email icon Email this citation

CIAO DATE: 12/99

Independent Task Force Report: Safeguarding Prosperity in a Global Financial System: The Future International Financial Architecture


Carla A. Hills and Peter G. Peterson, Co-Chairs
Morris Goldstein, Project Director

October 1999

Council on Foreign Relations

 

Table of Contents

Foreword

Acknowledgments

Executive Summary

I.Introduction

II.Why the International Financial Architecture Matters, Including to the United States

III.The Roots of Financial Crises and Weaknesses in the Existing Architecture

IV.Recommendations

V.Concluding Remarks: Moderate versus Radical Reform Plans

Dissenting Views

Members of the Task Force

 

 

Foreword

No event of the past 50 years has generated more calls for a reexamination of the institutions, structures, and policies aimed at crisis prevention and resolution than the Asian/global financial crisis that began in Thailand in July 1997. In September 1998, following a speech he delivered at the Council on Foreign Relations, President Clinton underscored this theme when he suggested that it would be worthwhile to convene a distinguished private-sector group to take a fresh look at the need for reform of the international financial architecture.

The Council was therefore enthusiastic about sponsoring this Independent Task Force on the Future International Financial Architecture. We were fortunate that Peter G. Peterson, chairman of both the Council and the Blackstone Group and secretary of commerce during the Nixon administration, and Carla A. Hills, CEO of Hills & Co. and US Trade Representative during the Bush administration, agreed to serve as co-chairs. We chose Morris Goldstein, a widely respected former deputy director of research at the IMF and now a senior fellow at the Institute for International Economics, to be the project director and to author the report. We also invited a stellar group of economists, bankers and financial experts, industrialists and labor leaders, political scientists, strategists, and regional specialists to join the task force. Suffice it to say that it would be difficult to assemble a group that could match for breadth and depth of experience on international financial policies the membership of this task force. The Council wishes to thank them all for their time and contributions.

The task force met regularly from January through June 1999. The first set of meetings focused on what was “broken” in the existing architecture, and the last set on how to “fix” it. Both moderate and more radical reform proposals were considered. In addition to its internal debates, the Task Force benefited from discussions with current and former economic policymakers. In this connection, the task force is especially indebted to Michel Camdessus, Andrew Crockett, Stanley Fischer, Tim Geitner, Alan Greenspan, William McDonough, Robert Rubin, George Shultz, and Larry Summers for sharing their views on the architecture. Likewise, the task force appreciates the valuable reactions and suggestions it received last April in a meeting with a group of central bank governors and finance ministers from a set of larger emerging economies and industrial countries.

In this final report, the task force argues forcefully that despite the sorry track record on banking, currency, and debt crises of the past twenty years, it would be a counsel of despair to conclude that little can be done to make crises less frequent and less severe. With the US economy now connected much more closely to the rest of the world than it was two or three decades ago, a strengthening of the international financial architecture is also very much in our national interest. The US economy performed impressively throughout the latest crisis because domestic spending was strong and inflation was low. Next time, we may not be so well positioned to weather the storm.

The task force favors a market-oriented approach to reform that would create greater incentives for borrowing countries to strengthen their crisis prevention efforts and for their private creditors to assume their fair share of the burden associated with resolving crises. This would place the primary responsibility for crisis avoidance and resolution in emerging economies back where it belongs: on emerging economies themselves and on their private creditors, which dominate today’s international capital markets.

Notwithstanding some dissents on specific findings and proposals, all 29 members of the task force endorse the broad thrust of this report. Seven key recommendations were able to command majority support:

  1. Greater rewards for joining the “good housekeeping club.” The IMF should lend on more favorable terms to countries that take effective steps to reduce their crisis vulnerability and should publish an assessment of these steps so the market can take note.

  2. Capital flows—avoiding too much of a good thing. Emerging economies with fragile financial systems should take transparent and nondiscriminatory tax measures to discourage short-term capital inflows and encourage less crisis-prone, longer-term ones, like foreign direct investment.

  3. The private sector: promote fair burden-sharing and market discipline. All countries should include “collective action clauses” in their sovereign bond contracts. In extreme cases where rescheduling of private debt is necessary, the IMF should provide financial support only if debtor countries are engaged in “good faith” rescheduling discussions with their private creditors, and it should be prepared to support a temporary halt in debt payments. The IMF should also encourage emerging economies to implement a deposit insurance system that places the main cost of bank failures on shareholders and on large, uninsured private creditors—not on small depositors or taxpayers.

  4. Just say no to pegged exchange rates. The IMF and the Group of Seven leading industrial countries should advise emerging economies against adopting pegged exchange rates and should not provide funds to support unsustainable currency pegs.

  5. IMF crisis lending: less will do more. For country crises, the IMF should adhere consistently to normal lending limits and should abandon huge rescue packages. For systemic crises that threaten the international monetary system, the IMF should turn to its existing credit lines when problems are largely of the country’s making and to special contagion funds when the country is an innocent victim.

  6. Refocus the IMF and the World Bank: back to basics. The IMF should focus on monetary, fiscal, exchange rate, and financial sector policies, not on longer-term structural reforms. The World Bank should focus on longer-term structural and social aspects of development, not on crisis management or macroeconomic advice.

  7. Generate political support for and ownership of financial reforms. A global conference of finance ministers should convene to reach a consensus on priorities and timetables for specific actions that countries will take to strengthen national financial systems.

The task force’s reform agenda is more ambitious than that being pursued by policymakers at present. It is tougher in the measures it proposes to reduce moral hazard and to induce private creditors to accept their fair share of the burden of crisis resolution. It is clear on the need for the IMF to return to more modest rescue packages for country crises and to activate very large rescue packages only in systemic cases with the agreement of a supermajority of creditor countries. It is stronger in its opposition to pegged exchange rates and more forthright in proposing tax measures to shift the composition of capital inflows to longer-term, less crisis-prone elements. It is more activist in urging the IMF to identify publicly which countries are and are not meeting international financial standards. It asks more of the major industrial countries in leading the way toward certain institutional reforms in capital markets. It calls for a stricter demarcation of responsibilities and leaner agendas for the IMF and the World Bank. And it suggests a vehicle for garnering political support and for regaining the momentum toward architectural reform.

As the Council forwards this report, we hope that it will contribute to the ongoing debate on how best to strengthen the international financial architecture. The more successful we are in that endeavor, the better are our chances of safeguarding America’s jobs, savings, and national security as well as of promoting global prosperity.

Leslie H. Gelb President Council on Foreign Relations

 

Acknowledgements

This final report represents a substantial amount of work and cooperation on the part of many individuals.

The task force’s co-chairs, Carla Hills and Pete Peterson, were instrumental both in putting this diverse and talented group together and in keeping the group focused on its objectives and timetable. They contributed valuable suggestions and ideas, which are reflected in the task force’s recommendations. They also made my job immeasurably easier by giving me enough room for maneuver to organize the task force’s deliberations and to draft the final report in the way I thought best, and by providing encouragement when it was most needed.

I am also indebted to the members of the task force. They couldn’t have been more helpful in sharing their views and ideas and in making suggestions on several previous drafts. While there was a lot of spirited debate during the task force meetings, that debate always took place in a constructive spirit (and with a lot of good humor). It was a pleasure to be a part of it.

I would also like to thank The Starr Foundation for its generous financial support of this project. At the Council, Les Gelb and his colleagues aided our work at every stage of the project. Special thanks go to Betsy Cohen, Tracey Dunn, David Jones, and April Wahlestedt. We could not have completed this report within such a short time frame without the wide-ranging and consistent support of Les and his team.

Finally, a few words of appreciation to my home institution, the Institute for International Economics. There, I want to thank Director C. Fred Bergsten for granting me leave from my normal Institute responsibilities to serve on the Council’s task force, Kara Davis, John J. Guardiano, and Christine Flint for doing their usual first-rate job in getting the manuscript ready for publication, and Trond Augdal for superb research assistance.

Morris Goldstein Project Director

 

Executive Summary

Certain passages in the executive summary are italicized to highlight the task force’s main findings and recommendations.

Introduction

When Thailand was forced to devalue its currency in July 1997, no one could have foreseen the turmoil that would follow. Over the succeeding two years, financial crises swept through the developing world like a hurricane. Indonesia, South Korea, Malaysia, the Philippines, Hong Kong, Russia, and Brazil were among the hardest hit, but few developing countries emerged unscathed. In the crisis countries, currencies and equity prices plummeted, economic growth turned into recession, wealth evaporated, jobs were destroyed, and poverty and school dropout rates soared. Private capital flows to emerging economies nose-dived, while industrial countries saw their export markets shrink. Last fall, after Russia’s debt default and devaluation and the near collapse of a large hedge fund (Long Term Capital Management, LTCM), international financial markets seized up for nearly all high-risk borrowers, including those in the United States. Global growth slowed sharply. In some quarters, doubts arose about the market as the engine of prosperity. Confidence in the official institutions that manage financial crises was shaken. No wonder, then, that President Clinton, speaking before the Council on Foreign Relations a year ago, characterized the Asian/global crisis as “the greatest financial challenge facing the world in the last half century.”

Financial crises are nothing new. In the past 20 years alone, more than 125 countries have experienced at least one serious bout of banking problems. In more than half these episodes, a developing country’s entire banking system essentially became insolvent. And in more than a dozen cases, the cost of resolving the crisis was at least a tenth—and sometimes much more—of the crisis country’s annual national income. As bad as it was, the US savings and loan crisis of the late 1980s cost US taxpayers about 2-3 percent of our national income. The debt crisis of the 1980s cost Latin America a “lost decade” of economic growth. Ten members of the European Exchange Rate Mechanism were forced to devalue their currencies in 1992 and 1993, despite spending upwards of $150 billion to defend them. Mexico suffered its worst recession in six decades after the devaluation of the peso in 1994-95. And in the recent Asian crisis, economies accustomed to annual growth rates of 6-8 percent suffered severe depressions, with output falling 5 to 14 percent last year. In the past six months, a number of the crisis countries have returned to positive economic growth and the functioning of global financial markets has improved. But the global recovery is still in its early stages and remains fragile—not least because most of the underlying vulnerabilities have been only partly addressed.

We cannot eliminate banking, currency, and debt crises entirely, but it would be a counsel of despair to argue that little can be done to make them less frequent and less severe. Strengthening crisis prevention and management—that is, the international financial architecture ("the architecture” for short)—is also very much in our national interest. The US economy is connected much more closely to the rest of the world than it was 20 or 30 years ago. The average share of exports and imports in our national output now stands at about 15 percent—twice as high as in 1980 and three times as high as in 1960. Two-fifths of our exports go to developing countries. US firms active in global markets are more productive and more profitable than those that serve only domestic customers. Exporting firms pay their workers better and have expanded jobs faster than firms that do not export. More than $2.5 trillion of US savings is invested abroad. Borrowing costs, including the monthly payments US households make for their home mortgages, are lower because of our participation in international capital markets.

But why worry, some might ask. After all, the US economy has continued to perform impressively throughout the latest crisis period. So it has. But to conclude that fragilities in the international financial system are somebody else’s problem would be dangerously complacent. In the recent emerging-market crisis, US exports to the most affected areas fell 40 percent. The Asian crisis struck when domestic spending in the United States was robust and when inflationary pressures were low. This meant that our economic growth was able to withstand a big jump in the trade deficit and that the Federal Reserve had scope to calm the turbulence in global markets by cutting interest rates. Next time we might not be so well positioned to weather the storm.

We should also take note of events that did not happen but could have. Americans have more of their wealth invested in the stock market than they have in their homes. The Asian crisis could have acted as a catalyst for a significant stock market correction.

The United States is not immune to financial crises abroad. There have been enough losses, close calls, and “might-have-beens” over the past few decades to remind us that international capital markets—despite their important contribution to our standard of living—can at times be risky places. The more successful we are in reducing the frequency and severity of financial crises—including in emerging economies—the better are our chances of safeguarding America’s jobs, savings, and national security as well as of promoting global prosperity.

Our Approach

If we are to make real headway in improving crisis prevention and management in the developing world, we must put the primary responsibility back where it belongs: on emerging economies themselves and on their private creditors, which dominate today’s international capital markets. If the behavior of debtors and creditors does not change, the poor track record on financial crises will continue. But wishing for change will not make it happen. Better incentives—including the prospect of smaller and less frequent official bailouts—can facilitate desirable changes in lender and borrower behavior.

Six principles guided our analysis. We wanted to:

  1. Encourage emerging economies to intensify their crisis prevention efforts.
  2. Permit savings to flow to the countries and uses where they have the best return.
  3. Promote fair burden-sharing among private creditors, official debtors, and official creditors when a crisis does occur.
  4. Increase the role of market-based incentives in crisis prevention and resolution.
  5. Make reform of the architecture a two-way street, with the major industrial countries also doing their part.
  6. Refocus the mandates of the IMF and the World Bank on areas they are best equipped to address.

Consistent with these principles, we offer seven key recommendations:

Recommendation 1. Greater rewards for joining the “good housekeeping club.” The IMF should lend on more favorable terms to countries that take effective steps to reduce their crisis vulnerability and should publish assessments of these steps for each country so the market can take note.

Recommendation 2. Capital flows—avoiding too much of a good thing. Emerging economies with fragile financial systems should take transparent and nondiscriminatory tax measures to discourage short-term capital inflows and encourage less crisis-prone, longer-term ones, such as foreign direct investment.

Recommendation 3. The private sector: promote fair burden-sharing and market discipline. To encourage more orderly and timely rescheduling of private debt where it is needed, all countries should include “collective action clauses” in their sovereign bond contracts. In extreme cases where rescheduling of private debt is needed to restore a viable debt profile, the IMF should require as a condition for its own emergency assistance that debtors be engaged in “good faith” (serious and fair) discussions on debt rescheduling with their private creditors. The IMF should also be prepared to support a temporary halt in debt repayments. To reduce moral hazard at the national level, the IMF should encourage emerging economies to implement a deposit insurance system that places the primary cost of bank failures on bank shareholders and on large, uninsured private creditors of banks—and not on small depositors or taxpayers. 1

Recommendation 4. Just say no to pegged exchange rates. The IMF and the Group of Seven (G-7) should advise emerging economies against adopting pegged exchange rates and should not provide funds to support unsustainable pegs.

Recommendation 5. IMF crisis lending: less will do more. For country crises, the IMF should adhere consistently to normal lending limits. This will help to reduce moral hazard at the international level. For systemic crises, the IMF should turn to its existing credit lines when problems are largely of the country’s making and to special contagion funds when the country is an innocent victim.

Recommendation 6. Refocus the IMF and the World Bank: back to basics. The IMF should focus on monetary, fiscal, and exchange rate policies plus financial-sector surveillance and reform and stay out of longer-term structural reforms. The World Bank should focus on the longer-term structural and social aspects of development, including the design of social safety nets. It should stay out of crisis lending and management. Recommendation 7. Generate political support for and ownership of financial reforms. Convene a global conference of finance ministers to reach a consensus on actions, priorities, and timetables for actions nations will take to strengthen national financial systems.

The Reform Agenda

Recommendation 1. Greater Rewards for Joining the “Good Housekeeping Club”

Emerging-market economies have a key responsibility to keep their houses in order, and the international community can encourage them to do so by enlarging the rewards for good housekeeping.

“Good housekeeping” covers a range of economic policies and institutional reforms. It means pursuing sound macroeconomic policies, including the avoidance of large budget deficits. It means prudent debt management that does not permit liquid liabilities of the public and private sectors to get way ahead of their liquid assets and that discourages the buildup of large currency mismatches. It means not being complacent about large current account deficits and highly overvalued exchange rates. It means maintaining a strong and well-regulated banking and financial system that extends loans on the basis of their expected profitability and of the creditworthiness of the borrower, and that complies with international standards for good public disclosure of economic and financial data, for effective banking supervision, and for the proper functioning of securities markets. It means shunning heavy reliance on short-term borrowing and on longer-term debt contracts with options that allow the creditor to demand accelerated repayment if conditions worsen. And it means holding enough international reserves and arranging contingent credit lines so that there is enough liquidity on hand to cushion against unexpected adverse shocks.

Suffice it to say that many of these elements of good housekeeping were not in order in the run-up to recent crises. In Russia and Brazil, for example, large government deficits and heavy reliance on short-term government borrowing were at the heart of their vulnerability.

In the Asian crisis countries, imprudent debt management, weak domestic banking systems, and premature and poorly supervised financial liberalization took a heavy toll when the external environment soured. Encouraged by interest rates lower abroad than at home, by exchange rates that had been relatively stable with respect to the US dollar, and by a history of strong economic growth, banks and corporations in the crisis countries stepped up their short-term foreign borrowing in the 1990s, much of which had to be repaid in foreign currency. On the eve of the crisis, short-term external debt was larger than international reserves in several of the crisis countries, and corporations had very high debt-to-equity ratios. Banks and finance companies in these countries had lax lending and accounting standards. Their lending decisions were also compromised by heavy government interference and by high levels of “connected” lending (to bank managers and directors and their related businesses). Bank supervision was weak. Reflecting all this, borrowed funds were not invested wisely, with heavy concentrations in real estate, equities, and industries with low rates of return. Lenders (domestic and foreign) did not monitor borrowers carefully, perhaps because they expected that governments and international organizations would be willing and able to bail them out if borrowers ran into trouble

.

And run into trouble they did. Exports from Asia slowed dramatically in 1996, prompted by a steep decline in semiconductor prices and a loss of competitiveness as Asian currencies followed the US dollar up against the Japanese yen. Property prices fell, leading to a surge in nonperforming bank loans. As foreign lenders began to recognize that Thailand’s weaknesses were shared by several other Asian emerging economies, a panic ensued in which foreign shareholders, bondholders, and banks scrambled to get their money out. Cash flow problems mounted as interest rates rose in vain attempts to defend currencies pegged to the dollar. Political instabilities and uncertainties added to the problem. And when currencies fell sharply, this made it much more expensive for companies to repay their foreign currency loans. Soon everything collapsed.

