia's vulnerability lay partly in the fact that it had the world's highest proportionate stock market capitalization, and high credit growth. The corporate debt to equity ratio was moderate, but this understated the leverage problem given amount of borrowing by favored Malays (bumiputra) to buy shares. The crisis created a dilemma about how far the government should go in protecting the bumiputra community. Singh said that the economic team realized the historic nature of the choice--whether to shift course and allow market mechanisms to work. He added that the team saw the Asian crisis as an opportunity to restructure. Anwar thought he had only six months to implement the strategy. The vulnerability of corporate borrowers began to be exposed by late 1997, however, and the need for public funds to assist financial restructuring became apparent. Thus the government was faced with another choice--whether to prevent failures. As the contradictions grew among the policy makers, the Malaysian risk premium increased, and it became clear that the country would not be able to maintain growth. This lead policy makers to shift to a policy of capital controls and economic stimulus. So far, few costs of the capital controls have been perceived. Going forward, a key question is whether Malaysia has used this period to enhance competitiveness and to restructure. To the extent that the controls have protected non-competitive firms, Singh fears that this will weigh on future growth.

Drawing an analogy with the flu virus, Ted Truman noted that markets go up and down (just as the flu comes around every year), but your underlying health determines whether you are likely to succumb. For Malaysia, he said, a key question is whether its problems stemmed from contagion or self-inflicted wounds.

Truman briefly reviewed the evolving situation in other crisis-affected countries in early 1997. In his opinion Thailand was a "crisis waiting to happen" given the mishandling of the economy before the crisis. It was thought that the initial conditions for Indonesia were most similar to Malaysia. Indonesia, the argument went, did not need an IMF program, but technocrats saw it as a way to bring about change. This backfired badly. Korea was in trouble by July 1997, having already suffered a substantial loss of reserves and deepening crises in the Chaebols.

Thus "environmental factors" played a role in Malaysia's difficulties, as the country was hit by the "backwash" of developments in the region, including yen depreciation and the withdrawal of funds by Japanese banks.

Takatoshi Ito observed that contagion requires an epicenter, and asked how do we identify the source of the shock. One approach is to look week-by-week from July 1997 to January 1998 and to see how many weeks a country's currency had the largest depreciation in the region. Limiting attention to weeks where some country had a depreciation of at least 4 percent the ranking is: Thailand (7), Indonesia (6), Korea (2), Philippines (2), and Malaysia (1). From this Ito concludes that Malaysia is mostly a follower, and rarely the source of the shock.

Ito continued that the markets knew that Malaysia was following an IMF-type approach of high interest rates and tight fiscal policy without an IMF program, following much the same strategy as Thailand. They concluded that this strategy was not good for the economy. Policy makers might have been worried that, given similarities in terms of ethnic tensions, an Indonesian-type situation would develop. In any case, there was a radical shift in the monetary policy stance from the autumn of 1997 to the spring of 1998.

General Discussion

Carmen Reinhart said that the approach being taken by the panel to answer the question "contagion or fundamentals?" was to compare Malaysia with countries A, B, and C. But she pointed out that Malaysia's fundamentals were deteriorating compared to its own past. With reference to Woo's data on spreads from the first session, she added that spreads are not good predictors of crises. Malaysia's performance was already deteriorating and then "along comes Thailand." There was also a sharp reversal in Japanese and European bank lending. So there were deteriorating fundamentals and two channels of contagion.

Peter Garber observed that what he heard in the first session suggested substantial misallocation of capital. When this happens, he said, there will eventually be a speculative flow, but this is "a cause, not an effect."

Homi Kharas noted that though the stock market and the currency moved down together the consensus forecast in January 1998 was 5.5 percent growth. With the large wealth loss that had taken place, he wondered why people thought it would not have a big impact on consumption. He added that the reduction in wealth was suffered by a broad section of the population. Turning to the contagion issue, he noted that Malaysia is most connected to the Philippines, mainly through their common reliance on electronics. It is important to understand that the electronic industry is subject to cycles, and thus the downturn in electronics should not be viewed as a true permanent deterioration in fundamentals.

