Following the Mexican crisis, and more intensively after the Asian and Russian crises, the IMF entered a period of far-reaching reform, which is still continuing. The reforms aim to help us prevent crises where we can and mitigate their effects where we cannot.
In its crisis prevention efforts, the IMF has responded to the enhanced role of the capital account in financial crises by taking steps in three areas:
Surveillance and the financial sector. The IMF has sharpened its scrutiny of national policies and international markets, focusing in particular on developments that can leave countries vulnerable to crisis. Of all the changes that have taken place in the IMF in recent years, the increase in two-way transparency between the IMF and the outside world is the most significant. Transparency not only helps ensure better-informed citizens and investors, but also encourages policymakers to strengthen their policies and institutions.
The structure of external debt and financial sector weaknesses played key roles in exacerbating capital account crises. We have significantly stepped up our surveillance of external vulnerabilities, including through scrutiny of the national balance sheet, particularly external debt and reserves, and the strength of financial systems, most notably through the Financial Sector Assessment Program. IMF surveillance is also being given greater focus and structure by the development of international standards and codes of conduct, monitored by the IMF or other relevant bodies.
The new International Capital Markets (ICM) Department is evidence of the increased attention the IMF is paying to surveillance over international capital flows. Our interactions with the capital markets have also been enhanced through the creation of the Capital Markets Consultative Group.
Exchange rate and capital account regimes. Every major financial crisis since Mexico’s in 1994 has in some way involved a fixed or a pegged exchange rate regime. Having said this, floating exchange rates are not sufficient to prevent crises, and I do not imply that policymakers can or should be indifferent to the exchange rate or necessarily refrain totally from intervention in the foreign exchange markets. If a country decides to float, it must decide on the monetary policy it will follow. Inflation targeting, which many recent converts have chosen, seems to be working well and has much to commend it.
On capital account regimes, it is surely no coincidence that the most advanced economies all have open capital accounts. The IMF has cautiously supported the use of market-based controls on inflows in some emerging markets. The controls seem to have been successful for a time in allowing some monetary policy independence and in shifting the composition of capital inflows away from short-term debt. But empirical evidence suggests that they lose their effectiveness over time.
Capital controls may at times appear attractive, especially during a crisis, but it is telling that even among the countries worst affected by the crises, almost all have resisted the temptation to close themselves off. When push comes to shove, policymakers have abandoned fixed exchange rates before capital mobility—and they have been wise to do so.
Foreign reserves and Contingent Credit Lines. The growth of international capital flows has prompted a rethinking of the way we assess the adequacy of a country’s reserves. In an era when crises are more likely to arise from the capital account than the current account, it makes more sense to argue that countries with open capital accounts need reserves sufficient to cover their short-term debt rather than three or four months of imports. In the Asian crisis, countries with very large reserves generally did better in avoiding the worst of the crisis than those with smaller reserves.
A potential source of supplementary reserves is the IMF’s recently enhanced Contingent Credit Line Facility, which offers precautionary credit lines to countries with demonstrably sound policies that nonetheless feel threatened by contagion.
I will describe three recent changes in the IMF’s response to crises:
The SRF. The SRF was introduced at the end of 1997 to enable the IMF to respond better to capital account-driven crises. Reflecting the potentially large need for financing, there are no formal limits on access to SRF resources. Reflecting the lender of last resort doctrine, SRF loans carry a significantly higher charge than normal Stand-By Arrangements (with the interest charged rising the longer the loan is outstanding). And, reflecting the likelihood that confidence can be restored relatively quickly, SRF loans also have a shorter repayment period than the normal Stand-By Arrangement.
Narrowing the scope of conditionality. Notwithstanding the importance of structural policies in the countries to which we lend, there is general agreement that our conditionality in this area has sometimes been too extensive and restrictive. Broader programs are more difficult for the IMF to monitor and for the member country to implement. Excessive conditionality may also undermine a country’s ownership of the program. We are seeking to focus our conditionality on the IMF’s key areas of responsibility: monetary, fiscal, and exchange rate policies, and the financial sector, and on those structural measures that are critical to achieving the macroeconomic objectives of the program.