Henceforth, the IMF should lend on more favorable terms to countries that take effective steps to reduce their vulnerability to crises. To increase the private market payoff for good crisis prevention, the IMF should make public a “standards report” in which it assesses periodically each member country’s compliance with international financial standards. It should also publish its regular assessments of each country’s economic policies and prospects (its Article IV reports). Loans to countries that make the extra crisis prevention effort should benefit from lower regulatory capital requirements for banks. Some initial, partial, and tentative steps in this general direction have already been made, but more should be done to strengthen the rewards for joining the “good housekeeping club.”

Recommendation 2. Capital Flows—Avoiding Too Much of a Good Thing

The freer flow of capital across national borders has been of considerable benefit to the world economy. It has loosened the constraints imposed by self-financing and improved the overall productivity of investment on a global scale. This finances development and raises living standards in borrowing countries while providing savers in lending countries with the opportunity to earn a better return on their money. It has permitted both borrowers and investors to obtain better diversification against shocks to their domestic economies. It has helped foster the transfer of best-practice production processes.

But experience indicates there are risks and costs along with the benefits. In recent years, private capital flows into emerging markets have been highly volatile. After mushrooming in the early 1990s, they reached a peak of $213 billion in 1996, before collapsing to just over $60 billion last year. This volatility shows up in price as well as quantity. During the 1990s, the interest rates paid on emerging-market bonds have fluctuated wildly in comparison with those paid on US bonds. For example, this interest rate spread was 1,200 basis points in January 1991; 400 points in January 1994; 1,600 points in January 1995; 400 points in mid-1997 (just before the Asian crisis began); 1,400 points in the fall of 1998 (just after Russia’s debt default); and 1,100 points in July 1999. 2

While some fluctuation in private capital flows to emerging economies is natural in light of changing investment opportunities and the way investors react to new information, experience suggests that “boom and bust” cycles in such flows create serious problems. When capital inflows are very large and occur at a pace that outstrips the domestic capacity to supervise the financial sector and to build a credit culture, they sow the seeds of subsequent banking crises. In several of the Asian crisis countries, the recent crisis was preceded by bouts of premature and poorly supervised financial liberalization. In Thailand, for example, the Bangkok International Banking Facility—although intended for another purpose—wound up serving as a conduit for local firms to vastly expand their loans from foreign banks, with unhappy results.

Turning to the “bust” side of the cycle, a sudden stop or reversal in private capital flows is often the forcing event that ushers in a crisis. This can turn liquidity problems into solvency problems, induce large cuts in spending that bring on recession, and spread crises considerably beyond their origin. For five Asian emerging economies heavily affected by the recent crisis, net private capital flows went from a net inflow of $65 billion in 1996 to an outflow of $43 billion in 1998; Japanese banks alone withdrew $21 billion from these five countries in 1997 (versus a $50 billion inflow the preceding year).

The challenge, therefore, is to find ways to moderate the boom-bust cycle in private capital flows and to tilt the composition of such flows toward longer-term, less crisis-prone components (such as foreign direct investment) while still preserving most of the benefits associated with greater market access.

Short-term capital flows carry a high risk because their short maturity makes it easier for investors to run at the first hint of trouble. Whereas net flows of foreign direct investment to emerging economies recorded only a slight decline during the recent Asian crisis, the fall in portfolio flows and even more so in bank loans was much more pronounced. Chile has addressed this problem by, in effect, taxing capital inflows if they are withdrawn after only a short time. This seems to have tilted the composition of its inflows toward the less risky, longer-term components. Admittedly, the effectiveness of such measures does tend to erode over time, and one side effect is that some desirable short-term flows—such as credits to support trade—can also be deterred. Nevertheless, if the alternative is a boom-bust cycle followed by a costly financial crisis, the choice seems clear.

The IMF should therefore advise those emerging economies with fragile domestic financial sectors to impose Chile-type holding-period taxes on short-term inflows until their ability to intermediate such flows is stronger. The measures should be transparent and designed not to impede the entry of foreign financial institutions, which can make a valuable contribution to strengthening domestic financial systems.

There are other things that can also be done to moderate the boom-bust cycle. For one, emerging economies should not impose controls or taxes on long-term inflows. One reason South Korea relied so heavily on short-term inflows was that it had controls that kept long-term flows out. For another, in revising the international agreement that specifies how much capital internationally active banks need to hold against various kinds of assets (the Basle Capital Accord), regulators should avoid weighting schemes, which provide incentives for short-maturity flows.

Hedge funds, which finance often very short-term investment decisions with huge sums of borrowed money, have a gained a certain notoriety in financial crisis episodes. But a review of these episodes—including the recent events in Asia—suggests that hedge funds are not the villains they are often made out to be. At the same time, given the threat highlighted by the near-collapse of LTCM last year, the official community is right to step up the “indirect” regulation of hedge funds by tightening risk-management guidelines for the banks and security houses that lend to them. Financial regulators should give this approach a fair trial. But if it does not produce results, they should consider going farther by imposing a higher regulatory capital charge (risk weight) for bank loans that go to offshore financial centers (where many hedge funds are located) that do not meet international financial standards.

Recommendation 3. The Private Sector: Promote Fair Burden-Sharing and Market Discipline

If a country faces an unsustainable burden of debt repayments, inevitably they will have to be rescheduled. It benefits neither debtor nor creditors if this takes a long time. But there are formidable institutional barriers to a quick resolution. There is no international bankruptcy code, and in many developing countries national bankruptcy laws either do not exist or function poorly. There are mechanisms through which debts payable to governments can be rescheduled cooperatively, but there is no equivalent for debts to the private sector.

If debt difficulties are resolved with large official bailouts—be it by national authorities or by the international financial institutions—then there will be problems of a different but equally serious nature. If market participants come to routinely expect such bailouts, then private creditors will have little incentive to monitor the financial condition of borrowers, too many resources will be channeled to the borrowers and lending categories viewed as implicitly “insured,” and taxpayers and legislatures in creditor countries may withdraw their support for such rescues because they are being asked to bear the consequences of poor lending and borrowing decisions by other parties. One reason it was so difficult to secure approval from Congress for an increase in the IMF’s financing last year was the perception that Wall Street—and particularly, large banks—was benefiting much more from official rescue packages than was Main Street.

These problems generally fall under the heading of “moral hazard.” As with other types of insurance, the appropriate responses to moral hazard are to limit the size of insurance payments and to charge risky policyholders more for the insurance—not to provide no insurance at all.

Problems with debt rescheduling and with moral hazard have become more pressing in recent years. The share of bonds in emerging-market financing has increased sharply while that for bank loans has declined. Bonds are “rescheduling unfriendly” compared to bank loans. For government bonds, one solution would be to include clauses in contracts to make it harder and less profitable for rogue creditors to impede a rescheduling; such “collective action clauses” already are included in syndicated bank loans. Successive reports by groups of industrial-country governments have in fact recommended such collective-action clauses. But it is unrealistic to expect emerging-market countries to take this step on their own if highly creditworthy industrial countries refuse to join in or to make it worthwhile to do so.

On the moral hazard front, significant difficulties exist at both the national and international levels. At the national level, the past several decades have witnessed rapid growth in the banking and financial sectors of emerging economies. Yet the vast majority of these countries do not have in place a good system of deposit insurance for their banks. Such a system should put the burden of resolving failed banks on shareholders and on large, uninsured creditors rather than on small depositors and taxpayers or on international institutions; it should place stringent accountability conditions on senior economic officials when they decide to rescue a bank because it is “too large to fail”; and it should give bank supervisors protection against strong political pressures to delay taking corrective actions. This is the kind of deposit insurance reform that was introduced in the United States after our savings and loan crisis. Without such reform, those most responsible for causing banking crises often get off the hook while others pay the tab.

At the international level, it is true that equity investors and bond holders experienced large losses in the Asian crisis, and banks took sizable hits from the Russian crisis. But too often, large rescue packages allow private creditors—particularly large commercial banks—to escape from bad lending decisions at relatively little cost. The $50 billion Mexican rescue package of February 1995 allowed holders of certain Mexican government securities (tesobonos) to get out whole. The international community committed about $190 billion in official rescue packages for Thailand, Indonesia, South Korea, Russia, and Brazil, one-third of which has so far been disbursed. The Miyazawa Plan has committed $30 billion more to Asian rescue packages. The Thai, South Korean, and Indonesian authorities issued broad guarantee announcements for bank depositors and creditors shortly after the outbreak of their crises, and the bulk of the rescheduling of the short-term debt of South Korean banks was done with a government guarantee.

One need look no further than private capital flows to Russia and the Ukraine in the run-up to the crisis—widely known on Wall Street as “the moral hazard play”—to see what happens when moral hazard effects become large. Despite serious underlying weaknesses in the economic fundamentals, investors were prepared to purchase large amounts of high-yielding government securities, presumably under the expectation that should conditions worsen, geopolitical and security concerns would prompt G-7 governments and the IMF to bail them out.

To be sure, when official rescue packages are evaluated, a balance must be struck between limiting systemic risk and encouraging market discipline. By providing emergency assistance to an illiquid but not insolvent borrower and thereby preventing a costly default and its possible spillover to other borrowers, an official crisis lender can limit the risk to the financial system as a whole. On the other hand, if such emergency assistance is too readily available, too large, and too cheap, lenders will not learn the lessons of their mistakes and market discipline will suffer. But in recent years that balance has tilted too far away from market discipline. Unless balance is restored, we will not be successful either in deterring future crises or in garnering popular support for official rescue packages.

To bring more order and timeliness to private debt rescheduling, all countries—including the G-7—should include collective-action clauses in their government bond contracts. The G-7 should match its words with actions by also requiring that all government bonds issued and traded in their markets include such clauses.

The IMF should encourage emerging economies to maintain comprehensive registers of their creditors. Creditors in turn should be encouraged to form standing committees that could help coordinate future debt reschedulings.

To reduce moral hazard at the national level, the IMF should advise emerging economies to enact sensible deposit insurance reform in their banking systems. At the international level, the IMF should provide emergency assistance only when there is a good prospect of resolving the applicant country’s underlying balance of payments and debt problems. In the extreme cases when this requires rescheduling of private debt, the Fund should make “good faith” discussions on such rescheduling a condition for its own assistance and should be prepared to support a temporary halt in debt repayments. No category of private debt (including bonds) should be exempt from such rescheduling, and it should be done in a way that does not discriminate between domestic and foreign creditors.

Recommendation 4. Just Say No to Pegged Exchange Rates

One of the most important steps an emerging economy can take to reduce the risk of a crisis is to get its exchange rate policy right. The events of the past two years have highlighted the risks of trying to defend a currency regime based on a publicly announced exchange rate target, and especially so for “adjustable peg” regimes (that is, a regime in which an emerging economy pegs its currency to the currency of a larger economy—usually the US dollar—with an option to adjust the peg when underlying conditions change). Thailand, Malaysia, the Philippines, Indonesia, Russia, and Brazil have all been forced to abandon announced exchange rate targets during the recent emerging-markets crisis. Among larger emerging economies with open capital markets, the list of those that have been able to maintain a fixed exchange rate for five years or more is now very short: Argentina and Hong Kong.

Pegged exchange rates have their attractions. They can be an effective instrument for reducing high inflation. But the potential risks of pegged exchange rates—particularly their vulnerability to crises—outweigh the benefits. Pegged rates become problematic when they become highly “overvalued.” This can happen either because the country’s inflation rate (even if it is much reduced from earlier periods) remains higher than that of its trading partners, or because the currency to which it is pegged is rising and is dragging it up against other currencies. In either case, a highly overvalued exchange rate translates into poor competitiveness, making the currency a target for speculators. But there is no easy way to exit gracefully from a pegged exchange rate: when the overvaluation is small, there is apt to be little political support for upsetting the applecart with a change in the pegged rate, and by the time the overvaluation has become large and obvious, it is often too late to avoid a crisis.

Once a country runs low on international reserves, the brunt of the defense of a pegged exchange rate falls on high domestic interest rates (to make assets denominated in its currency more attractive). But high interest rates slow economic growth and raise unemployment, make things worse for fragile banking systems, exacerbate the fiscal problems of governments with large fiscal deficits and lots of floating-rate debt, and add to the cash flow problems of highly leveraged corporations. Speculators, many of whom can finance very large positions in the foreign exchange market, know that there is a limit to how long countries can keep interest rates sky-high. In most of these battles, David and his sling (that is, his fixed exchange rate and high interest rates) have been crushed by Goliath (the international capital market), and it is not easy to see why this asymmetry would disappear over the foreseeable future.

In some recent crises (for example, Brazil and Russia in 1998-99, Mexico in 1994-95, and some member countries of the European Exchange Rate Mechanism in 1992-93), exchange rate overvaluations were sizable. In the case of the Asian crisis countries, their currencies appeared to be only modestly overvalued in mid-1997, but large current account deficits, sharply falling export growth, weak banking systems, and highly leveraged corporations made them vulnerable. In addition, because their exchange rates had been relatively stable for a long time, banks and corporations did not protect themselves against currency risk, and hence the consequences of large foreign currency exposure were that much more painful when large devaluations finally occurred.

None of this is to claim that other currency regimes are not without their own problems. Rather, it is simply to argue that an adjustable peg regime seems particularly crisis prone for emerging economies.

Despite the risks, history suggests that emerging economies will be tempted to defend an overvalued pegged exchange rate if IMF and G-7 funds are available to finance that defense. The IMF and the G-7 therefore should go beyond advising emerging economies not to adopt an adjustable peg regime. If asked to support an unsustainable peg, the Fund and the G-7 should “just say no.” The mainline currency recommendation for emerging economies should instead be one of “managed floating,” with the use of currency boards and single currencies reserved for particular situations.

Recommendation 5. IMF Crisis Lending: Less Will Do More

The International Monetary Fund was created to help countries tackle balance of payments problems without resorting to draconian austerity measures, beggar-thy-neighbor exchange rate policies, and trade barriers. This remains an extremely valuable goal. Indeed, as costly as the recent emerging-market crises have been, we would have seen deeper recessions, more competitive devaluations, more protectionism, and far more human suffering had there been no financial support from the IMF and from other official creditors.

But this does not mean the bigger the better. Rescue packages for country crises should be large enough to reduce the recessionary impact of the crisis, to finance some smoothing operations in currency markets, to contribute modestly to the cost of bank restructuring, and to provide a social safety net that provides some protection against the hardships of the crisis for the most vulnerable groups in the economy. In the overwhelming majority of cases, the Fund’s normal lending limits (100 percent of a country’s quota or IMF subscription annually and 300 percent of quota cumulatively) ought to be sufficient. Rescue packages should not be so large as to provide cover for holders of short-term external debt to escape the consequences of poor lending decisions—lest they generate the kind of moral hazard problems emphasized above. IMF loans to Mexico (1995), to Thailand and Indonesia (1997), and to Brazil (1999) were in the neighborhood of 500-700 percent of Fund quotas, while the loan to South Korea (1997) was 1,900 percent of its quota.

We are not persuaded that smaller rescue packages would necessarily make it more difficult for emerging economies to regain the “confidence” of investors, as experience suggests that this owes more to the speed and determination with which underlying economic problems are addressed. The expectation of smaller rescue packages may well reduce somewhat the flow of private external finance to emerging economies and increase somewhat its cost. But since interest rate spreads on emerging-market borrowing have been too low and the flow of capital to them too high during much of the 1990s, some moderate movement in the other direction would be no bad thing (especially for those emerging economies with relatively high domestic savings rates).

But what about rare situations of widespread cross-border “contagion” of financial crises where failure to intervene would threaten the performance of the world economy and where private capital markets are not distinguishing well between creditworthy and less creditworthy borrowers?

To counter such systemic threats effectively, the international community needs quick access to an adequately and securely funded international backup facility that can assist the victims of contagion. This would supplement the existing credit lines (the New and General Agreements to Borrow, or NAB/GAB) that the IMF already has available from a set of creditor countries. In April 1999 the IMF established a new lending window, the Contingency Credit Line (CCL), to offer assistance to well-behaved countries that feel threatened by contagion. But the CCL contains no new money and its operational guidelines seem unnecessarily complex.

We propose that the IMF return to normal lending limits (100-300 percent of quota) for country crises, that is, for crises that do not threaten the performance of the world economy. In the unusual case of a systemic crisis, the IMF should turn to its systemic backup facilities—either the existing NAB/GAB or a newly created “contagion facility” that would replace the existing Supplemental Reserve Facility (SRF) and the CCL. Activation of the systemic facilities would require a decision by a supermajority of creditors. The NAB/GAB would be used when the country’s problems are largely of their own making and an IMF program is needed to correct those problems. The contagion facility would be used for victims of contagion. Loans from the contagion facility would be agreed on expeditiously, would be disbursed quickly, would be free of policy conditions, and would be priced more expensively than normal loans from the Fund. This contagion facility would be funded by pooling a one-off allocation of Special Drawing Rights—the IMF’s artificial currency. 3 The US contribution to that contagion facility would be made after extensive consultation with the Congress.

Recommendation 6. Refocus the IMF and the World Bank: Back to Basics

The emerging-market crises have shaken public confidence in the Bretton Woods institutions. But their roles are crucial—the Fund as a key crisis lender and manager and also increasingly as a monitor of compliance with international financial standards, and the World Bank as a promoter of poverty reduction and sustainable economic development.

Calls for their abolition are misplaced. There is a moral hazard problem associated with large IMF-led financial rescues, but this can be reduced significantly by altering the IMF’s lending policies along the lines sketched above. Although hindsight reveals that the Fund’s monetary and fiscal policy recommendations to the Asian crisis countries were by no means flawless, these are best regarded as judgment calls in a difficult situation in which there were no easy solutions. For example, while an earlier move to lower interest rates would have helped counter the recession by reducing debt burdens and cash-flow vulnerabilities, it carried the risk of accelerating currency depreciation in a context in which banks and corporations had large, unhedged foreign-currency liabilities and confidence was already very weak.