Caroline Atkinson said she agreed with Ted Truman that it is the "weak that get sick." She added that current account deficits of the size Malaysia was running must make you vulnerable.

Nor Mohamed Yakop said fundamentals had deteriorated and there was contagion as investors responded to what they saw in Thailand. But this does not explain the depth of the financial changes in Malaysia. He said that speculation is quite distinct from contagion, and noted that short selling was quite easy in Malaysia. Using offshore markets he said it was possible to make money without putting any money in at all. This type of short selling was stopped. The money available for short selling had vastly exceeded the central bank's reserves. Businessmen were complaining about the high interest rates, which led to falls in the stock market, and to further depreciation, to calls from the IMF to raise interest rates, and so on in a vicious cycle.

Nouriel Roubini asked what the role of the controls was in a period of contagion. He noted that there was a big difference between on-shore and off-shore interest rates, with the latter providing a better indicator of depreciation expectations. This big difference created an incentive to lend money out and so was a source of "bleeding funds."

David Malpass noted that in early 1997 a lot of investors were worried about Thailand's current account deficit and poor government. When Thailand devalued, those who had been arguing for short positions had their positions within their institutions strengthened. "Thai shorting had made a lot of money," he said. This led to a re-evaluation in peoples' minds of the risk-return opportunities in emerging markets. In October of 1997 the contagion process spread as people talked with each other at the Hong Kong IMF meetings. The focus then shifted to the vulnerability of pegs as opposed to current account deficits. It also made sense to short the stock markets in pegged countries given the perceived general vulnerabilities.

Carmen Reinhart said she was confused about the idea that initially there was an interest rate defense. She said she is not able to see it in the interest rate charts. Lin See Yan replied that the defense is more apparent in high frequency data. Lin also said he is not convinced that fundamentals were weak coming up to the crisis, and pointed in particular to the large budget surpluses.

Anoop Singh weighed in on the interest rate question by saying "there was no interest rate defense in Malaysia." There were short periods of higher interest rates, but there was no sustained defense. Such a defense was discussed but it was resisted by Malaysian decision-makers.

Ted Truman reiterated that, though Malaysia was affected by contagion, the biggest problem was the self-inflicted wounds. He also recalled that the international institutions were pushing for lower growth rate forecasts in the crisis-affected countries, but the national authorities resisted lowering the forecasts. Finally, he said that the focus on the damage done by high interest rates is a red herring. Once the financial crisis hit the weakness of the financial system was such that lower interest rates would not have helped.

Takatoshi Ito disagreed that interest rates had not been increased. He said that as at qualitative statement it is true that interest rates went up and fiscal policy was restrictive. He added that he would not necessarily call it "an interest rate defense." But it was an attempt to signal prudent macroeconomic policies to the market.

Nor Mohamed Yakop had the final word in the session. He stressed that interest rates did go up because spreads went up. He also raised the issue of the appropriate counterfactual for judging the interest rate stance, with the claim that interest rates should actually have been lowered during this period.

Session 3. Controls: The Imposition of Controls on Capital Outflows in September 1998

Chair: Martin Feldstein, Harvard University and NBER

Panelists: Nor Mohamed Yakop, The Prime Minister's Department

Dani Rodrik, Harvard University and NBER

Caroline Atkinson, Council on Foreign Relations, formerly US Treasury Yung Chul Park, Korea University

Nor Mohamed Yakop began the session by noting that the standard term for the package of measures instituted in Malaysia is capital controls. But the responsible officials did not view themselves as implementing capital controls. Rather, they saw themselves as pursuing measures that would create stability in the financial system, which was being destabilized by activities in the offshore market. Offshore accounts were frozen--but the ringgit continued to be a convertible currency. In his view, all that was done was a limiting of the access of speculators to domestic currency. He stressed that there are now no capital controls, so that direct and portfolio investors are free to move their money in and out. What exists is a "fixed exchange rate regime and efforts to deter speculators."