Private sector involvement. This issue remains one of the thorniest on the IMF’s agenda. The goal is to ensure that the private sector contributes to the resolution of financial crises by providing financing, rather than exacerbating the crisis by seeking to withdraw funds at the same time as the public sector is injecting them. More graphically, the argument is that the public sector should not bail out the private sector. The IMF framework seeks to rely on the catalytic approach and voluntary private sector involvement in cases of normal access to IMF resources and in cases of exceptional access where the crisis is likely to be quickly reversed because it is fundamentally one of liquidity. Where the balance of payments outlook over the medium term is not viable without debt restructuring, a coordinated approach will have to be made to ensure private sector involvement. In certain extreme cases, a temporary payments suspension or standstill may be unavoidable.
The crisis would have been far less virulent if exchange rates had been floating for some years before 1997. But the changes would only have made a difference if it had been possible to persuade governments to act in time.
Revisiting the Asian crisis
What difference might all these changes have made to the course of the Asian financial crisis if they had been in place in 1996? We cannot know, but let me speculate.
First, the exchange rate regime. If exchange rates had been flexible, Asian currencies would have appreciated less in late 1996 than they did as the dollar appreciated. The balance of payments difficulties faced in early 1997 by the countries that later went into crisis would therefore have been mitigated. Further, with more flexible exchange rates, short-term capital inflows—and the subsequent outflows—would probably have been smaller. And neither Thailand nor Korea would have used up essentially all their foreign exchange reserves to defend the exchange rate.
Second, if there had been more transparency, with reserves data meeting the standards of the Special Data Dissemination Standard, far more would have been known about the state of the foreign reserves in the crisis countries. This, too, could have prevented countries using up all their reserves and would have forced earlier action on the exchange rate. Further, more would have been known about the composition of external debts. That should have cut off capital inflows earlier and therefore reduced the disruption caused by subsequent outflows.
Third, if more attention had been focused on the health of the financial system, action could have been taken earlier to begin strengthening weak institutions, avoiding or reducing the extent of liquidity support that was extended in all the crisis countries. For instance, it would have been possible in Indonesia to begin dealing with weak banks before the crisis began and to have been better prepared with bank resolution mechanisms. It would also have been possible to spot the deterioration of corporate finances, and it might have been possible to begin dealing earlier with the disastrous interaction of corporate financial and banking sector weaknesses.
Fourth, if IMF surveillance had been more vigilant, and more attuned to market developments, it is possible that the financial attack on Korea would have rung the alarm bells a month or two earlier than November 1997, and perhaps the Korean government could have been persuaded to let the exchange rate float earlier. In Thailand, we would probably have struck a different balance in informing markets about our increasingly urgent dialogue with the authorities there in the runup to the crisis.
Perhaps all this means that it would have been possible to advance adjustment, thereby mitigating if not preventing the crisis, if the crisis prevention measures taken since the crisis had been in place in 1996. I do believe that the crisis would have been far less virulent if exchange rates had been floating for some years before 1997. But the changes would only have made a difference if it had been possible to persuade governments to act in time. It is not clear that would have happened. After all, many of the warning signs were around before the crisis: some of them were brought to the attention of governments, and others must have been known to governments (including central banks) even though they were not made public.
Why did governments not act? That is hard to know. Some may have been lulled into a false sense of security by many years of success. They had surely heard many earlier warnings of disasters, which had failed to materialize. Some of the governments were politically weak. Some of them were caught up in the familiar syndrome in which an exchange rate peg takes on a political significance that transcends its economic importance. And some of them must have thought that any action they might take would only precipitate the crisis they were trying to avoid.
The interactions between politics and economics in any economic crisis are illustrated by the striking fact that in all three IMF-supported programs in Asia in 1997–98, the economy began to turn only when new governments came into power: first in Thailand, then in Korea, and last, and with the most disruption, in Indonesia in May 1998.
Next, what would have happened had the crises played out exactly as they did to the point where countries turned to the IMF for assistance, but the subsequent changes in methods of crisis response had then been in place. First, conditionality would have been more focused—on macroeconomics, on the exchange rate, and on financial sector restructuring. On the macroeco-nomic side, we would probably not have asked for as much fiscal tightening as we did initially—though that request was soon reversed in the Thai and other cases.