But to reject the abolition of these institutions is not to deny that there is a need for reform of the Bretton Woods twins. Both the Fund and the Bank have tried to do too much in recent years, and they have lost sight of their respective strengths. They both need to return to basics.

We argue that the Fund should normally lend less and concentrate more on encouraging crisis prevention. It should also focus on a leaner agenda of monetary, fiscal, and exchange rate policies, and of banking and financial-sector surveillance and reform. It should leave more detailed structural reforms to the World Bank and other international organizations with the requisite expertise in those areas.

The World Bank, in turn, should focus on the longer-term structural and social aspects of economic development. It should not be involved in crisis lending or crisis management, and it should refrain from publicly second-guessing the Fund’s macroeconomic policy advice.

One area where the Bank can and should do more is in the design of social safety nets. When crises strike, the burden of economic adjustment usually falls hardest on those least able to cope. The recent crises in Asia have provided fresh affirmation of this danger, and the need to protect the poorest and most vulnerable would be even more pressing if official rescue packages became smaller in the future. While this report concentrates on the financial architecture and on the role of the IMF, it is well to remind ourselves that financial stability is not an end in itself but rather a means to broadly shared global prosperity and that it is a fantasy to believe that financial stability can be maintained without attention to the social aspects of development.

Recommendation 7. Generate Political Support for and Ownership of Reforms

Many of the cracks in the international financial system reflect weaknesses in national policies. Remedying these defects—be it by strengthening financial systems in emerging economies or by altering some institutional practices (for example, collective-action clauses in bond contracts) in industrial economies—might be unpopular and will on occasion demand that powerful vested interests be confronted. In addition, experience suggests that reform programs are most successful when the countries most affected participate directly in the design of those measures and when they take “ownership” of the reforms. If the emerging economies are not full partners in the reform exercise, it will not work.

Intensive discussions on strengthening the financial architecture have been under way for about five years (since the Mexican peso crisis of late 1994). With prospects for recovery from the global crisis brightening, there is a danger that “architecture fatigue” and complacency could combine to stall the push for reform—and this before most of the measures set out in this report could be implemented.

For all these reasons, governments will have to demonstrate considerable political will to carry the reform agenda through. To help foster the necessary political commitment, the international community should directly involve the nations whose behavior we wish to change. The IMF’s Interim Committee, the Financial Stability Forum, and the presidents of the regional development banks therefore should convene a special global meeting of finance ministers to establish priorities on architectural reform measures and to agree on a specific timetable for specific corrective steps.

Conclusion

Many of the themes emphasized in this report have also been part of the official sector’s plans and suggestions for the future architecture. Nevertheless, our approach differs from theirs in several important respects:

The recommendations outlined in the report are those that were able to command majority support within the task force. Many other proposals, however, were actively debated. A large group of task force members felt that there could be no serious reform of the architecture without fundamental reform of G-3 currency arrangements. They argued that the impact of the global economy on emerging economies is driven significantly by swings among the G-3 currencies, and that in recent years these swings have been enormous, volatile, and frequently unrelated to underlying economic fundamentals. They favored a system of rather broad target zones or reference ranges for the dollar, the euro, and the yen. Some other task force members favored stronger regulation over highly leveraged institutions; a global summit on architectural reform to be held by heads of state and governments; greater incentives for crisis prevention and greater use of early warning indicators of financial crises; a more structural approach to reform of the architecture; a link between core labor standards and international financial standards; a different approach to collective-action clauses in bond contracts, to taxes on capital inflows, and to private-sector burden-sharing; or new measures to encourage sound long-term lending to emerging economies.

We also discussed more radical alternatives. These included comprehensive controls on capital flows, the adoption of single currencies, more far-reaching reforms of the IMF (ranging from its abolition to the creation of a much larger and more powerful Fund), and the establishment of new, supranational regulatory institutions. In the end, the more radical proposals seemed either undesirable or impractical. We therefore opted instead for what we would characterize as “moderate plus” proposals: proposals that, taken together, would make a significant difference to crisis prevention and management but that would still have a reasonable chance of acceptance.

 

I. Introduction

Financial crises—banking crises, currency crises, debt crises, or some combination of the three—have occurred with disturbing frequency and intensity over the past 20 years. The Asian/global financial crisis that first erupted in Thailand in July 1997 is the most serious of these crises. Indeed, in his speech before the Council of Foreign Relations in September 1998, President Clinton characterized it as “the greatest financial challenge facing the world in the last half century.”

Financial crises can impose enormous costs and hardships on the countries involved. At their worst, these crises can destroy within the space of a year or two much of the economic progress that workers, savers, and businesses have achieved over several decades. They can also lead to greater questioning of the verdict of the marketplace and to a decline in popular support for the official institutions that are responsible for crisis management. It is therefore a matter of high priority for governments and the private sector alike to find ways both of reducing susceptibility to financial crises and of dealing with crises more effectively when and where they occur. Unless we can do better at reducing the number of serious accidents that take place on the superhighway of international finance, many—including those emerging economies that could potentially benefit most from using it—will be tempted to take an alternate route.

Governments from both the industrial countries and leading emerging economies, in concert with the international financial institutions (the IMF, the World Bank, the Bank for International Settlements, and others), have been hard at work on plans for improving crisis prevention and crisis management. 4 This collaborative international effort has come to be widely known as strengthening the “international financial architecture” (hereafter, the architecture).

In this report, the task force sets out its own assessment of the existing architecture and puts forth a package of recommendations for improving it.

We favor a market-oriented approach to reform that would create greater incentives for borrowing countries to strengthen their crisis prevention efforts and for private creditors to assume their fair share of the burden associated with resolving crises. This would place the primary responsibility for crisis avoidance and resolution in emerging economies back where we think it belongs—on emerging economies themselves and on their private creditors, which dominate today’s international capital markets.

We also think a greater effort should be made to distinguish “country crises” from multicountry “systemic crises,” and to treat the two differently. By withholding IMF financial support for overvalued fixed exchange rates and by making greater use of private debt rescheduling under appropriate circumstances, it should be possible for the IMF to become “smaller” in its emergency lending for country crises. In our view, an IMF that adhered consistently in its lending for country crises to normal access limits (100 percent of Fund quota on an annual basis and 300 percent of a quota cumulatively) would be more compatible with lenders and borrowers doing the right things on crisis prevention and crisis resolution than an IMF that engaged more frequently in very large loans (such as the loans for 500-700 percent of Fund quotas extended to Mexico, Thailand, and Indonesia). “Moral hazard” is by no means the only problem in the existing architecture but any sensible reform plan should contain recommendations for reducing it. 5

At the same time, it would be imprudent to assume that private market excesses and widespread cross-border contagion of crises can never happen again. And if it does happen, the international community needs to have the tools available to combat it. We therefore also support an adequately and securely funded international backup facility that could deal expeditiously with rare but truly “systemic” multicountry crises that threaten to undermine the performance of the world economy.

In framing our specific recommendations, the task force has been guided by six principles.

First, changes in the architecture should encourage national governments to intensify their own crisis prevention efforts—and should not serve as a substitute for, or as a means to delay, policy reform.

Second, it is essential to preserve the ability of international capital markets to channel savings to the places and uses where they have the highest long-term return. This is not inconsistent, however, with counseling those emerging economies that need them to take measures to limit the vulnerability associated with surges in short-term capital inflows.

Third, when financial crises occur, the burden involved in resolving the crisis needs be shared equitably among debtors, private creditors, and official creditors. Overborrowing does not occur without overlending. For market discipline to operate effectively, there must be an expectation that no class of private creditors or of debt instruments will be exempt from bearing the consequences of poor lending and investment decisions.

Fourth, a larger role for market-based incentives in crisis prevention and resolution does not mean that governments and international financial institutions will cease to play an important role in the future architecture. Quite the contrary. They can improve the infrastructure needed to make markets function better. They can police the integrity of markets against manipulation and other abuses. They can help to rein in excessive risk taking by enforcing appropriate prudential regulations. They can promote good behavior by promulgating and monitoring compliance with international codes and standards, and by rewarding countries who observe them. They can induce countries to respond to crises in ways that are not destructive of their own or their neighbors’ prosperity. And they can help to limit runs in financial markets when herd behavior or other problems prevent private-market participants from discriminating between creditworthy and less creditworthy borrowers.

Fifth, architectural reform must be a two-way street. The major industrial countries (including the United States)—no less than emerging economies—should be willing to make changes in their own financial markets and supervisory practices to encourage better crisis prevention and crisis management. In the case of certain institutional reforms, such as inclusion of “collective action clauses” in sovereign bond contracts, the chances of success will be much higher if the Group of Seven (G-7) countries lead the way. Similarly, stronger incentives for inducing creditors and investors in industrial countries to improve their risk assessment and management would contribute to moderating the “boom-bust” cycle in private capital flows to emerging economies.

And sixth, public institutions function best when they have a clear mandate, when they concentrate on their comparative advantage, and when they avoid duplication of each other’s efforts. Before we consider either abolishing the international financial institutions we have or creating new ones, we ought to try to refocus the mandates of the IMF and the World Bank to make them more compatible with the needs of today’s global economy. Both the Fund and the Bank have tried to do too much in recent years.

The remainder of the report is divided into four parts. In Section II, we explain why it is crucial to do better at preventing and managing financial crises and why the United States itself, despite its continued impressive overall economic performance since the outbreak of the Asian crisis, has a large stake in improving the future architecture. In Section III, we examine the factors that often give rise to financial crises and offer our assessment of what parts of the existing architecture are most in need of repair. In Section IV, we outline a package of seven interrelated recommendations for strengthening the future architecture. We also indicate where our recommendations differ from those put forward by the official sector. Finally, in Section V, we offer brief concluding remarks on the choice between moderate and radical reform of the architecture.

 

II. Why the International Financial Architecture Matters, Including to the United States

Crises Happen

In the past 20 years, more than 125 countries have experienced at least one serious bout of banking problems. In developing countries, there have been at least 70 cases where these banking problems were so extensive that the entire banking system essentially became insolvent. In more than a dozen of these episodes, the crisis was so deep that taxpayers had to spend 10 percent or more of the country’s total output (i.e., its gross domestic product, or GDP) to resolve the crisis. As bad as it was, the US savings and loan crisis of the late 1980s cost US taxpayers approximately 2-3 percent of our GDP, a figure that would not even make the “Misfortune 50”—the list of the 50 worst banking crises (relative to the size of the economy) of the past two decades. On top of this, we know that economic growth is typically much lower during banking crises than during normal times, and that a banking crisis increases the odds that a country will undergo a currency crisis as well.

The debt crisis of the 1980s had such a severe contractionary effect on economic growth in the heavily indebted developing countries that the 1980s is still frequently referred to in Latin America as “the lost decade.”

In 1992-93, the countries of the European Monetary System spent $150-200 billion on intervention in foreign exchange markets in an unsuccessful effort to stave off the devaluation of 10 European currencies. During its currency crisis of 1994-95, the Mexican economy contracted by 6 percent, its worst recession in six decades.

For the Asian emerging economies at the epicenter of this global crisis, the toll has been heavier yet. After the fall of the Thai baht (in early July of 1997), currencies and equity prices in the region plunged by 30-75 percent in the first six months. Accustomed over the past three decades to annual growth rates in the neighborhood of 6-8 percent, these economies entered into severe depressions. Indonesia’s economy shrank by almost 14 percent in 1998. Malaysia, Thailand, Hong Kong, and South Korea each contracted by 5-8 percent. And with the depression came setbacks in living standards, including rises in unemployment, poverty, and school dropout rates and deteriorations in health. Taxpayers in the Asian crisis economies are facing bills on the order of 20-40 percent of GDP to rebuild shattered banking systems. Despite the commitment of nearly $120 billion in IMF-led official rescue packages (for Thailand, Indonesia, and South Korea), confidence (until quite recently) remained depressed.

The region’s largest economy, Japan, was already suffering from a protracted period of near-zero economic growth and a massive bad-loan problem in its banking system when the Asian crisis hit, making both problems worse. The weakness of the Japanese economy and its banking system in turn exacerbated the plight of the emerging Asian crisis countries, which faced both low demand for their exports in Japan and a large-scale withdrawal of loans by Japanese banks. The Japanese economy contracted by almost 3 percent in 1998. Reflecting a shift to more stimulative fiscal and monetary policies, economic activity rose sharply in the first quarter of 1999 (almost 8 percent), but it remains to be seen whether the long-awaited recovery will be sustained, and the consensus growth forecast for 1999 is still close to zero. The Japanese government has made a commitment to spend 60 trillion yen (over $500 billion, or 12 percent of Japan’s GDP) to clean up the banking system mess, and the final tab could be even larger.

The region’s other large economy, China, was more successful last year in maintaining economic growth—in part because its more restricted capital-account regime gave it more leeway than other emerging economies to undertake expansionary monetary and fiscal policies. But China too is now feeling the strain. Exports were flat last year and are falling this year; depreciations in other crisis countries have reduced its competitive position; there has been a huge deterioration in the capital account of the balance of payments; retail prices are falling; its state-owned banks are saddled with a bad-loan problem as large as any in the region; it is confronted with a huge restructuring job in its state-owned enterprises; and it needs robust economic growth to put a rapidly expanding labor force to work. Most analysts expect Chinese growth this year to decline to about 6.5-7 percent—down from 7.8 percent in 1998 and 8.8 percent the year before. 6

Many other emerging economies—from Latin America to eastern Europe to South Africa—found their homegrown economic problems exacerbated by the crises that began in Asia. Net private capital flows to emerging economies (as a group) collapsed last year—to 70 percent below their peak in 1996. That decline in capital flows was particularly marked for portfolio capital flows (bonds and equities) and for bank lending. Interest rate spreads on emerging-economy bonds also soared last year, to over four times the spread prevailing in the immediate run-up to the crisis, and this at the same time as export revenues were falling under the pressure of weak global commodity prices.

In the fall of 1998, after the Russian default and devaluation and following the near-collapse of Long Term Capital Management, a large hedge fund, the turmoil in international financial markets intensified to an almost unprecedented degree. The flight to quality and liquidity was so strong and pervasive that practically all higher-risk borrowers—including those in the United States—saw their borrowing costs surge upward, and many saw their market access curtailed. There was talk of a “global credit crunch” and a “global margin call.” In the end, it took a series of interest rate cuts by the US Federal Reserve and other major central banks to restore order to international financial markets.

In mid-January of this year and notwithstanding a large IMF-led rescue package agreed on three months earlier, Brazil was forced to abandon its crawling peg regime, and for several months its currency, the real, fell sharply. The use of high interest rates to defend the real, in combination with the decline in capital flows and continuing weakness in primary commodity prices, added to contractionary forces in the region. In a pleasant surprise, contagion from the Brazilian crisis has so far been less severe than feared. In addition, the Brazilian situation itself has improved markedly over the past six months. Still, Argentina has been hard hit by the Brazilian crisis; Chile, Colombia, Ecuador, and Venezuela are in recessions; nonoil commodity prices remain weak; and growth in Latin America as a whole this year is expected to be only marginally positive at best.

To be sure, the outlook for recovery from the global crisis is brighter now than it was a year or so ago. Spurred by sharply improved external accounts, lower interest rates, expansionary fiscal policies, and a down payment on financial-sector and corporate restructuring, revisions to growth forecasts for the Asian crisis economies have switched from negative to positive. South Korea, Taiwan, Thailand, Singapore, the Philippines, and Malaysia are now all expected to be firmly in the positive growth column this year. In addition, oil prices have strengthened, equity markets in emerging economies have rebounded sharply, emerging-market borrowing costs have fallen significantly, and a number of crisis countries have been able to reenter the capital markets. Last fall’s seizing-up of some international financial markets, so alarming at the time, has passed.

Nevertheless, many developing economies remain fragile, the global growth outlook is still relatively weak, and important downside risks are evident. In the April 1999 issue of Global Development Finance, the World Bank estimated that economic growth in the developing world would be lower in 1999 than at any time since 1982. The growth of global merchandise trade this year is expected to be less than half of what it was in 1997. The Institute of International Finance projects overall net private capital flows to emerging economies to be only slightly higher this year than last year’s depressed figure. The IMF’s May 1999 World Economic Outlook projected global growth of only 2.3 percent this year—the worst outcome since 1991. Even if the consensus forecast for global growth in 1999 is now closer to 3 percent, this would be the poorest result (excepting 1998) since 1993. In addition, Japan is not yet out of the woods, and the risk of a currency devaluation in China has increased markedly. At its meeting on 30 June 1999, the Federal Reserve’s Open Market Committee raised its target for the federal funds rate by 25 basis points. In late July, Federal Reserve Chairman Alan Greenspan, in his Humphrey-Hawkins testimony before Congress, emphasized that the Fed would “act promptly and forcefully” at the first hint of inflation dangers. If US interest rates were to rise appreciably, the negative impact on growth and external financing conditions in the developing world could be substantial. At this stage, it remains uncertain whether the recovery pattern from the Asian/global crisis will be rapid and sharp (V-shaped) or subject to a second dip.

To sum up, it is sometimes said that there is no better motivation for strengthening crisis prevention and crisis management than undergoing a financial crisis. If that is the case, then given the track record of the past 20 years and the events of the past 26 months, the international community should consider itself highly motivated.

We Are Not Invulnerable To Crises Abroad

Because the US economy has continued to record impressive overall economic performance, some may be tempted to conclude that the US economy is invulnerable to financial crises abroad and that the US stake in strengthening the global framework for crisis prevention and crisis management is small. We reject that conclusion.