Dani Rodrik asked if the controls allowed Malaysia to recover more quickly. He noted that the prevailing view is that the controls, though not as damaging as predicted, did not help Malaysia. Critics point to the fact that Thailand and Korea began to recover at about the same time as Malaysia, indicating that it was a change in the external environment rather than the controls that explains the turnaround. He argued that these countries do not provide the relevant comparison. Capital market pressures were increasing in 1998 in Malaysia, but decreasing elsewhere. It is not appropriate to compare what was happening in countries that were 10-14 months into IMF programs with Malaysia in September 1998. Once appropriate adjustment is made for timing in a more formal econometric analysis, the positive impact of capital controls is more apparent.

Caroline Aktinson said that the Malaysian capital controls were neither a triumph nor a disaster, nor do they provide general lessons. Responding to Rodrik, she noted that Malaysia started off in better shape than other crisis-affected countries, and questioned the assumptions he used to control for timing. She said that if the package had been put in place in April 1998 rather than five months later we would have better knowledge of their impact. On the speed of recovery, she pointed out that Malaysia had an undervalued exchange rate, which allowed for a faster recovery. Turning to her doubts about general lessons, she noted that the central bank was able to exert substantial influence over bank behavior, an influence that won't be easy to replicate elsewhere.

Yung Chul Park asked what instruments Asian governments have available in the context of a managed exchange rate system. He said that governments would like to be able to use monetary policy for domestic stabilization purposes. Sterilized intervention, however, could be very expensive. Because of these tradeoffs some Asian governments are willing to entertain capital flows. They are not willing to move to a true floating rate regime. Turning to the more general question of capital account liberalization he asked: "What are the benefits?" Until we have a clear answer to this question, he said countries will continue to consider capital controls. Finally, he complained about the many foreign "noise traders" operating in Asian markets, noting that by looking only at current account balances and levels of reserves they often have a destabilizing impact on markets. The focus on the current account also forces governments to adopt costly policies to manage the size of the deficit; likewise with the management of foreign exchange reserves. This, he said, is a heavy price to pay for protection from speculators.

General Discussion

Wing Thye Woo said external assessments of the Thai bhat and Korean won were improving in mid 1998. But the assessment of the ringgit may also have been improving. As evidence he noted that the absence of a black market premium combined with the increases in the foreign reserves of the central bank suggested the exchange rate was being pegged at an undervalued level. He wondered if the reason for the strong recovery of the economy had less to do with the fiscal-monetary reflation pf the economy and more to do with Malaysia's strong export performance, which was caused by the undervaluation of the Ringgit.

Michael Bordo said that Dani Rodrik's counterfactual exercise is similar to work that he has done with Anna Schwartz in a recent NBER WP on the impact of IMF programs. They have found that countries with IMF programs did not show a stronger recovery from crises than countries with similar characteristics that did not have IMF programs.

Takatoshi Ito agreed that it is a misnomer to call what Malaysia did capital controls. The goal was to kill off the offshore market. A prohibition on the repatriation of foreign exchange was needed to kill off supply to the offshore market. He stressed that the goal is not the normal restriction of capital outflows, but rather a prohibition on speculative short positions.

Nouriel Roubini said capital controls had no impact on financial conditions or on real activity. By the end of August all speculative pressure had disappeared. This easing of speculative pressure also was apparent in countries as different as South Africa and Australia. He feels that controls had no impact on interest rates in Malaysia. He added that given that 1997 was a year of recession in Asia (including Malaysia), Rodrik's time shift analysis makes his comparison overly favorable to Malaysia. Rodrik responded that this ignores the different experiences of the countries before September 1998. Interest rates were not falling in Malaysia as they were in Korea and Thailand. Moreover, hedge funds weren't the only ones moving money off shore; domestic residents were also getting out of the domestic market with the result that there was very little domestic lending.

Michael Klein said we don't think that wage-price controls are good for an economy in the long term, but they might offer a window of opportunity to get through a crisis. The same might be true of capital controls.