On the structural side, we would certainly still have urged rapid progress on financial sector restructuring. And we would have worked with greater urgency with the World Bank to try to ensure rapid progress on corporate restructuring. That is to say that the core of structural conditionality in IMF-supported programs would not have been very different; but some other desirable, but not critical, policy changes would probably have been omitted from the programs.
Quite likely, we would have sought to involve the private sector more rapidly in all cases—that would have saved a few weeks in Korea and a few months and some distress in Indonesia and Thailand—and we would have had debt-monitoring systems ready. Possibly, had the SRF then existed, we would have suggested making it available to Thailand. All this means that capital outflows might have been better contained than they were, and exchange rates would likely have stabilized even more quickly than they did.
In brief, with today’s methods of crisis response, the crisis would have been better handled. But there is again a “but.” The core structural measures in the Asian crisis countries involved financial and corporate sector restructuring. Those are still the main items on the policy agenda, and it is proving very difficult for the affected governments to implement those reforms thoroughly. Thus, it is unlikely that conditionality would have been better implemented even had it been focused on these key measures. Nor are the governments of the crisis countries unique in their reluctance to move speedily on financial sector reforms, for governments all over the world are slow to undertake such politically difficult structural changes, be they in Japan at present, or Russia over the past few years, or the United States during the savings and loan crisis of the 1980s. Nonetheless, it remains urgent to move determinedly ahead on these fronts.
Asia and the IMF: looking ahead
In the wake of the crisis, Asian countries are moving to put in place new regional arrangements, including various swaps, and are considering alternative currency arrangements. The management of the IMF welcomes this enhanced regional cooperation, which should be complementary to more global arrangements, such as the IMF. We see potential advantages in regional currency cooperation, even though it will likely take many years—perhaps as many as it took in Europe—for such arrangements to come to fruition.
The IMF stands ready to cooperate fully in helping make these regional arrangements more effective—for instance, by assisting in the surveillance process and by cooperating in the financing arrangements envisaged under ASEAN [Association of South East Asian Nations] + 3 [China, Japan, Korea], where the activation of loans beyond 10 percent of the agreed lines will take place in the context of IMF-supported programs. And, to be sure, the lessons learned from recent crises will be applied when these programs are negotiated.
More also needs to be done to enhance the role of Asian countries in the working of the IMF. The under-representation of Asia in the IMF’s quotas is indeed a serious problem—recently somewhat mitigated by increasing China’s quota to reflect the return of Hong Kong SAR. But votes are rarely taken in the Executive Board of the IMF, and the effectiveness of Executive Directors is more related to their persuasiveness than to the size of their vote. Asia has 5 out of the 24 seats in the Board (Japan, China, and constituencies headed respectively by India, Indonesia, and Australia/Korea). It is important, if Asia’s voice is to be heard, that these positions be occupied consistently by the highest quality candidates, and that the countries they represent take a lively interest in the matters discussed in the Board and seek to develop independent policy positions on the key issues. Japan, our second largest member, always plays an important role in guiding the institution.
Since the Executive Board far prefers making decisions by consensus than by voting power, those directors who have coherent positions and the ability to advance them will usually see their views reflected in IMF positions on the issues.
It is also important that the IMF appoint more staff members from currently underrepresented Asian countries. This appears to be a matter more of supply than of demand, for the IMF is always on the lookout for qualified candidates for its staff. Many countries in other regions, and some in Asia, see an appointment to the staff of the IMF for some years as an important way to build up needed human capital about the international financial system inside their governments.
The IMF as a global institution is incomplete if Asia is not playing a full role. I believe Asia needs the IMF if it is to continue to benefit, as it has so spectacularly over the years from its integration into the global economy. For one thing, there is much unfinished business left over from the crisis, including how best to reduce the volatility of international capital flows—an issue on which we have to make further progress. Moreover, while we have implicitly been concentrating on emerging market countries, we also need to ensure that countries in Asia and elsewhere that have not so far been able to enjoy the benefits of integration into the global economy can do so.
To sum up: there is much for the IMF to do in Asia, and much for Asian countries to do inside the IMF.