The Asian financial crisis struck the US economy at a time when the domestic sources of US economic growth were unusually strong and when inflation was very low. The strength in domestic demand permitted the US economy to absorb a large decline in our net exports (much of it directly linked to the Asian crisis) without suffering a fall in overall economic growth. Our nontradable industries (housing and construction and services) were also buoyed by low long-term interest rates and by a strong US dollar—reflecting large capital inflows seeking a “safe haven” from the crisis and devaluations in the crisis-stricken economies. Thus, while some US states, industries, and companies were hit hard by the Asian crisis, overall economic activity remained robust. In addition, low inflation gave the Federal Reserve the room to cut interest rates last fall when it was needed to calm the turbulence in international financial markets.

If a future international financial crisis were to occur at a time when the US economy was in a much weaker cyclical position and/or when US inflation was less under control, the impact on the US economy could be much more severe. Our defense against crises should not be predicated on the assumption that crises will occur abroad only when the US economy is well positioned to absorb them.

More fundamentally, the United States has a large stake in more effective global crisis prevention and crisis management because the US economy is now more much “connected” to the rest of the world economy—including emerging economies—than it was two or three decades ago, and because today’s global financial system is one where financial disturbances can be transmitted quickly from one location to another. Efforts to safeguard economic prosperity in the United States will therefore have a much higher chance of success if the international community as a whole—and emerging economies in particular—can do a better job of reducing both the frequency and severity of financial crises. As Federal Reserve Chairman Greenspan has aptly put it, the United States cannot expect to remain “an oasis of prosperity” if the rest of the world is in financial chaos.

In addition, financial crises should not be seen exclusively in economic terms. They also have significant political and security dimensions. How assured would the future of democratic governments in Mexico and in the rest of Latin America be if there were little financial stability in the region? How would a prolonged financial crisis in Russia affect prospects for further reducing the global nuclear threat and for encouraging Russia to increase its cooperation with the United States on global security questions? Would the political and security map of Asia (where the United States has fought three wars during the past 50 years and still has 100,000 troops) be unaffected if China’s efforts to avoid a serious financial crisis and Indonesia’s effort to climb out of its deep financial crisis were to falter? These are not questions the United States can afford to ignore.

Today, the average share of exports and imports of goods and services in US national output stands at about 15 percent—twice what it was in 1980 and three times what it was in 1960. Forty percent of our exports go to developing countries, and one-third of them to Asia alone. Forty percent of what our farmers export goes to Asia.

America has benefited greatly from the lower prices, the wider range of choices, and the spur to efficiency that international trade has brought with it. Our exporting plants have higher (labor and total factor) productivity than nonexporting plants of the same size, industry, and location. Similarly, during the 1990s, the most globally dependent 200 firms in the Standard & Poor’s US stock index earned a rate of return that was 5 percent higher per year than the least globally dependent 200 firms in the same index. Studies show that, controlling for everything else, economies that are more open to international trade grow, on average, more than 1 percent per year faster than less open ones.

Exports have accounted for more than one-quarter of US economic growth over the past 15 years. American workers in exporting firms earn 5-15 percent more than workers elsewhere, and that goes for low-skill workers and workers in small firms as well. Employment has grown 15-40 percent faster in US firms that export than in those that do not.

Yes, some Americans have been displaced or had their wages reduced as a result of foreign competition. But the answer to that legitimate concern is not to attempt to wall off America from the rest of the world. It is instead to give those displaced workers the education and training they need to compete better in the global economy, and to continue to push both for the dismantling of trade barriers and the implementation of core labor standards in other countries.

When our trading partners are undergoing banking, debt, or currency crises, their economies are apt to be contracting, not growing, and a shrinking economy is not a promising outlet for US exports. In 1983, at the worst point in the Latin American debt crisis, US exports to that region were almost 40 percent lower than before the onset of the crisis in 1981. Similarly, US exports to Mexico were 11 percent lower in 1995 than they were the previous year, before the peso crisis. The same pattern has been repeated during the Asian crisis. US exports to five Asian crisis economies (Indonesia, South Korea, Malaysia, the Philippines, and Thailand) were 39 percent lower in the second quarter of 1998 than they were immediately before the crisis, in the second quarter of 1997.

As much as our trade connections with the rest of the world have grown over the past 20 to 30 years, our capital flow connections have grown even faster—propelled by, among other things, financial liberalization at home, the dismantling of capital and ex-change controls abroad, dramatic decreases in the costs of telecommunications and of information gathering and processing, and the ascent of institutional investors. A variety of indicators mirror this increasing capital flow interconnection—including cross-border transactions in bonds and equities (up twentyfold as a share of GDP since the early 1970s), inward and outward flows of foreign direct investment (almost double, as a share of GDP, the level of the mid-1960s), the share of pension fund and mutual fund assets invested abroad (more than double the share of 1990), the share of our public debt held by foreigners (double the share of 20 years ago), and average daily turnover in (global) foreign exchange markets (up sixfold since the mid-1980s). 7

Over $2.5 trillion of American savings is now invested in portfolio investments abroad. Overseas (nonbank) affiliates of US corporations hold over $3 trillion of assets and have over $2 trillion in sales. 8 US banks as a group derive 10-15 percent of their profits from foreign operations; for our five largest banks, that profits share is much higher—approaching 45 percent. There are over 700 foreign banks operating in the United States. Almost 350 foreign companies are listed on the New York Stock Exchange (NYSE), and American Depository Receipts (representing shares listed on foreign stock exchanges) traded on the NYSE cover more than 300 foreign companies headquartered in 42 different countries.

Make no mistake: the overwhelming bulk of America’s financial activity and wealth continues to be conducted and invested in the United States itself. But the foreign component is already important and is growing rapidly.

Over the same time horizon, developing countries have become real players in the international financial system. They account for approximately 45 percent of global output, more than one-third of global foreign investment inflows and of global capital portfolio flows, and one-eighth to one-tenth of global stock market capitalization, global issuance of international bonds, and global banking assets. They take one-quarter of industrial-country exports. They include two of the world’s six largest foreign exchange markets (Hong Kong and Singapore) and its third-largest futures exchange (Brazil). During the 1990s, they have been the recipients of over $1.2 trillion in net private capital flows from the industrial countries. In June 1998, industrial-country banks had $370 billion in claims on Brazil, the five East Asian crisis economies, and Russia. All IMF loans since 1976 have gone to developing countries.

Again, this change in financial markets should be kept in perspective. America’s financial links are still much larger with industrial countries than they are with emerging economies. But the time when economic and financial developments in emerging economies could be considered a minor sideshow has passed.

Like our increasing involvement in international trade, America’s increasing integration with world capital markets has paid us sizable dividends. It has reduced the cost of borrowing for households, businesses, and the US government. Lower mortgage rates in turn have helped more Americans to realize their dream of home ownership. Integration has offered American savers and investors an opportunity to obtain the benefits of greater diversification. It has helped to foster the adoption of best-practice production processes in our plants. And because we have (since the early 1970s) followed a regime of flexible exchange rates, it has not unduly constrained the Federal Reserve from lowering interest rates when needed to combat recessions or from increasing them to rein in inflation.

But there is also a potential liability—or risky side—to having closer financial links with other countries. The same links that are so welfare enhancing during normal times expose US households and businesses to greater risk when other countries experience financial crises. If that risk exposure is not well managed, then we too can be seriously affected. For instance, if a financial crisis impairs the debt-servicing capacity of foreign borrowers, then US banks and investors who lent to those borrowers will be hurt. Likewise, when foreign financial crises are characterized by steep, unexpected falls in these countries’ currencies, bond prices, and stock markets, US investors who are “long” in those assets will suffer losses. In addition, there is a real but harder to quantify risk that portfolio losses and declines of confidence in one crisis may lead to wider-scale lending pullbacks, liquidations, and losses of confidence in other markets—thereby widening the arc of exposure to US creditors and investors.

In the 1970s, US money-center banks made a large amount of syndicated loans to governments in Latin America. When the debt crisis broke out in Latin America in the early 1980s and the full repayment of those loans became highly suspect, the market value of US banks’ claims on these countries sank. Because the banks’ exposure to Latin America was larger than their capital, the developing-country debt crisis put the solvency of the largest US banks into question. In the end, the stability of the US banking system was preserved. Nevertheless, US financial regulators were sufficiently chastened by that close call to seek an international agreement (the Basle Capital Accord) on the minimum amount of capital that internationally active banks must hold.

The Asian and Russian crises provide another salient example of potential risks when other countries go into financial crisis. Reflecting the sharp declines in equity and bond prices in the crisis countries, one recent study estimated that foreign equity investors as a group suffered potential losses of roughly $240 billion from the East Asian and Russian crises; the corresponding figures for foreign banks and foreign bondholders were $60 billion and $50 billion, respectively. 9 Between June 1997 and January 1998, US investors are estimated to have lost about $30 billion on Asian equities alone.

In assessing risk, we should also include events that did not happen but could have. Exhibit A concerns the behavior of the US stock market. Americans now have more of their wealth invested in the stock market than they have invested in their homes. With price-to-earnings ratios in the stock market way above their historical average and with dividend yields at record lows, the turmoil in emerging economies over the past two years might well have acted as a catalyst for a major stock market correction. In the event, there were sharp drops on a few days, but no sustained correction. Anyone who thinks that it is only stock markets in emerging economies that are susceptible to large declines has a short memory. On 19 October 1987, the US stock market fell by 20 percent—with little advance warning.

Yet another perspective on risk in today’s international capital markets comes from considering three key characteristics of these markets. One is the multiplicity of channels by which financial strains in one market or region can be transmitted to others. We used to think that crises would be transmitted mainly by bilateral trade and investment links between countries. It is now apparent that the transmission possibilities are much wider than that. For example, a financial crisis in one country may “wake up” investors to similar long-standing but previously ignored weaknesses in other countries, thereby inducing a wider sell-off. As one country after another in a region devalues its currency, other countries that have not devalued suffer a loss in competitiveness, which increases their vulnerability to attack. The fall in aggregate demand in the crisis countries as a group can significantly weaken the world price of primary commodities, thereby increasing the vulnerability of primary-commodity-exporting countries. Chile, for example, has been hurt by the fall in copper prices. When institutional investors or banks suffer a loss on the securities of (or loans to) one or more crisis countries, they may be induced by redemptions, margin calls, internal risk-management guidelines, or regulatory constraints to sell other countries’ securities or call in loans from them—thereby adding to contagion. Even investors who are relatively optimistic about a country may, upon seeing many other investors flee, conclude that others know something they don’t and thus join the group. And if one country undertakes an unconventional policy response to the crisis that is seen as possibly foreshadowing a change in the “rules of the game” (say, by implementing a unilateral rescheduling of its domestic sovereign debt, or by imposing controls on capital outflows), then emerging markets as an asset class can face a higher cost of borrowing. Neither Thailand nor Russia were “large” economies in terms of their direct trade and investment links with the rest of the world; yet crises in those two countries had large repercussions for other emerging economies.

One also sees echoes of these multiple transmission channels in both the players and instruments that participate in today’s international capital markets. Just as it is now more accurate to describe many of our largest companies as global corporations headquartered in the United States rather than as US corporations, it is likewise necessary to think in terms of global financial firms that operate worldwide and transfer their books across time zones to permit 24-hour trading. The troubles of financial trader Nick Leeson at Barings involved a British merchant bank taking a large accumulated unhedged position in Nikkei index futures on the Singapore International Monetary Exchange and on the Osaka Securities Exchange. When Korean banks got into trouble last year and had to shrink their balance sheets, it turned out that they had been large holders of Russian and Brazilian government securities; and when they unloaded those Russian and Brazilian bonds, it added to the strains in those markets.

Today, it is much easier for borrowers and lenders to obtain the combinations of risk, return, and liquidity they want. The menu of available assets is much wider. There are American or General Depository Receipts (negotiable certificates issued by a US bank that represent claims on debt or equity of emerging-market companies), emerging-market country funds, equity options, bond coupons that depend on credit ratings, loans with call options, put options exercisable on credit downgrades, exchange-rate-linked issues that provide protection against devaluations, and many more varieties. It has been observed that there are now at least a dozen different ways of achieving the equivalent economic exposure of a leveraged position in the Standard & Poor’s 500 stocks, and in the United States alone at least six different types of institutions (brokers, mutual funds, commercial banks, investment banks, insurance companies, and futures exchanges) engage in those transactions.

Speed is a second key characteristic of today’s global capital markets. Information now moves with lightning speed around the globe. Financial liberalization and innovation have also made it possible for market participants—including those in emerging economies—to alter quickly the asset and currency composition of their portfolios. “News” that leads to a significant change in investor sentiment or “confidence” can therefore produce large, rapid changes in capital flows and in asset prices—sometimes triggering financial crises in the process. Speed also means that officials face a more compressed time frame for deciding what form of intervention, if any, would be desirable.

Between 1987 and the summer of 1992, approximately $300 billion of private capital flowed into higher-interest-rate European currencies under the assumption that with conversion into a single European currency surely coming, the risk of devaluation (for the higher-interest currencies) was minimal. Why, then, accept the lower rate of interest on a deutsche mark bond when you could get the much higher interest rate on a lira or peseta bond? But then in the summer of 1992 came the negative referendum vote in Denmark on European Economic and Monetary Union (EMU), and what looked inevitable before didn’t look so any more. All of a sudden, currency risk resurfaced, investors rediscovered certain weaknesses in fundamentals among the ERM countries, and there was a massive rush to the exits, which ushered in the ERM crisis. 10

Other sudden and massive shifts in expectations occurred in global bond markets in the first quarter of 1994 (when an unexpected turn in US monetary policy set in motion an almost unprecedented rise of 50-170 basis points in yields on 10-year benchmark government bonds in twelve major industrial countries—all within 50 days), in the yen/dollar foreign exchange market in October 1998 (when massive deleveraging after the Russian crisis and an unwinding of “yen carry trades” led the yen to appreciate vis-à-vis the US dollar by 12 percent on 7-8 October—the largest daily gain since the abandonment of the fixed exchange rate regime in 1971), and of course in private capital markets upon the outbreak of the Asian crisis in the summer of 1997 (when net private capital flows to the Asian crisis economies fell by $100 billion in the second half of the year). 11

Asymmetries in size are a third important characteristic of today’s international capital markets. Two of those asymmetries—that between the financial resources available to the private capital markets and liquid assets of the official sector, and that between private capital flows and the size of emerging-economy financial markets—are worthy of special note.

Private international capital flows to emerging economies have been six times as large (over the 1990-98 period) as official flows. Average daily turnover in global foreign exchange markets is now roughly $1.5 trillion. The global over-the-counter derivatives market is larger than $70 trillion (in notional value). Institutional investors in G-7 countries manage over $25 trillion in assets. US commercial banks alone have over $4 trillion in total assets (at home and abroad), and US insurance companies more than $3.7 trillion. The hedge fund industry is estimated to manage $200-300 billion in capital and to have $800 billion to $1 trillion in total assets. There are individual financial firms that can take (and have taken) positions in the foreign exchange market of $5-10 billion or more.

Unrecorded capital flight from emerging economies was estimated to amount to as much as $70 billion in 1997 and $140 billion during the 1990s. Before the outbreak of the Asian crisis, neither the Philippines, Indonesia, Malaysia, Thailand, nor South Korea had as much as $35 billion worth of gross (nongold) reserves (and in some cases, usable or net reserves were much lower). Mexico had about $29 billion of international reserves in March 1994; eight months later, 80 percent of those reserves were gone. Last year, Brazil lost almost half of its reserves in six months’ time. From 1985 to 1993, about one-third of middle-income developing countries and industrial countries experienced a monthly maximum reserve loss equal to 100 percent or more of their IMF quotas.

At year-end 1997, there were more than 50 developing countries with entire banking systems that were smaller than the credit union for World Bank and IMF employees, and 30 more that were smaller than a medium-sized ($4 billion in assets) metropolitan savings and loan—the same kind of institution that would probably be advised to avoid engagement in international markets because it is too small. The equity market capitalization of large emerging economies, such as South Korea, Brazil, Malaysia, and Taiwan, each totaled $150-200 billion in the mid-1990s—compared to a capitalization for the United States of $6-7 trillion (and now near $13 trillion). A 1 percent shift in the international (not the total) portfolios of G-7 institutional investors would amount to roughly $60 billion. At the height of the capital inflow period preceding both the Mexican and Asian crises, these flows represented 6-8 percent of the recipient country’s GDP. Portfolio flows from the United States alone sometimes represent 10 percent or more of the monthly trading volume in emerging-economy stock markets.

Because of the abundant opportunities for high leverage (that is, high ratios of debt to equity) in international financial markets, it is also possible for individual financial firms to mount very large positions. The classic case in point is the hedge fund Long Term Capital Management (LTCM). According to a recent report by the President’s Working Group on Financial Markets, at the end of August 1998 LTCM had a balance-sheet leverage ratio of more than 25 to 1, which translated its less than $5 billion of capital into total assets of in excess of $125 billion. There were over 60,000 trades on LTCM’s books, including long and short securities positions of over $50 billion each, and positions on some markets that were greater than 10 percent of the daily turnover. It is estimated that LTCM’s top 17 counterparties would together have lost $3-5 billion if LTCM defaulted. While LTCM was unusual in many respects, it was by no means the only large highly leveraged player in international financial markets. At year-end 1998, the five largest US commercial bank holding companies had an average leverage ratio of nearly 14 to 1; the five largest investment banks had an average leverage ratio of 27 to 1; and there were ten other large hedge funds (i.e., with $100 million of capital or more) that had leverage ratios of at least 10 to 1.

Two implications of these size asymmetries stand out. One is that it is much harder now for individual countries to resist the verdict of the private capital markets than it was in the past. For example, if an emerging economy attempts to defend an overvalued exchange rate in circumstances when private markets reach a concerted opposite view, than the chances of a forced devaluation and a currency crisis are high. In this kind of table stakes poker game, the private sector simply has a much bigger stack of chips. The second implication is that moderate-sized allocation shifts by G-10 banks and institutional investors can have a disproportionate (large) impact on emerging-economy financial markets, often producing very large changes in asset prices.