Carmen Reinhart observed that if you fix the exchange rate you only have an independent monetary policy with capital controls. She reported that she has studied other instances of capital controls and has found that countries have difficulty de-coupling the domestic interest rate from the international rate. Malaysia apparently succeeded in doing this; but she cautioned about extrapolating the Malaysian experience to other circumstances

Jomo Kwame Sundaram noted that in imposing controls Malaysia closed down markets that had been a major part of the process of financial deepening. He also pointed out that the US interest rate reduction in August of 1998 helped to strengthen the East Asian currencies and to reduce exchange rate volatility. Turning to the Rodrik argument, he said that it supposes there was a crisis yet to happen. He agrees that the real economy, which had been shrinking from the fourth quarter of 1997, could have fallen deeper into crisis. But he noted that many people think that the decline was the result of the "virtual IMF policy" being followed before the change of policy course in September 1998. This suggests that the policy stance was not so different from the policies being followed by countries with formal IMF programs before September, and raises a question about the appropriateness of the time-shifted counterfactual in the Rodrik account.

Peter Garber noted that the controls were strongly signaled before being implemented. This allowed many investors to settle their positions, and caused the currency to appreciate. Garber added that the controls were an "exit tax" on investors with long positions.

Nor Mohamed Yakop said that traditional thinking said you cannot control interest rates and exchange rates, but Malaysia showed that you could. He reiterated that he thinks what Malaysia did is "not capital controls." Apart from limits on short selling there are no controls on asset flows, he said.

Caroline Atkinson said that the Malaysian capital controls don't provide general lessons. And she doubts they helped Malaysia. They did not lead to a more appreciated exchange rate, and FDI has fallen. She said that the authorities did not realize the environment was changing for the better. If they had tried six months earlier the costs would be clearer, and she thinks we would be having a different debate now.

Dani Rodrik said the main pro argument for capital controls is not that they allow the maintenance of an undervalued exchange rate. The main benefit of controls is that they allow the government to deal with a situation of panic. Looking at the period one year after the imposition of controls, we see that the largest contributor to Malaysia's differential performance has been consumption, not exports. This suggests the main benefit has not come through under-valuation. The stability of the exchange rate has been more important than the level, allowing for a longer planning horizon and for reduced uncertainty.

Yung Chul Park said that the combination of capital controls and a fixed exchange rate has been a success. But he has not heard from the Malaysian officials why they were successful. Expressing some exasperation, he added that he now hears that Malaysia didn't really have capital controls after all.


Session 4. Aftermath: 1999-2000 And Current Outlook

Chair: Dani Rodrik

Panelists: Yusuke Horiguchi, International Monetary Fund

Angus Armstrong, MIT; formerly Deutsche Bank

David Malpass, Bear Stearns Co. Inc.

Peter Garber, DeutscheBank Morgan Grenfell

Notwithstanding Malaysia's strong recovery, Yusuke Horiguchi said the country faces the challenges of protecting itself better in the future and regaining its position as one of the world's fastest growing economies. The past approach involved government-directed bank lending with an exchange rate peg to facilitate investment in the export sector. The strategy was successful in generating high growth, but the quality of investment diminished over time. In the future, the government must rely more on markets--deregulating competition, reducing the protection of domestic firms, and ensuring a better environment for FDI.

To reduce its vulnerability, Malaysia needs banking and corporate sector reform and a sound external policy (including a stable exchange rate regime). The banking and corporate sector reform is moving ahead, but a lot remains to be done. The restructuring agency cannot be allowed to become a warehouse for bad loans. He also complained that the banking and corporate sector reforms have not always been applied in an even-handed way, and questioned whether the government is truly committed to market discipline. He added that as long as well-connected firms continue to get preferential access to credit, the health of the financial sector will continue to deteriorate. Turning to the external front, Horiguchi noted that levels of external debt are well-regulated and the country has substantial international reserves. Although reserves are adequate at present, he said that they have declined despite large current account surpluses. This implies that capital outflows have increased, raising questions about the sustainability of the exchange rate peg. He added that there are few precedents for achieving a smooth exit from a pegged exchange rate regime, and he advised the government to exit from a position of strength, noting that the current position is weaker than it was six months ago though still relatively strong. There is little argument for waiting much longer, he concluded.

Angus Armstrong observed that Malaysia's star has slipped. He said people focused too much on the high GDP growth numbers and neglect of key issues like corporate finance. There are still deep problems with corporate structures and policies such as renationalizations and government interference have not dealt with the root of the problem. Turning to the fiscal accounts, he said that Malaysia is already running a primary deficit even with recent growth above trend and interest costs from bank re-capitalization. On the external side, he said things now look good but once one has in sight the end of monthly surpluses, foreign currency reserves will not last long. In fact reserves are already falling which shows net capital outflows as investors are discouraged from committing funds to Malaysia. So he thinks that the pegged exchange rate should be changed before reserves are wasted on an unnecessary defense. As an alternative he favored a managed float which would be preferable to inflation targeting which was unnecessary.

David Malpass said that comparing overall output performance from 1996 to 2000 Malaysia has done better than Thailand or Indonesia. Their strategy has been primarily an exchange rate peg, supported, at least initially, by capital controls. They need to engage in some damage control with foreign investors as a result of the capital controls and the labeling of investors as harmful speculators. He said that he does not agree with the IMF approach that a crisis is the best time to push through structural reform. Malaysia is in a better position now, having stabilized somewhat. He advised that those pushing countries to adopt a managed float with inflation targeting note that Asian countries are doing this with capital controls while Latin American countries are doing it with high interest rates. In neither case is inflation targeting working very well from a growth standpoint.

Peter Garber pointed out that FDI has fallen to roughly half the pre-crisis amount since the controls were implemented. Portfolio flows have also fallen off, with the result that growth is financed with internal saving. The boom in the US has also helped to support growth, particularly in the electronics sector. But now Malaysia is competing with Thailand and Malaysia in these sectors, and it is entering the first business cycle downturn of the new economy (and possibly even the end of the higher growth trend). Garber noted that Malaysian exports had fallen recently. This will put downward pressure on growth, which in turn will increase pressure for an expansionary policy. The result would be pressure on the exchange rate, and, with capital controls hovering in the background, a reluctance on the part of investors to put or keep money in Malaysia. Thus Garber predicts that soon there will be some combination of devaluation and capital controls. This he said is the legacy of controls--the belief that they will be imposed again. Finally, he noted that Malaysia still has a good debt rating. His concern, however, is not with default, but with the sustainability of the exchange rate.

General Discussion

Carmen Reinhart said that income elasticities for imports into the US are particularly high for Malaysia. Thus Malaysia would be hurt by a US downturn. She also said that in the pre-crisis years many countries had what were officially called managed floats, but their exchange rates were essentially pegged to the dollar. Labels can be changed, but that does not necessarily change the underlying policy. What many countries have now, she said, amount to crawling pegs though they are labeled as managed floats.

Andrew Warner asked the panel what they look at when assessing overvaluation. He said that the answer has important implications for the appropriate exit strategy from an exchange rate peg. Yusuke Horiguchi said that he looks at reserve losses. David Malpass said that the traditional methods -- purchasing power parity calculations and the historical movement of the exchange rate - are of limited help in deciding what to do. He noted that there is a lot of leeway in deciding what the correct exchange rate is.

Simon Johnson said that the main structural weakness in Malaysia is the corporate finance arrangements of politically connected firms, and added that the misallocation of capital got worse with the imposition of capital controls.

Dani Rodrik asked to what extent the success of the Malaysian program is tied to its pegged exchange rate.

Wing Thye Woo said that an important issue is the long-term impact of the capital controls on FDI flows. A lot of the FDI that went to South East Asia reflected the uncertainty that China faced with annual renewals of MFN. He said that FDI into China is now picking up, and added that India, with its English speaking workforce, provides a powerful competitor for investment flows. However, Woo thinks the historical evidence shows that "after default money does return." So if FDI is below trend, as Garber reports, it is because of diversions due to increased competition and not because of the controls.

Lee San Yin reported that businesspeople are happy with the peg. This is the message the government is getting. He questioned the assertion that FDI is not coming in, and noted that a number of major investments had taken place. Lin also said that corporate structuring had to be looked at "in perspective." There is a lot to do, but it takes time, and he is "optimistic that it will get done." Overall, he thinks the controls worked well, and that it is better not to dismantle completely them until there is a better understanding of what is happening in the US economy.