To sum up, contrary to recent appearances, the United States is not immune to financial crises abroad. If a serious foreign financial crisis were to occur at a time when our economy was weak or was actually in recession, the impact would likely be much more severe than it has been during this recent episode. The US economy is now much more connected to the rest of the world than it was two or three decades ago. There have been enough losses, close calls, and “might-have-beens” over the past twenty years to remind us that international capital markets—despite their important overall contribution to our standard of living—are risky places. The more successful we are in putting in place an architecture that can reduce the frequency and severity of financial crises—including in emerging economies—the better are our chances of safeguarding America’s jobs, savings, and national security as well as of promoting global prosperity. Greater financial stability on the part of our trading and investment partners can only be good for us. Like the rest of the world, we have a big stake in a new architecture that can make global financial markets safer.

 

III. The Roots of Financial Crises and Weaknesses in the Existing Architecture

Sources Of Vulnerability

No two financial crises are the same. Macroeconomic imbalances, particularly fiscal policy excesses, were prominent in the 1980s debt crisis but not in the Asian crisis. The Mexican crisis was about sovereign debt, whereas the Asian crisis has been mainly about private debt. The global economic environment was much weaker on the eve of the 1980s debt crisis than it was on the eve of the Asian crisis. Foreign direct investment was more resilient in the Asian crisis than it was in either the Mexican crisis or the 1980s debt crisis.

Yet, as interesting as these differences are, it is the similarities and recurring patterns across financial crises that are important for assessing shortcomings in the existing architecture. There is no shortage of these common factors. Our own list would give pride of place to the following eight interrelated sources of vulnerability: (1) weak national banking and financial systems in emerging economies, along with premature and poorly supervised financial liberalization; (2) poor public and private debt management, with inadequate liquidity defenses against shocks; (3) vulnerable exchange rate regimes in emerging economies; (4) shortcomings in market discipline stemming from inadequate information and from “moral hazard” problems; (5) institutional problems with private debt rescheduling; (6) inappropriate macroeconomic policies in emerging economies—either before or during a crisis; (7) high volatility of private capital flows to emerging economies and occasional bouts of cross-country contagion of financial crises; and (8) threats to the effectiveness of and popular support for the IMF and the World Bank.

(1) Weak national banking and financial systems in emerging economies, along with premature and poorly supervised financial liberalization. Any banking system can come under strain if the macroeconomic environment becomes severe enough. Yet the frequency of serious banking crises in developing countries over the past 20 years tells us that something more than that is in need of repair. While there are important differences among emerging economies, the Asian crisis put the spotlight on many of the common culprits.

Lending standards in the crisis countries were too lax. Most of the Asian crisis countries experienced a lending boom in the run-up to the crisis, with much of that lending going into real estate and equities. Exposure to property accounted for 25-40 percent of bank loans in Thailand, Indonesia, Malaysia, and Singapore, and an even higher proportion in Hong Kong. Bank exposure to equity price movements was high in Malaysia and South Korea. When cyclical conditions later deteriorated and interest rates rose, property prices fell and nonperforming bank loans soared.

Lending standards were also compromised by heavy government involvement in and ownership of banking systems and by high levels of “connected lending” (that is, lending to bank owners, directors, managers, and/or their related businesses). Banks in many of the Asian economies were pressed by the government to allocate some proportion of credit to particular sectors or industries without much regard to creditworthiness. State-owned banks almost invariably have worse loan-loss experience and lower profitability than their commercial counterparts. Connected lending leads to overconcentration of credit risk and the substitution of “insider” interests for arm’s-length transactions. When the rate of return falls in the companies receiving bank loans (as it did in some Asian crisis countries), loan officers are either unaware of it or take little account of it. Collateral (for example, real estate for bank loans) became a substitute for credit assessment. In short, capital was not being channeled to its most productive uses.

A weak accounting, disclosure, and legal framework added to problems. Bad loans were made to look good by lending more money to troubled borrowers (the so-called evergreening of bad loans). Reflecting these lenient loan-classification practices, official published estimates of nonperforming bank loans in several of the crisis countries (for example, Thailand and South Korea) were way below the estimates of independent banking analysts. The legal structure made it difficult for banks to seize or transfer collateral when loans became delinquent.

With several significant exceptions (Hong Kong, Singapore, and the Philippines), bank capital was low relative to the riskiness of the operating environment. Low capital means a small cushion to absorb loan losses. It also means that bank owners have relatively little of their own money to lose if the bank ultimately goes bust.

In many of the crisis countries, bank supervisory agencies lacked the independence, resources, and legal authority to carry out their mandate. In particular, they did not have enough clout to withstand strong political pressures for regulatory forbearance when the financial condition of banks and/or their customers was worsening. Finance companies were subject to even weaker supervision.

With governments having maintained a disciplined fiscal position and with a history of generous support to financial institutions that ran into trouble, there was a widespread expectation that if banks failed, the government would have both the means and the inclination to bail out depositors, creditors, and shareholders. Because there was no well-developed debt market in these economies, banks were the dominant source of intermediation. When the banking system crashed, there were few alternative sources of credit. The impact of the banking crisis on real economic activity was therefore that much greater.

Taken together, these problems might well have been sufficient to generate a banking crisis solely from domestic sources. In the recent crises in Asia, however, as in many others, this fragility was exacerbated by ill-prepared financial liberalization and large-scale capital inflows. Liberalization gave financial institutions greater leeway to engage in excessive risk taking, and large inflows provided much of the fuel to do so. In Thailand, for example, the Bangkok International Banking Facility (BIBF) was established in 1993 and armed with tax and regulatory inducements in order to promote Bangkok as a regional financial center. It was supposed to raise funds from nonresidents and to lend them to other nonresidents (referred to as “out-out” transactions).

Instead, it wound up acting mainly as a conduit for local firms to obtain foreign bank loans (“out-in” transactions). Foreign liabilities of the Thai banking system increased from 11 percent of GDP in 1993 to 27 percent in 1996.

It’s not that financial liberalization per se is dangerous or undesirable; nor is it to deny that foreign ownership of banks offers many advantages to emerging economies (ranging from greater diversification to importation of better risk-management practices). The point instead is that liberalization in the presence of heavy capital inflows becomes a source of vulnerability if it is implemented in a way and/or at a pace that outstrips the domestic capacity to supervise the financial sector and to build a credit culture.

None of the financial-sector weaknesses outlined above are unique to the Asian crisis; they have appeared in many other emerging-market banking crises and in more than a few industrial-country banking crises as well.

As part of their efforts to recover from the crisis, many Asian emerging economies have started to address their financial-sector vulnerabilities by, inter alia, closing insolvent banks and recapitalizing others, setting tougher accounting and loan classification standards, reducing government involvement in the banking system, liberalizing market access to foreign financial service providers, and strengthening banking supervision. In Latin America, banking systems are much improved from their condition in the 1980s—a process that has been aided both by consolidation and by increased foreign ownership.

In addition, countries now have before them a comprehensive international banking standard on which to model the upgrading of banking supervision. The Basle Committee’s “Core Principles of Effective Banking Supervision” was agreed on in September 1997. International standards are also now available for, inter alia, cross-border listing of securities and securities regulation and for supervision of internationally active insurance companies; others are expected soon for international accounting standards and for corporate governance.

But promulgation of international financial standards is only part of the job. To have an impact on crisis prevention, countries must implement and enforce the standards. Their incentives to do so should operate through three channels. One is the expected market payoff. If markets regard implementation of the standards as reducing the risk of lending to these countries, and if markets can tell who is and is not implementing them, then the countries that do so should benefit from a lower market cost of borrowing. The second channel is the IMF-World Bank channel. Specifically, the Bretton Woods twins could give member countries that implemented the standard a better insurance deal (that is, larger access or lower interest rates) when they need loans. A third channel is via the risk weights assigned to various kinds of bank loans under international agreements for regulatory capital. Loans to countries implementing the standards could qualify for a lower (preferred) risk weight, helping these countries to borrow in the market at lower cost.

At present, however, only initial, partial, and tentative steps have been taken to make any of these incentive channels operational. With the exception of the IMF’s data standard, the international financial institutions do not identify publicly which countries are meeting various voluntary financial standards or codes of good behavior. 12 Nor have they been willing thus far to give members that are implementing the standards preferred access to their resources. Progress toward compliance with financial standards is supposed to be taken into account in determining eligibility for the IMF’s new lending window, the Contingency Credit Line, but it remains to be seen what “taking into account” means. The Basle Committee on Banking Supervision has proposed (in its revision of regulatory capital requirements for banks) that countries that do not implement international financial standards be made ineligible for the most preferred risk weights. But the committee did not indicate who is to make that determination of compliance with the standards.

(2) Poor public and private debt management, with inadequate liquidity defenses against shocks. One answer to the question of what have we learned from the Asian crisis is that countries are a lot more like banks than we thought. Like banks, they are susceptible to “runs” if liquidity, maturity, and currency mismatches are permitted to get too large, if leverage is allowed to get too high, and if some shock causes creditors to lose confidence in them.

At the heart of the Asian crisis was a large buildup of short-term, unhedged, foreign currency debt by banks and/or their corporate customers. In June 1997, just before the onset of the crisis, ratios of short-term foreign debt to international reserves had risen to well above 100 percent in South Korea, Indonesia, and Thailand. Other measures of liquidity and currency mismatch, such as ratios of broad monetary aggregates to reserves, pointed in a similar direction. In Thailand and South Korea, it was the banks and finance companies that did most of the foreign borrowing; in Indonesia, it was the corporate sector that took the lead. Debt-to-equity ratios for nonfinancial corporations, which were already very high in these three countries (200-300 percent versus 100 percent in the United States), rose further in 1995-96. In South Korea, short-term foreign debt was as large in 1996 as corporate equity.

It is not hard to see why these unsustainable borrowing patterns developed. Local interest rates in these Asian economies were much higher than those abroad (particularly compared to those available in Tokyo), thereby creating a sizable incentive for foreign borrowing. Since their currencies had been relatively stable vis-à-vis the US dollar in the 1990s, currency risk did not seem to be high. The last time Thailand had experienced a growth rate below 5 percent was in 1972; for South Korea it was 1980, and for Indonesia, 1985. Governments encouraged the banks to continue making loans to corporates. By assuming rollover and currency risk (that is, by borrowing short-term and in foreign currency), Asian borrowers reckoned that they were taking a limited risk for a high return.

The trouble with a debt management strategy that condones large liquidity and currency mismatches and high leverage—even if that strategy has seemingly performed well over a long time—is that it is highly vulnerable to shocks. Any shock that significantly reduces cash flow and net worth and that leads private creditors to lose confidence and “run” can upset the applecart—and there were plenty of such shocks in 1996-97. Growth of merchandise exports slowed dramatically in 1996, to 0.5 percent in Thailand (versus 23 percent in 1995), and to 4 percent in Indonesia (versus over 30 percent in 1995). US dollar prices of semiconductors fell by roughly 80 percent in 1996—in a region where electronics comprise a large share of total exports (above 40 percent for Singapore, Malaysia, and the Philippines). Competitiveness declined because these countries’ currencies followed the US dollar up against the Japanese yen. Property prices fell. Several of the largest Korean corporations (chaebol) went bankrupt and nearly half of the 30 largest suffered losses. Foreign equity investors began to flee, followed by foreign bondholders and foreign banks. Higher interest rates employed to defend currencies added to cash flow strains, and belated hedging by corporations in the foreign exchange market drove currencies lower and magnified debt payments in local currency. Everything collapsed.

Again, the Asian case is not unique. A similar gap between short-term debt and international reserves—this time for public rather than private debt—opened up in the run-up to the devaluation of the Mexican peso in late 1994. Sixty percent of the liabilities of large and medium-sized Mexican companies were then denominated in foreign currency, versus less than 10 percent of their total sales. Similarly, in 1980, just before the outbreak of the developing-country debt crisis, banks in those countries had a net foreign liability exposure of $81 billion; this subjected them to large losses when they subsequently had to devalue.

This problem is still with us. The World Bank estimates that external public and publicly guaranteed liabilities of developing countries currently stand at about two and a half times their international reserves, that 60 percent of these liabilities are at floating interest rates, and that about 20 percent carry maturities of less than a year. More worrisome, a recent survey reveals that 70 percent of sovereign borrowers in developing countries do not currently hedge their interest rate and exchange rate exposures.

On top of this, many emerging economies have issued medium and long-term debt instruments with put options that allow creditors to demand repayment ahead of the scheduled contract date. For 1999-2000, such puts on sovereign and private external debt of emerging-market borrowers amount to more than $30 billion (equal to roughly one-third of scheduled bond redemptions during this period).

It doesn’t have to be this way. There are many things borrowers can do to improve their liquidity, to reduce their currency exposure, and to control leverage. They can lengthen the maturity structure of their debt so that they face lower rollover risk. They can build up their stock of international reserves so that they have an adequate cushion against external shocks in both goods and capital markets. They can make their banks subject to rigorous liquidity and reserve requirements so that they have enough liquid assets on hand to meet sudden deposit withdrawals. They can use derivatives and other market instruments to hedge their interest rate and currency exposure. They can limit the share of new public and private debt that is denominated in foreign currency. They can arrange contingent lines of credit with private banks and other commercial lenders to give them an assured source of liquidity if they need it. They can avoid using medium and long-term debt with put options (that is, debt with an accelerated repayment clause, exercisable at the option of the lender). They can work with their private creditors to design new financial instruments that either forge a closer link between earnings and payments to creditors (as do equities), or extend maturities in the event of trouble, or use export earnings as collateral for credit lines. In extreme cases (as in South Korea), they can set targets for overleveraged corporations to reduce debt-to-equity ratios and instruct banks to withhold loans to corporations that do not meet those targets. They can diversify their funding sources. And they can work to develop a well-functioning domestic debt market so that reliance on external sources of funding is reduced.

None of this can be accomplished overnight. But it can be done. Argentina, for example, has increased the maturity profile of its government debt to such an extent that short-term debt now accounts for only 3 percent of total debt. In addition, it has arranged a $6.7 billion contingent credit line with 14 international banks, it holds international reserves larger than the requirements set by its currency board, and it has imposed a stiff liquidity requirement on its banks. Argentine officials estimate that these liquidity enhancements would be sufficient to meet a 40 percent withdrawal of the deposit base—considerably beyond the 18 percent withdrawal experienced in 1995 during the “tequila effect” of the Mexican crisis. Some other emerging economies also have begun to buttress their liquidity defenses.

But the key point is that the overwhelming majority of developing countries have not yet put in place the prudent debt management and liquidity arrangements needed to cope with today’s volatile international capital markets.

(3) Vulnerable exchange rate regimes in emerging economies. It makes little sense to talk about currency crises without also talking about currency regimes.

Over the past 25 years, there has been a striking trend in exchange regimes for developing countries: more and more of these countries have moved from pegged regimes to flexible regimes. In 1975, about 90 percent of developing countries had pegged rates and only 10 percent followed flexible rate regimes; by 1997, the pegged rate group had fallen to about half. If we weight countries by their output, the trend is even more striking. In 1975, the pegged rate group accounted for about 80 percent of developing-country output; by 1997, their output share had fallen to about 25 percent. Among larger emerging economies with open capital markets, the list of those that have maintained a fixed exchange rate for five years or more is now very short: Hong Kong and Argentina. The Asian/global crisis has contributed further to the trend away from publicly declared exchange rate targets. Thailand, Malaysia, the Philippines, Indonesia, Russia, and Brazil have all been forced into following a regime of managed floating.

Experience suggests that there are at least four sources of vulnerability associated with fixed exchange rate regimes. First, there is a tendency to underestimate currency risk and to avoid hedging against it as the duration of the fixed rate regime increases. In contrast, when a more flexible exchange rate regime is in force, the higher volatility of exchange rates usually makes domestic players more aware of currency risk. This is human nature: the longer something (e.g., a fixed rate) goes on, the more convinced we become that it will continue to go on. One reason why the Asian crisis has been so costly in terms of output losses is that so much of business and banking communities in those countries bet (implicitly or explicitly) against an exchange rate change. A 20 percent devaluation with largely hedged exchange rate exposures is a different animal from one with largely unhedged exposures.

A second source of vulnerability occurs in exchange-rate-based stabilization programs when the nominal exchange rate becomes the “anchor” for the country’s disinflation effort. A common outcome is that the inflation rate does indeed fall dramatically. But the reduced inflation rate is still often higher than that of the country’s trading partners. The country thereby loses competitiveness and the external position worsens—inviting an attack that can bring the currency down. This is more the story of devaluations in Latin America in the 1980s than in Asia in the 1990s.

The third source of vulnerability has as much to do with politics as with economics. When a fixed exchange rate becomes overvalued, there is no graceful way to exit to a more flexible regime. It is a political Catch-22. If the market is not challenging the overvalued peg, there is no political support for devaluing. On the other hand, by the time the markets have begun to apply pressure, the authorities have to deny that any devaluation is being contemplated. By the time the problem is obvious, it is too late. The British, Italian, and Spanish devaluations of 1992, the Mexican devaluation of 1994, the Russian devaluation of 1998, and the Brazilian devaluation of early 1999 are all cases in point.

Interestingly enough, the Asian crisis devaluations of 1997 do not fit that well into this box, because the overvaluations themselves were relatively modest. 13 For example, relative to the 1987-97 average, the real (trade-weighted) effective exchange rate of the Thai baht was overvalued by perhaps 7 percent at the end of June 1997 (just before it was forced to float). The corresponding figures for the Indonesian rupiah, the Malaysian ringgit, and the Philippine peso were 4, 9, and 12 percent, respectively. The fact that there were other associated risk factors—including large current account deficits, a sharp export slowdown in 1996, an upward trend in these exchange rates over the 1995-97 period, and weak banking systems with large prospective recapitalization costs—probably made those modest overvaluations more potent.

The fourth source of vulnerability for fixed rates is what might be called the “David vs. Goliath” problem, that is, the confrontation between a huge global capital market and relatively small economies that can defend a pegged rate with high interest rates only for limited periods.

The key instruments used to defend an exchange rate target are exchange market intervention and interest rates. After international reserves have fallen, high interest rates assume the brunt of the defense. But there are strong limits to how long most emerging economies (or industrial countries for that matter) can keep interest rates sky- high. Vulnerability will be particularly high if the banking system is fragile and has high exposure to property and equity markets, if the government has a large fiscal deficit with a high share of short-term floating-rate debt, if the corporate sector has a high debt-to-equity ratio, and if the economy is experiencing slow growth or recession. In those conditions, speculators will recognize that they can sink the peg if they can push the costs of holding on to the peg beyond the (credibility) costs of reneging on the exchange rate commitment.

Suffice it to say that in most of such battles David and his sling (that is, his fixed exchange rate and high interest rates) have been crushed by Goliath. As suggested earlier, the Asian crisis countries had many of the characteristics that make it difficult to sustain a high interest rate defense. So did Russia and Brazil (although not the same ones). Perhaps even more telling, few believe that this David vs. Goliath asymmetry will disappear in the foreseeable future.

Because they allow the exchange rate to absorb market pressure and because they present less of a “one-way bet,” for speculators, more flexible exchange arrangements are a less crisis-prone regime for emerging economies. Flexible exchange arrangements usually are interpreted to mean floating rates with occasional intervention aimed at leaning against the wind of market forces. Some would also include in that category “crawling band” regimes, in which the authorities are obligated to maintain the rate within a publicly announced band and in which the parity within the band is periodically adjusted to keep the band in line with economic fundamentals.

Although flexible rates are less crisis prone, many developing countries are hardly enthusiastic about them. Given the limited breadth and depth of their financial markets and their high degree of trade openness, they worry that floating rates will be subject to sharp fluctuations, that there will be a tendency for their currencies to “overshoot” a sensible equilibrium in periods of turbulence, and that all of this will be damaging to their economies. Mexico, Turkey, and South Africa are some of the emerging economies with floating rates that have been hard hit during this global crisis—albeit not so hard hit as the worst crisis cases.

The track record of crawling band regimes is also probably best regarded as mixed. Although they tend to produce smaller exchange rate misalignments than adjustable peg systems, their Achilles’ heel is that they typically do not have enough flexibility to handle large capital outflows during a crisis. Chile and Israel have fared pretty well with their crawling band regimes, whereas Brazil, Indonesia, and Russia have been forced into adopting managed floats during this crisis.

Currency boards and single currencies (including “dollarization”) lie on the opposite end of the spectrum from floating rates. A currency board operates like the traditional gold standard without gold. By embedding the currency regime in the country’s legal and constitutional framework, by abolishing the central bank, by putting monetary policy on a rule, and by strongly backing the currency with international reserves, a currency board is said to “lock in” a fixed exchange rate. With a single currency, there is no exchange rate at all.

Currency boards have two attractions. First, for countries that have either just emerged from hyperinflation or may be headed toward hyperinflation, casting aside discretionary monetary policy in favor of putting it on automatic pilot has a definite appeal. Second, because currency boards are harder to “undo” than a simple fixed exchange rate, they can command a lower currency-risk premium (interest rate) in the market. Because both Hong Kong and Argentina have currency boards and their exchange rate regimes are still standing after strong pressures from the Asian crisis, the stock of currency boards has recently been on the rise.

But currency boards also have pitfalls. If the costs of holding on to the currency board get too high, it too can be abandoned. When the currency is under attack and private capital is fleeing the country, a currency board (like the old gold standard) will lead to a rise in interest rates. And if the banking system is fragile or economic activity is very weak, that rise in interest rates can create the same strains it would under other regimes. Hong Kong’s economy contracted by 5 percent in 1998 as high interest rates put a heavy dent in the property sector. In Argentina growth held up rather well in 1998 (just below 3 percent), but for 1999 the economy is expected to shrink by 3-4 percent. Similarly, in 1995 during the tequila effect of the Mexican crisis, Argentina suffered a recession of over 4 percent in 1995. Moreover, the resilience of the Hong Kong and Argentine economies during this crisis is subject to multiple interpretations, because their banking systems are among the strongest of the developing countries.

The single currency option takes the currency board argument one step further. It is supposed to be even harder to “undo” than a currency board and hence should command a lower currency-risk premium in the market. Its proponents also emphasize that low currency risk is important for issuing long-term bonds in the financial markets of the major industrial countries, where creditors have a strong preference for instruments denominated in their own currency. Like a currency board, a single currency also preempts discretionary monetary policy by the national authorities; however, instead of putting domestic monetary policy on a rule (as in a currency board), a single currency puts monetary policy in the hands of the anchor country’s central bank.

Here too there are concerns and problems. Specifically, if cyclical asymmetries, financial-sector problems, or other factors were to call for a monetary policy in the emerging economy that differed from that of the anchor country, a single currency could not accommodate that need. In this connection, US monetary and treasury officials have recently made it clear that while they are open to discussing the option of “dollarization” with emerging economies in Latin America and elsewhere, they would not be prepared to alter the course of US monetary policy to reflect economic developments in dollarized economies—nor would they be prepared to expand US lender-of-last-resort responsibilities to include financial institutions in those economies. 14

The implications of the euro zone for single currencies elsewhere are also less than straightforward. While some see the launch of the euro as tangible proof that currency zones are made and not born, others point out that agreement on European Monetary Union came about only after a 50-year period of preparation and that it reflects a larger political initiative that simply does not exist in the Americas or in Asia. Individual countries could choose to dollarize on their own without a monetary treaty, but then the reduction in currency risk would be smaller (that is, if countries dollarize without either a voice on monetary policy or any arrangements for emergency financial assistance, the reduction in their borrowing costs from dollarization would be smaller). Single currencies are a live and maybe even an attractive option for many emerging economies a decade or more down the road—but not right away.

Yet another perspective on the currency regime issue is that whatever emerging economies choose to do about their own exchange arrangements, we will not achieve greater systemic stability without some reform of the G-3 exchange rate regime. With respect to the Asian crisis, critics of the existing G-3 floating exchange rate regime contend that the very large (50 percent plus) swings in the yen/dollar rate during the 1990s—with the dollar depreciating in the first half of the decade and then appreciating vis-à-vis the yen from 1995 to 1997—put the competitiveness of the Asian crisis economies on a yo-yo course that contributed significantly to their boom-bust cycle; that the gyrations of the yen added substantially to Japan’s economic instability; and that the rise of the dollar has stoked the huge and rising US trade deficit.

More generally, they argue that high short-run variability of G-3 exchange rates has discouraged international trade and investment and has resulted in costly hedging and plant location decisions; that the present system has generated large and persistent exchange rate “misalignments” that have induced large-scale resource misallocation and have fanned protectionist pressures; and that the absence of announced exchange rate targets has impeded both stabilizing speculation and more effective international economic policy coordination.

They would prefer a system of wide (10-15 percent on either side) “target zones” among the dollar, the yen, and the euro. This, in their view, would represent a third way between the crisis-prone rigidities of fixed rates and the overshooting and misalignments of floating rates.

Many others are not moved or convinced by these arguments. They maintain that the Asian crisis was mainly about financial-sector weaknesses and corporate borrowing excesses in the crisis countries—not about large exchange rate misalignments; that there is little reason to believe that wide and changing target zones for the G-3 currencies would induce stabilizing speculation when much tighter arrangements in Europe—with greater political will behind them—produced precious little of it; that orienting G-3 monetary policies toward the exchange rate would produce inferior bottom-line results for economic growth and inflation than does the existing approach to monetary policy; that it is simply not credible to suppose that a G-3 country will raise interest rates in the midst of a recession for the sake of a target zone; and that the effectiveness of (sterilized) exchange market intervention in today’s huge international capital markets is very limited.

To sum up, debates over the exchange rate regime have been going on in economics for over a century. The outcome of this debate has sometimes been described as a pendulum that swings (every decade or two) between the poles of fixed and floating exchange rates. If that is the right analogy, then the Asian crisis would seem to have pushed the pendulum in the developing world closer to both poles simultaneously, that is, toward greater flexibility and toward more polar cases of fixed rates such as currency boards.

The IMF’s Articles of Agreement give countries wide scope in their choice of exchange arrangements. But experience strongly suggests that some of those choices are wiser than others. The IMF and the G-7 countries can influence the exchange rate regime choices of developing countries—not least by the financial support that they are prepared to extend to support adjustable peg exchange rate regimes during periods of market stress. Global crisis prevention will not be able to gain more traction until the IMF and the G-7 are prepared to say “no” more often to defending overvalued fixed rates.

(4) Shortcomings in market discipline stemming from inadequate information and from “moral hazard” problems. In a market-based system of finance, there should be powerful forces operating to rein in errant behavior by borrowers and lenders. When a firm or a bank overborrows, it should face a rising cost and/or a reduced availability of funds; that increased risk premium, in turn, should induce the borrower to reduce its exposure and get its house in order. Similarly, creditors, investors, and managers who make bad credit and investment decisions should be penalized sufficiently such that they do not make the same mistakes in the future.

In practice, however, inadequacies of information and moral hazard problems associated with official explicit and implicit safety nets (that is, guarantees and bailouts) can either blunt the impact of market discipline or cause it to operate in a delayed and draconian way. If market participants cannot obtain timely and comprehensive information on the borrower’s creditworthiness, they will not be able to price the borrower’s obligations correctly. And if borrowers know that certain kinds of financial information do not have to be disclosed publicly, they will be more likely to hide their problems under those rugs (statistics).

In our context, “moral hazard” means the provision of insurance by the official sector that weakens investors’ and borrowers’ sense of responsibility for their own actions. If market participants expect an official bailout of troubled borrowers, then the interest rate will reflect the creditworthiness of the guarantor—not that of the borrower—and creditors will have little incentive to monitor the financial condition of borrowers. Both of these problems have been present in recent financial crises.

Published data on gross international reserves gave a very misleading impression of the net international reserve positions of Thailand and South Korea. This was because there were substantial (unpublicized) commitments in the forward exchange market and because the Bank of Korea had placed foreign currency deposits with overseas branches of Korean banks that became illiquid. While data on international borrowing by banks were good enough to see trouble coming for South Korea and Thailand—if investors had actually looked at it—data on the external debt of the nonbank corporate sector (which was important particularly in Indonesia) were less adequate. The official estimate for the share of nonperforming loans in the Korean banking system in 1996 was 0.8 percent—10-20 times less than estimates of independent analysts. Thailand did not publish a figure for nonperforming loans in 1996. Publicly reported figures on the share of nonperforming loans similarly gave little hint of banking crises in Chile and Colombia in the early 1980s. The Chinese government has long funneled its aid to loss-making state-owned enterprises, not through the budget but through the state-owned banks, and it is only recently that official estimates of nonperforming loans have begun to approach those of private-sector analysts. Before it collapsed in January 1998, Peregrine Investments (based in Hong Kong) had become Asia’s largest investment firm outside Japan. Its off-balance sheet activities were approximately 10 times larger than its on-balance sheet activities, and the former included about $10 billion of interest rate and currency swaps in Indonesia. Moreover, if investors were looking for help in evaluating credit risk from either the major credit-rating agencies or the IMF, they would have found little indeed. Long-term sovereign credit ratings issued by Moody’s and Standard & Poor’s were essentially unchanged during the 18-month run-up to the crisis, and the IMF’s policy at the time was not to publish the part of its Article IV country reports that contained the assessment of a country’s economic policies and prospects.

None of this is to say that lack of information was the key factor in the Asian crisis. Faulty analysis and plain euphoria probably were more important. Nevertheless, we have a wide body of experience suggesting that markets function much better when they have better information. US Treasury Secretary Lawrence H. Summers has argued that the single most important innovation shaping the American capital market was the idea of generally accepted accounting principles.

During the past 24 months, progress has been made in addressing information and disclosure weaknesses. The IMF’s Special Data Dissemination Standard (SDDS) has been amended to include data on reserve-related contingent liabilities (that is, on net international reserves) and to provide better coverage of the foreign liquidity position of the corporate and government sectors. The Bank for International Settlements (BIS) has upgraded its international banking statistics, and the Basle Committee on Banking Supervision has prompted large international banks and securities firms to be more forthcoming about their derivative positions. Several of the Asian economies—including Japan and China—are also in the process of revamping their loan classification procedures. And, in a break with tradition, in April 1999 the IMF initiated a pilot program under which member countries can agree to have their full Article IV country reports published.

Turning to events that could be seen as generating moral hazard and weakening market discipline, the $50 billion Mexican rescue package of February 1995 allowed those holders of tesobonos who held on to get out whole. The international community committed $190 billion (about one-third of which has been disbursed thus far) in the official rescue packages for Thailand, Indonesia, South Korea, Russia, and Brazil, and some of the disbursed funds allowed private creditors with short-term debts to make a much less costly exit than would otherwise have been the case. The Miyazawa Plan added $30 billion more to Asian rescue packages. The Thai, South Korean, and Indonesian authorities issued broad guarantee announcements for bank depositors and creditors shortly after the onset of their crises. When South Korea did reschedule $24 billion of short-term bank debt in January 1998, $20 billion of it was guaranteed by the South Korean government, and the interest rates charged (of 225-275 basis points over six-month LIBOR) were considerably above immediate precrisis rates. When finance companies in Thailand were closed, there was originally an effort (rebuffed successfully by the IMF) to pay off owners with government bonds at full market interest rates. A recent survey estimated that $20 billion of Korean government guarantees have been issued to the chaebol. The World Bank and the Asian Development Bank have issued partial guarantees to help several Asian economies reenter capital markets, and in mid-May 1999 the Japanese government announced a $17 billion plan with the same intent.

On the other side of the moral hazard ledger, equity investors and, to a lesser extent, bondholders have (as mentioned in Section II) experienced large losses in the Asian crisis. 15 Banks suffered losses on their lending to Asian corporates and on their off-balance-sheet activities; they escaped with relatively small losses on their loans to banks in the Asian crisis countries because they cut their exposure in the second half of 1997 and because most bank loans have either been paid or rescheduled (with government guarantees). As of August 1998, total reported provisions (for losses) by US banks amounted to less than 3 percent of their claims on East Asian countries; the corresponding figures for Japanese banks was 3 percent, and for French, German, and UK banks, less than 8 percent. Banks were much harder hit in the Russian crisis. In fact, it has been said, with some justification, that while Russia was a terrible model for how to reschedule debt, and while it acted as a trigger for widespread turbulence in international financial markets, it did have one important silver lining: if in the end the only thing that will deter excessive risk taking is the memory of earlier losses, then the Russian crisis refreshed the memory of many banks.

So as not to get too carried away with moral hazard, it is well to note two caveats. First, moral hazard is a potential problem with all insurance arrangements—emergency financial rescues and otherwise. The usual response is not to provide no insurance at all but rather to limit the amount of payment (e.g., deductibles, coinsurance, etc.) and/or to price the insurance appropriately (i.e., higher insurance premia for more risky policyholders) so that moral hazard effects are kept under reasonable control. The second caveat is that by providing emergency assistance to an illiquid (but not insolvent) borrower and thereby preventing a costly default and its possible spillover to other borrowers, a lender of last resort serves a useful function for the economy as a whole. This latter consideration applies with more force when the illiquid borrower is a bank, because of the special role banks have in the payments system and because—particularly in emerging economies—banks dominate the credit intermediation process. This also helps to explain why rescheduling of bank debt often takes place more quickly than rescheduling of corporate debt—that is, the former is perceived as having more systemic urgency than the latter.

Notwithstanding these caveats, there are six points on the moral hazard issue that merit emphasis.

First, it is not convincing to argue that because the experience of the Asian crisis has been so wrenching and costly for borrowers, a repetition of overborrowing there or elsewhere in emerging economies is unlikely. This is the same argument that was made after the costly Mexican peso crisis in 1994-95. The Asian crisis (with all its excesses in short-term borrowing) erupted less than two years later. Moreover, as documented in Section II, there have many costly banking crises over the past 20 years; apparently the “demonstration effect” of earlier crises was not strong enough by itself to deter later ones. It is one thing to have undergone a costly financial crisis, but it is quite another to be sufficiently motivated by that crisis to make a large enough investment in crisis prevention to deter the next one. And that motivation on the part of borrowers may not be independent of the scope and size of the official safety net.

Second, while it is true that the proper functioning of financial markets rests on the willingness of borrowers to pay in full and on time when they can, it is also true that default or rescheduling should not be considered a largely unanticipated event when creditors are sometimes receiving interest rates on emerging-market securities anywhere from 300 to 5,000 basis points above US treasuries. Over 60 percent of the balance of payments crises since 1970 have involved debt restructuring or an increase in arrears. The market system doesn’t say that private creditors should get two scoops for assuming risk—one in the form of a higher risk premium when the debt is purchased, and a second in the form of an official bailout if things work out badly.

Third, restoration or noninterruption of market access cannot be the sole criterion for judging the success of financial rescue packages. Speedy restoration of market access can always be assured by having G-7 governments and/or the IMF guarantee all external private claims on emerging-market borrowers, public and private. This may appeal to both private creditors and emerging-market borrowers. But it would transfer an inequitable burden to the guarantors, would likely encourage overlending and overborrowing in the future, and could result in a much larger (taxpayer- financed) bailout down the road. There are three players in this game, not two, and the third—industrial—country governments and the IMF—has as much right to ask for equitable burden-sharing as do the other two. This is why the Paris Club of official creditors has long demanded “comparable” concessions from private creditors when it agrees to reschedule official debt. Market access is important, but it needs to be balanced against other objectives in crisis resolution—including the need to discourage future crises.