Edwin Truman said that the issues of the third and fourth panels are closely related. He said capital controls are often the "third best policy." After the first best policies the second best approach is to "clean up your problems when you get into trouble." The third best policy is to shut yourself off from the world. Malaysia is now between the second and third best responses. The controls are not designed to shield Malaysia from the world, but rather are a means of buying time to straighten things out. The big question, he said, is whether the next move will be towards orthodoxy or towards greater restriction.

Yung Chul Park asked about the alternative of inflation targeting. He said that such targeting amounts to wage targeting when there is a large traded sector, and wondered if this is a good idea given the relationship between wages and unemployment. Addressing those who suggest moving to a managed float without inflation targeting, he asked what the nominal anchor would be.

Kathryn Dominguez said that many countries have exchange rate pegs no matter what they say. But she said there is an important difference between an explicit and an implicit target under a managed float in terms of their vulnerability to speculators. Managed floats with an implicit peg are more feasible.

Nor Mohamed Yakop asked what Malaysia should do now. He noted that not many people are complaining about shutting down the offshore market. This is because the only ones being hurt are the speculators. On exchange rate policy, he said that Malaysia pegged its exchange rate to ensure a period of stability to push through reforms. Yakop believes that more time is needed; although the reforms could be moving faster. He also pointed out that there is a lot of volatility in a floating rate system, and wondered why Malaysia would move in this direction when few in the business community are complaining at present. Malaysia likes exchange rate stability. And monetary independence as well. On capital controls, he reported that the authorities would like to get rid of them, especially given the difficulty of distinguishing between long-term and short-term flows. Malaysia likes open capital markets, but it is easier to maintain the pegged exchange rate when there are limits on short selling in the offshore market. Finally, he said that even with limits on short selling there are other factors that move the exchange rate. The result of a move to a floating rate, he concluded, would be a very volatile exchange rate. Ted Truman responded that what we have just heard suggests the country may have another crisis within a few years.

Jomo Kwame Sundaram said he was pleasantly surprised that no one had mentioned Malaysia's lack of policy credibility. There is, he said, a lot of uncertainty about policy continuity. On the costs of the pegged exchange rate, he noted that there is little knowledge of what the true costs are, and without firmer evidence the government cannot be convinced that the costs are high. On foreign direct investment, Jomo said that the flows have decreased since 1996. There is a lack of credibility here also, with the minister in charge claiming that Malaysia is becoming more selective. He added that much new FDI is in the form of acquisitions, and he questioned how valuable this is in terms of replacing bad managers, etc. Lastly, he said that in the past it was believed that government policies had ushered in the East Asian miracle. What is being done now, he asked, to generate the next stage of growth?

Peter Garber responded to the claim that closing down the offshore market does not really hurt any one. This is true, he said, but it does change the perception of the risk of investing in Malaysia. He added that investors are also put off because they find it harder to hedge their risk without the offshore market.

David Malpass rejected Malaysia's apparent claim that it can violate the "international trilemma" by simultaneously controlling the exchange rate, controlling monetary policy, and maintaining an open capital account. He said that all countries have to figure out the "least costly" method of managing their exchange rate. But the international community's views on best practice affects the cost of any particular strategy through the resulting interest rate spreads. He noted that the IMF was at that time supporting different regimes, for example a pegged regime in Turkey [subsequently abandoned] and a currency board in Argentina.

Angus Armstrong observed that it is difficult to determine the "right level" of exchange rate for such an open economy for all time. He added that fixed rates do not bear the primary responsibility for the problems experienced by a number of Asian economies. Weakness in corporate governance and inadequate banking systems were more important. As a rule of thumb, he said weak banks lead to weak currencies, leading him to conclude it is better to concentrate on fixing the banks.

The final word went to Yusuke Horiguchi, who stated that the policy requirements for sustaining an exchange rate peg are too daunting for most emerging market economies. But countries must have a nominal anchor. He suggests inflation targeting, but warns you "can't turn it on and off."