Fourth, if one class of emerging-market debt (say, eurobonds) were to be automatically excluded from rescheduling, then a distortion in the composition of capital flows would likely develop over time (in favor of the protected category). It is hard to see why such a distortion should be promoted by public policy unless that category of debt generates some wider benefits. And if we are looking for wider benefits, they are probably larger for foreign direct investment than for other types of financial flows to emerging economies.

Fifth, if one wants to see what happens when moral hazard effects become large, one need look no further than recent private capital flows to Russia and the Ukraine—widely known on Wall Street as “the moral hazard play.” Here, despite serious underlying weaknesses in the economic fundamentals, investors were prepared to purchase large amounts of high-yielding government securities—presumably in large part under the expectation that if conditions worsened, geopolitical and security concerns would prompt G-7 governments and the IMF to bail them out. In the first half of 1998, Russia placed four issues of eurobonds that doubled the outstanding stock. More generally, because governments in developing and industrial countries alike find it very difficult to credibly commit to pulling the plug on large borrowers, there is often an opportunity for private investors to capture some of those insurance benefits for themselves if they can get the timing right. Interestingly enough, some of the new models of financial crisis predict that the rush for the exits will come just when the govern-ment’s reserves (cum any resources from outside official creditors) fall to the level of its contingent liabilities.

And sixth, public support for the institutions that manage crises and that dispense emergency financial assistance is not independent of perceived “fairness” in crisis resolution. One of the reasons why it proved so difficult in 1998 to get US congressional approval for an IMF quota increase was that members on both sides of the aisle apparently felt that recent rescue packages were benefiting Wall Street—and particularly large banks—much more than Main Street. Likewise, if the burden of crisis resolution in the borrowing countries falls more heavily on households than it does on the banks and corporates that are regarded as being most responsible for the crisis, the institutions managing the crisis are apt to become unpopular.

Less progress has been made over the past 24 months in addressing moral hazard concerns than in dealing with transparency and disclosure shortcomings. Two problems stand out.

One is that we do not yet have enough momentum or a timetable for deposit insurance reform in emerging economies—even though the major source of moral hazard occurs at the national and not the international level. “Reform” in this context means an insurance system that puts large uninsured creditors of banks at the back of the queue when failed banks are resolved, that places more stringent accountability conditions on senior economic officials when they invoke “too large to fail,” and that gives banking supervisors more protection against strong political pressures for regulatory forbearance. Here, there is much that can be successfully exported from the banking reforms adopted in this country in the wake of the savings and loan debacle. Those reforms go under the heading of structured early intervention and resolution (SEIR) and prompt corrective action. They were included in the Federal Deposit Insurance Corporation Improvement Act (FDICIA) of 1991.

FDICIA retains deposit insurance for “small” depositors (up to a maximum of $100,000 per depositor), since they are least likely to be able to ascertain the true financial condition of banks and thus are most likely to engage in uninformed bank runs. FDICIA also makes it harder for regulators to bail out large uninsured creditors of banks. Specifically, FDICIA requires the FDIC to evaluate all resolution alternatives and pick the one that carries the lowest cost to the deposit insurance fund. There is a systemic override to protect all bank creditors in exceptional circumstances, but this requires the explicit consent of the secretary of the treasury in consultation with the president of the United States, two-thirds of the governors of the Federal Reserve, and two-thirds of the directors of the FDIC. FDICIA also specifies that banks be subject to progressively harsher regulatory sanctions as their capital falls below multiple capital-zone tripwires, that well-capitalized banks receive “carrots” in the form of greater banking powers and lighter regulatory oversight, and that there be an explicit exit rule for banks that closes them down while they still have positive net worth.

Most emerging economies either have no explicit deposit insurance scheme at all or have one that does not contain the right kind of incentives. The Group of 22 (G-22) supported deposit insurance reform along the lines of SEIR in its October 1998 report on strengthening national financial systems, but it seems not to have received much emphasis in recent G-7 statements on the architecture.

The second problem has to do with the international sources of moral hazard. Here, a good argument can be made for having the IMF return as soon as possible to normal access limits (i.e., 100 percent of IMF quota on an annual basis and 300 percent on a cumulative basis) in dealing with country crises. As noted in the Introduction, some recent Fund-led rescue packages have gone well beyond these normal access limits—ranging from the 500-700 percent of quota packages for Mexico, Thailand, and Indonesia to the 1,900 percent of quota package for South Korea. 16 In addition, the Contingency Credit Line created in April 1999 envisages access limits for qualifying countries in the range of 300-500 percent of IMF quotas.

Smaller IMF loans for country crises would place more of the responsibility for crisis prevention and resolution on debtors in emerging markets and on their private creditors. In thinking about the implications of a smaller IMF, three points should be noted.

First, adherence to normal access limits would still permit Fund programs to reduce the recessionary impact of a balance of payments crisis, to finance some smoothing operations in the foreign exchange market, and to make a contribution toward the costs of banking reform and recapitalization. But it would curtail—desirably in our view—the scope for supporting overvalued fixed exchange rates and for bailing out large, uninsured private creditors. Some of the large rescue packages (IMF plus bilateral support) of the past four years (Mexico and Indonesia) have been big enough to cover all short-term debt of the recipient country; some (South Korea, Russia, and Thailand) have not. In any case, the reality is that private capital flows (including potential capital flight) are now just too big to expect Fund-led rescue packages to cover fully all financing gaps faced by emerging economies.

Second, it is not obvious that there is a unique level of Fund financial support that is associated with regaining “confidence” in a crisis country. In recent experience, conditions did not seem to stabilize right after the signing of a Fund program. Instead, the return of confidence came later, when there was stronger evidence both of political leadership and of concrete policy actions to address the underlying economic problems. In this sense, we are not persuaded that a somewhat smaller IMF would be less effective as a crisis manager.

Third, a smaller IMF that could provide less support to private creditors in crisis situations would probably also imply that developing countries would in the future face a somewhat higher cost of borrowing and perhaps a smaller flow of external financing. This is just what one would expect to happen if an official (implicit) subsidy to that lending were removed. Since spreads on emerging-market borrowing have been too low and the flow of capital to them too high during much of the 1990s, some moderate movement in the opposite direction would be no bad thing (especially for those economies with relatively high domestic savings rates).

The IMF recognized the implications of its rescue operations for creditor behavior when in the late 1980s it adopted a policy of “lending into arrears.” This policy reversed an earlier position that barred the IMF from lending to countries that were in arrears to their private creditors. The problem with the earlier policy was that it gave banks too much of an advantage in their debt negotiations with sovereign borrowers (since failure to reach agreement with the banks also disqualified the country from getting IMF loans). Under the new policy, the Fund could lend to countries that were in arrears to private bank creditors so long as the country was meeting the conditions of the IMF program and was engaged in “good faith” negotiations with private creditors. This helped to level the playing field. In September 1998 the policy was extended to cover bond and other nonbank credits as well.

The issue now is whether “lending into arrears” goes far enough to induce equitable burden-sharing by private creditors. In our view it does not, and in Section IV we offer some concrete suggestions on how Fund lending might be conditioned on greater private-sector burden-sharing.

(5) Institutional problems with private debt rescheduling. One of the lessons of the financial crises of the past two decades is that long delays in rescheduling an unsustainable debt burden that is unlikely to be paid anyway benefits no one—least of all the real economy of the debtor country. But wishing for private debt rescheduling to become more timely and orderly will not make it happen, particularly when there is neither an international bankruptcy code nor, in many cases, good national bankruptcy laws, and when there is no private-sector analogue to the Paris and London Clubs (which deal with rescheduling of officially held public debt).

A notable feature of the composition of private capital flows to emerging economies as one moves from the 1980s to the 1990s is the much-reduced share of syndicated bank loans and the rising share of other types of flows—mainly bonds in the case of gross financing flows, and foreign direct investment and portfolio flows (equities and bonds) in the case of net flows. For example, in the 1980s, syndicated bank lending represented over three-quarters of gross private market financing to emerging economies; in the 1990s, the proportion has fallen to less than half. Over the same time span, bonds increased their share from about one-quarter to nearly one-half. Most analysts expect the trend toward greater reliance on bond finance to continue, driven by a desire for liquidity and long maturities and by the increasing importance of institutional investors.

But the increasing importance of bonds and of securitization brings with it a problem: relative to syndicated bank loans, sovereign bond contracts are rescheduling-unfriendly. More specifically, unanimous consent usually is required to restructure them, individual bondholders can sue the issuer, successful lawsuits can trigger both cross-default clauses on other securities and accelerated repayment schedules, and there is no requirement that proceeds recovered in litigation with other bondholders be shared. Also, ownership of bonds tends to be diffuse and—unlike bank loans, where there are bank advisory committees—there are no standing steering committees to handle negotiations from the creditor side. This can make it costly and time consuming to organize creditors on the spot (as the creditor discussions after the Russian default of August 1998 illustrated). On top of this, federal regulators have less leverage to lean on bondholders than they do on their banks. Faced with all these obstacles, the debtor may just see rescheduling as too daunting an undertaking to embark upon—even if such rescheduling would benefit the majority of both debtors and creditors.

The solution to this problem is straightforward, at least in principle. The terms of existing bond contracts could be altered to include “collective action clauses” (i.e., majority-voting, sharing, and nonacceleration clauses) that would make it harder and less profitable for a few rogue creditors to block a rescheduling; and standing steering committees (representing bondholders, banks, and others) could be organized to conduct future creditor negotiations. Both the G-10 and the G-22 countries have expressed the view in recent reports that these institutional changes would be welcome. The only objection has come from private creditors, who have opposed mandatory inclusion of rescheduling clauses in bond contracts as making default too easy.

The rub is that neither the major creditor countries nor the IMF has thus far been prepared to take actions to make these institutional changes occur. The G-10 countries have not been willing, for example, to include these collective-action clauses in their own sovereign bond contracts. The difficulty is that if only emerging-market bonds include these clauses, it might be taken by the markets as a signal of their need to restructure; they will therefore not do it alone. No signal would be forthcoming if the bonds of the most creditworthy borrowers also included such clauses. The G-10 countries have likewise not been willing to restrict issuance or trading of bonds in their markets to those with collective-action clauses; nor has the IMF made any provisions for countries whose bonds include these clauses to get “carrots” from the IMF. And nothing much has been done by the official sector to encourage the formation of standing steering committees of creditors. Thus, while there is considerable intellectual agreement on what should be done, little has changed on the ground.

(6) Inappropriate macroeconomic policies in emerging economies—either before or during a crisis. Usually emerging economies are chastised for their loose macroeconomic policies in the run-up to a financial crisis and praised for their tighter macroeconomic policies after the outbreak of the crisis. For the Asian crisis countries and the IMF, the pattern of criticism this time has been just the opposite.

Thailand, Indonesia, Malaysia, the Philippines, and South Korea were in fiscal surplus or in fiscal balance for most of the 1990s. Nor was inflation a serious problem. There were, as noted earlier, some large current account deficits and some bad lending booms in several of these countries; but the overall macroeconomic situation was much better than either in the heavily indebted developing countries in the early 1980s or in Mexico in 1994.

The fact that the Asian crisis economies did not conform to the fiscal deficit model of financial crises does not imply that the latter has become an extinct species. On the eve of their respective financial crises, both Brazil and Russia had large fiscal deficits (equal to 8 percent of GDP for Brazil and 6-7 percent for Russia), and questions about the medium-term sustainability of their debt positions were at the heart of their vulnerability. More generally, there is considerable empirical evidence to suggest that countries that exhibit relatively high degrees of macroeconomic instability—particularly a high frequency and severity of recessions—have a higher susceptibility to banking and balance of payments crises than countries that do not.

But what about the proper conduct of fiscal and monetary policy once a financial crisis has already struck? No other issue in the Asian crisis has been debated more. Much of the criticism takes the line that the crisis in Asia would have been shallower and would have ended sooner if the IMF had prescribed, and the crisis countries had taken, the standard Keynesian medicine for recessions, that is, expansionary fiscal policy and easier monetary policy (lower interest rates).

Even though the Asian crisis countries were running disciplined fiscal positions, the IMF felt that some fiscal tightening was necessary to cover the interest-carrying costs of bank restructuring and to help restore a sustainable current account position. The original fiscal adjustment was planned to be about 3 percent of GDP in Thailand, 0.5 percent in South Korea, and 1 percent in Indonesia. The IMF acknowledges that it—like practically everyone else—underestimated the depth of the crisis in each country (and in the region as a whole). As evidence came in that these economies were going to be weaker than anticipated, the IMF loosened the fiscal targets and continued to do so as growth continued to fall below forecasts. The last agreement with South Korea, for example, called for a fiscal deficit no greater than 8.5 percent of GDP. On the basis of its own review of the evidence, the Fund concluded that fiscal stringency was not a major factor accounting for the output decline in the Asian crisis economies. Critics of the IMF argue that it was apparent earlier on that the decline in consumption and investment in the crisis countries would be very large and that therefore an expansionary fiscal policy should have been proposed right away.

The debate on monetary policy has been even more contentious—and in many ways even less decisive. The Asian crisis countries were suffering from two serious problems at the same time: first, a recession overlaid on a business and financial sector with a high debt burden and cash-flow vulnerabilities; and second, a rapidly falling currency with a banking and corporate sector that was saddled with large unhedged foreign-currency liabilities. The conventional remedy for the first is lower interest rates, and for the second, higher interest rates. Several of the Asian crisis countries initially vacillated in raising interest rates but then raised them sharply as their currencies plunged and as they entered into stabilization programs with the IMF.

Critics of high interest rates point out that a higher interest rate will not necessarily strengthen the currency if it reduces the probability of payment even more. Defenders counter that what matters is not only the ability to pay but also the willingness to pay. Furthermore, an expansionary monetary policy for an emerging economy with debt problems could be seen as an effort to inflate or devalue away debt obligations denominated in domestic currency. While equity holders may see higher interest rates as having a negative effect on growth and earnings in the medium term and thus as making investments there less attractive, fixed-income investors—especially in short-maturity bonds—may see higher interest rates as crucial to compensate for currency risk.

Economic theory tells us that a country can attain any two of the following three objectives: a fixed exchange rate, independent monetary policy, and open capital markets. The implication is that the scope for lowering interest rates (without weakening the exchange rate) is increased if restrictions are put on capital outflows. Sure, such restrictions have adverse resource allocation effects, become more porous the longer they are in place, and will discourage capital inflows in the future, but—so the argument goes—maybe these costs will still be smaller than what would be gained by cushioning the depth of the recession.

This was apparently Malaysia’s reasoning in imposing such controls on capital outflows last summer. While it is hard to know the counterfactual, Malaysia’s experiment does not look like a resounding success: the economy contracted by over 7 percent in 1998. China’s case is more intriguing. As suggested in Section II, its lack of capital account convertibility gave it more room to pursue antirecessionary fiscal and monetary policies, and it was much more successful than its neighbors in sustaining growth last year. But its relevance is less clear for countries that have already dismantled their capital account restrictions and are not eager to go back. Thailand and South Korea have by now reduced interest rates to below their precrisis levels without the aid of restrictions on capital outflows and are expected to show some recovery this year; but they also experienced deep recessions last year.

To muddy the waters further, there is not enough empirical evidence on the relationship between interest rates and exchange rates in emerging-market crisis situations to warrant a strong conclusion. Nor do individual case studies offer much guidance. For example, we know that Sweden’s gambit of raising overnight interest rates to 500 percent per annum during its 1992 crisis was not credible given its banking problems. But that tells us little about prospects after raising short-term interest rates from 6 percent to, say, 50 or even 80 percent for a few months until things begin to stabilize (as both Mexico did in 1995 and Brazil did this year), and then bringing them down steadily after that. The United Kingdom (in 1992) and Australia (in 1998) did much better by taking more exchange rate depreciation and lowering (or leaving unchanged) interest rates—but their banking systems and central bank independence have probably acted to limit overshooting of exchange rates in ways that are not yet replicable for most emerging economies.

In the end, we suspect that even after the Asian crisis, most emerging economies that are hit simultaneously by strong currency pressures and recession will opt to raise interest rates (for at least a short period) and then “feel their way” toward lowering them step by step. No one thinks this is a particularly appealing strategy—but the others look even less appealing. Despite all that has gone on, very few emerging economies have chosen to apply capital-outflow controls. And even fewer have been willing right from the outset of the crisis to challenge the conventional market view of “confidence raising” by going for low interest rates and expansionary fiscal policies. Maybe we would all be better informed if someone did—but so far no emerging economy in crisis wants to be that white rat.

(7) High volatility in private capital flows to emerging economies and occasional cross-border contagion of financial crises. Whatever the underlying causes of financial crises in emerging econ-omies, the forcing event is often a sudden stop or reversal in private capital flows. In addition, there are some episodes in which crises occur in clusters—seemingly displaying “contagious” behavior. Some observers define contagion simply as the spread of crises or financial difficulties across countries. Others want to go further and distinguish regular contagion from “true” contagion, in which crises spread beyond what would be implied by trade and financial linkages, that is, herding behavior.

During the postwar period, there have been two waves of large private capital flows to emerging economies. The first was in the late 1970s and early 1980s and was fueled by the recycling of petrodollars. It ended with the onset of the developing-country debt crisis in 1982. Net flows remained depressed for the rest of the 1980s (0.3 percent of emerging-market GDP for 1983-90 vs. 0.7 percent for 1980-82). The second and much stronger wave was in the 1990s, when net private flows mushroomed (to 2.3 percent of emerging-market GDP), only to come crashing down with the outbreak of the Asian crisis. In absolute terms, the peak capital inflow year was 1996; relative to emerging-market GDP, it was in 1993. For the five East Asian crisis countries, the upsurge in flows was much more marked—from 1.1 percent of their GDP for 1983-89 to 4.8 percent for 1990-95.

In contrast to the aftermath of the Mexican crisis, in which the rebound in private flows was quick enough to prevent any year-on-year decline for emerging markets as a group, this crisis has produced a whopping fall in aggregate flows—from $213 billion in 1996 to just over $60 billion (preliminary) in 1998; for the five Asian crisis countries, the drop has been sharper yet—from a net inflow of $65 billion in 1996 to a net outflow of $43 billion in 1998.

The same kind of volatility is evident in the prices or terms of financing. For example, in the 1990s, the average secondary market spread for Brady bonds fluctuated from 1,200 basis points above US treasuries in January 1991, to a little over 400 basis points in January 1994, back up to over 1,600 basis points after the outbreak of the Mexican crisis in January 1995, back down to less than 400 in mid-1997, back up to over 1,400 after the Russian default, and receding to the neighborhood of 1,100 basis points in July 1999.

There have been four major episodes of contagion in the 1990s: the 1992-93 crisis in the European Monetary System, the 1994-95 Mexican peso crisis, the 1997-98 Asian crisis, and the 1998-99 Russian/Brazilian crisis.

Recent research suggests that contagion has at least five characteristics.

First, contagion is stronger on the downside. While it can be evident in periods of euphoria as well as in crises, it appears to operate with greater impact during periods of turbulence than during normal times.

Second, contagion is mainly regional. While the global aspect of contagion may be growing, most contagion still takes place at the regional level. For example, in the 1992-93 crisis in the Exchange Rate Mechanism (ERM) of the European Monetary System, it was mainly European industrial countries and emerging markets in central and eastern Europe that were involved. Similarly, the Mexican crisis had a Latin American focus, the Asian crisis an Asian focus, and so on.

Third, contagion is nonlinear. That is, whereas the presence of a single crisis somewhere else in the world or even in the same region only increases country X’s susceptibility slightly, once many crises have occurred elsewhere, it becomes quite unlikely that country X will emerge unscathed.

Fourth, contagion is not random. As in a flu epidemic, the weakest go first. Some stronger countries will also eventually be affected. More specifically, the prime candidates for contagion will be countries that have overvalued real exchange rates, large current account deficits, high shares of short-term debt, high ratios of liquid liabilities to liquid assets, slow growth or recession, and an important creditor in common with the country first affected.

And fifth, contagion is temporary. It rarely lasts longer than a year, and true contagion usually is over within several months.

High volatility of private capital flows and contagion of crises, then, have been facts of life in the 1990s. This raises two questions. What is driving this volatility and contagion? And if it is excessive, what is not being done now that could be done to reduce it?

As for what is driving the volatility, the short answer is lots of things—some good, some not so good. Underpinning the surge in inflows of the 1990s were some sizable improvements in economic fundamentals that made emerging economies more attractive to investors than they were in the late 1970s and early 1980s—better growth, inflation, and export performance; much reduced fiscal deficits; removal of the previous debt overhang; opening of capital accounts; more privatization and trade liberalization; and greater liquidity and wider choice of assets in developing-country securities markets. Improvements in credit ratings also expanded the investor base, and growing recognition that returns in emerging economies were weakly correlated with those in industrial countries strengthened diversification motives.

The not-so-good part is that the upsurge in flows to emerging economies sometimes also represents an overreaching “search for yield” that, in addition to fluctuations in industrial-country interest rates, is driven by unsustainable exchange rate and debt policies, or asset price bubbles, or poorly designed tax and investment breaks, or expectations of official bailouts.

As for the contagion, it too (as discussed in Section II) is driv-en by multiple factors. Part of it is attributable to common global or regional shocks (e.g., changes in world interest rates, primary-commodity prices, or G-3 exchange rates) that are largely beyond the control of an individual emerging economy. Part of it is attributable to trade spillovers (e.g., country A has a crisis-induced recession and therefore reduces its imports from countries B and C). Part of it results from financial linkages of various types. For example, on the eve of the Thai crisis, over half of the foreign liabilities of the Thai banking system were to Japanese banks, and Thailand represented over 20 percent of the emerging-market exposure of Japanese banks. Reflecting losses on both their domestic and external loans, Japanese banks withdrew $21 billion from the five East Asia crisis countries in 1997 (versus a $50 billion inflow in 1996). Finally, part of it simply reflects sharp changes in investor sentiment.

What this list of causal factors tells us is that some degree of volatility in private capital flows (and in asset prices) and some contagion across countries and markets are natural; they reflect the ways market participants react to new information and changing opportunities, and the way countries and markets interact with one another. The concern therefore is with excessive volatility and contagion. In this sense, an assumption behind the push for reform of the architecture is that whatever is “normal” for volatility and contagion, the experience of the 1990s goes well beyond that.

In addition to the vulnerabilities discussed earlier in this section, concern about excess volatility and contagion focuses on four problems: weak defenses against short-term capital inflows in emerging economies, existing policies that have the effect (intended or otherwise) of encouraging short-term capital flows, poor credit and risk assessment on the part of industrial-country lenders, and inadequate emergency liquidity support for actual or potential victims of contagion.

Within the overall volatility of private capital flows to emerging economies, short-term flows provide a particular risk because their short maturity makes it easier for investors to run at the first hint of trouble. One commentator has likened them to cars sitting in the parking lot with the engines running. Some support for that characterization comes from empirical tests that show that portfolio capital flows (bonds and portfolio equity) into emerging economies over the past 15 years have been considerably more volatile than flows of foreign direct investment (FDI), and that short-term capital flows have responded more dramatically to financial disturbances than has FDI. Behavior during the Asian/global crisis also supports this conclusion: whereas net FDI flows to emerging markets increased from $119 billion in 1996 to $135 billion in 1998, portfolio flows fell from $81 billion to $36 billion, and bank loans from $13 billion to minus $109 billion. Among the Asian crisis countries, only in Indonesia did FDI register a sharp decline. One explanation for this finding is that many important determinants of FDI (e.g., natural endowments, supply of human and physical capital, infrastructure, etc.) are not necessarily disrupted by a financial crisis.

This concern that short-term capital flows generate negative externalities for the host economy that are not taken into account in individual borrowing decisions has prompted some emerging economies to impose holding-period taxes on inflows. Chile’s system, which utilizes an unremunerated reserve requirement at the central bank for inflows that stay less than a given time period, has stoked the most interest. 17 Such a tax discourages short-term inflows but leaves long-term inflows unaffected; it also implicitly penalizes foreign lenders that do a lot of short-term in-and-out trading of emerging-market securities.

The available empirical evidence suggests that Chile’s holding taxes have tilted the composition of inflows in the intended direction, that is, toward longer-term flows, which are less crisis prone. It also appears, by driving a wedge between domestic and foreign interest rates, to offer some additional scope for independent monetary policy. On the other hand, there is less evidence that it has either reduced the total capital inflow or affected Chile’s exchange rate. Moreover, the effectiveness of the measures appears to erode over time, and it generates some distortions (e.g., sectors more dependent on bank finance are relatively hard hit). Other drawbacks are that the system can discourage certain types of short-term inflows (e.g., trade credit) that are desirable (either on their own or as a complement to FDI), and that it could act as an instrument of corruption in some economies.

Still, when all is said and done, such holding taxes on short-term inflows offer some defense against the surges in short-term inflows and ill-prepared financial liberalization that have had so much to do with financial crises in emerging economies in the 1990s. The IMF, which was strongly opposed to such capital-inflow taxes in the 1970s and 1980s, has recently adopted a more permissive tone. But neither the IMF nor the G-7 countries have been willing as yet to recommend them to emerging economies that need them, and relatively few emerging economies now have such holding-period taxes in place.

While some economies have not done enough to discourage short-term inflows, others have actually encouraged them. One of the reasons South Korea experienced such a flood of short-term inflows was that it had discriminatory controls that kept long-term investment out.

Another disincentive to inflows of longer-term capital comes from an unlikely source, namely, the risk-weighting system for commercial bank assets that is embedded in the 1988 Basle Capital Accord. In this system, which sets minimum capital requirements for the world’s internationally active banks, short-term claims on banks from any country carry a relatively low risk weight, thereby encouraging banks to engage in interbank lending. This is a double whammy. Not only can short-term borrowing generate a negative externality (for reasons outlined above), but also moral hazard problems are more serious for interbank lending, since governments are loath to permit a default in this market. In the proposed revision of the Basle Capital Accord issued for comment in June 1999, this bias toward short-term interbank lending has been reduced but not eliminated. 18

To the extent that excess volatility in private capital flows reflects poor risk management, borrowers in emerging economies do not have a monopoly on it. Lenders in the major industrial countries have done their part too. It is clear that during the run-up to the Asian crisis, many lending officers did not know their customers, many investors gave little weight to published information on the buildup of large foreign currency exposures, and stress tests and scenario analyses of market risks either were not up to the job or were not given much weight by senior management (or both). Supervisors of banks and securities houses in the major industrial countries ought to be asking themselves (and those they regulate), “Why?”

In the past several years, supervisors in the G-10 countries have allowed banks to rely more heavily on their internal models for assessing market risk. To the extent that those models have done poorly in the Asian and Russian crises, there is a serious question of what went wrong. Specifically, it needs to be determined whether the failure of many risk-management systems during 1997 and 1998 was simply the result of extreme observations of volatility that could not reasonably have been anticipated on the basis of the relevant asset price history, or whether there were also other deficiencies in these models. Similarly, G-10 supervisors ought to be asking themselves too whether the lessons of the recent crisis for risk management would be driven home better if banks had been following a “precommitment” approach to market risk (where penalties or increased capital charges were imposed on banks whose losses exceeded their value-at-risk calculations) rather than the present approach.

Special concerns have also been voiced about the activities of hedge funds, particularly in the wake of the near-collapse of LTCM. One worry is that these firms possess enough market power to destabilize currency, bond, and equity markets—especially in emerging economies—and have done so even when the economic fundamentals in these economies were sound.

Several studies have recently been completed on the role of hedge funds in episodes of market turbulence, and the results do not support this contention. Specifically, the studies find that hedge funds had a leading role in the ERM crisis of 1992-93 but were responding to market fundamentals; that domestic residents, rather than hedge funds, were the dominant players in the Mexican peso crisis of 1994-95; that hedge funds participated in but were not the dominant players in the large increase in the yen carry trade; that they also had significant short positions on the Thai baht but were at the rear of the herd rather than the lead when the selling accelerated; and that they did not take collectively significant short positions on any other Asian crisis currencies in the summer of 1997. On this charge, not guilty. Further study of the past activities of hedge funds is apparently being undertaken by the newly formed Financial Stability Forum, a group composed of the senior financial regulators from the industrial countries and the international financial institutions.

A second assertion is that hedge funds are “manipulating” the international financial system when they engage in speculative “double plays.” The leading example was the simultaneous short positions on the Hong Kong dollar and the Hong Kong stock market that hedge funds and other large players took in 1998. The assumption behind such a trading strategy is that when the monetary authorities raise interest rates to defend the currency, those interest rate increases will depress the property and equity markets, thereby making a winner out of the short equity position. Gains on the short equity position therefore help finance the siege against the currency. In the end, the Hong Kong authorities decided (in contradiction to their free-market principles) to fight those double plays by intervening in the stock market itself (so as to frustrate the equity short sellers).

Such double plays, especially if they are well financed, certainly make it harder for authorities to defend a fixed exchange rate. But so long as these markets are open to all comers, and so long as hedge funds do not “collude” with one another in their position taking, it is hard to see why such trading strategies should be regarded as “manipulation.” Double plays can, after all, be taken on a variety of markets. Hence, unless there is a challenge to the integrity of markets, savvy trading strategies should not be confused with manipulation.

A third worry is that the kind of “character lending” and “large customer bias” that allowed LTCM’s risk profile to get out of hand could happen again, with unhappy consequences not only for its lenders but also for those markets in which large positions might have to be liquidated into a falling market over a short period. This is a real concern.

To prevent a repetition of the LTCM affair, regulators in the G-10 countries have begun to lean on the lenders to highly leveraged institutions (HLIs) to, among other things, tighten up their counterparty risk management and to disclose publicly their financial exposures to all HLIs, including hedge funds. In addition, legislation will apparently be drafted (at least in the United States) that would require hedge funds to improve the frequency and quality of public disclosure about their activities.

A tightening of counterparty risk management is needed because risk management in the over-the-counter markets is governed by more discretionary procedures than it is on official futures exchanges, where there are rules for margin requirements and formal loss-sharing arrangements to limit systemic risk. A very large hedge fund that is a good customer of the lenders and that has had a recent string of good results may not be monitored as closely as it should be. Since the main lenders to hedge funds (large commercial banks and large securities houses) are regulated, supervisors can “indirectly” police hedge funds by leaning harder on their lenders to do a better job of risk management. This in turn may increase the cost of borrowing to hedge funds and thereby help to reduce their leverage (and possibly volatility as well).

While publication of lenders’ aggregate exposure to hedge funds may dissuade undue concentration of exposures, the effects of improved disclosure by hedge funds themselves are not easy to gauge. Some central banks have made a level-playing-field argument for it—that is, if new financial-data standards are to compel them to publish data on their net international reserves with high frequency, than those they face on the other side of the foreign exchange market (including hedge funds) should be required to do the same. Put in other words, I’ll show you mine only if you show me yours. The rub is that because hedge funds (like proprietary trading departments of banks and securities houses) can change their positions quickly, a disclosure statement that is, say, two to three months old may not tell you much. Also, because hedge funds sometimes have a leadership role, one cannot rule out the possibility that publication of their positions could induce more “copy- cat” trading. Still, more information would probably be helpful.

Thus far, G-10 officials have shied away from more direct approaches to regulating hedge funds (e.g., setting capital requirements for them), because they were worried that this would only induce the funds to move more of their activities offshore. This is part of the larger problem of dealing with offshore financial centers where financial supervision is weak.

But there is a potential weapon that has not yet been unsheathed. G-10 supervisors could set higher risk weights (for banks’ regulatory capital) on lending to offshore centers that do not implement minimum international financial standards (including public disclosure). So long as financial firms located in regulation-shirking offshore centers fund themselves in G-10 markets, this would increase their cost of capital and provide an incentive to meet these standards. There is also an equity issue at stake. If financial standards are going to become the new “rules of the road,” and if emerging economies are being called upon to observe them, then emerging economies might well ask that those who operate in their financial markets do no less (or at least pay a penalty if they do less).

Next, contagion. We have argued that emerging economies and their private creditors should carry the primary responsibility for crisis prevention and resolution in emerging markets. But central banks in the national context and the IMF in the international context also carry an important responsibility to prevent or contain financial crises (including limiting the scope of contagion). Indeed, the rationale for a lender of last resort is that it is capable of lending to solvent but illiquid borrowers when no other lender is capable or willing to lend in sufficient volume to prevent or to end a financial panic. The IMF is not a true lender of last resort because, unlike national central banks, it cannot create base money; nevertheless, its access to a pool of financial resources and its mandate to help solve balance of payments crises and to oversee the functioning of the international monetary system enables it carry out emergency lending functions.

In addition to its normal lending windows, the IMF has recently added two facilities that have a contagion orientation. One is the Supplementary Reserve Facility (SRF). The SRF was established in December 1997 to deal with situations in which large capital outflows from a country create a risk of contagion and threaten the stability of the international monetary system. The SRF must be used in conjunction with an existing IMF program, and its financing essentially permits the country to obtain larger than normal access to IMF resources. The SRF carries penalty interest rates (relative to normal borrowing cost from the Fund) and a short repayment period. Funds are dispensed in two tranches, the first of which comes on approval. The SRF was used in conjunction with recent IMF-led rescue packages for South Korea and Brazil.

In contrast to the SRF, which applies to countries in the throes of a crisis, the newly established (April 1999) Contingency Credit Line (CCL) is to be used as a preventive measure for countries concerned with potential vulnerability to contagion but not facing a crisis at the time of commitment. It too provides for exceptional access to IMF resources (expected to be in the range of 300-500 percent of Fund quota but higher in exceptional circumstances); it carries above-normal interest rates and a short repayment period, and its disbursements can be front-loaded. Countries qualify for using the CCL on the basis of, inter alia, good policies, progress toward meeting internationally accepted standards, and submission of a quantitative framework laying out future economic and financial policies.

The CCL is authorized for a two-year period, and experience with it is to be reviewed after the first year of operation. The IMF’s Executive Board has acknowledged that it needs to be approached in an experimental way.

In thinking about appropriate responses to cross-border contagion, there are at least four key operational issues to settle: whether prequalification should be used, whether there should be a systemic threshold for activation, whether to apply policy conditionality, and how funding for the contagion facility should be handled.

The main argument for prequalification—that is, making countries eligible for a line of credit before they ask for it—is that the knowledge that a country has qualified for a sizable line of credit may deter a speculative attack. Objections to prequalification can be made on three counts.

One is that earlier efforts to ward off attacks by enlarging the war chest and then opening its lid have had at best mixed success. Both the United Kingdom and Sweden arranged—and publicized—contingent credit lines from private banks in 1992; but in neither case did this succeed in deterring large (and subsequently successful) attacks on their currencies in the fall. Hong Kong had $60-100 billion of reserves in 1997-98 and pledges of support from Beijing; yet it too faced strong attacks on the currency during that period. Given the potential resources on the private side of the market, the deterrent effect of prequalification may be limited.

A second objection to prequalification is that it makes it harder to disqualify a country if it behaves badly between the time the credit line is committed and the time it is drawn. For example, if Brazil had been prequalified to draw in, say, the beginning of 1998 but was regarded as a more risky borrower in the late summer, would the IMF really have been willing to deny the line?

A third objection has to do with competition with contingent credit lines offered by the private sector. If one wants emerging economies to rely more in the future on their own resources and on private sources of liquidity, then care must be taken that the IMF not offer a sweeter deal. Admittedly, there are potential problems with the private contingent credit lines (e.g., creditors may reduce exposure in other areas when they are drawn, they may seek to renegotiate tougher terms when the lines are called, etc.), but this market needs time to develop and may not develop if the public sector undercuts it deeply.

The CCL tries to have it both ways on prequalification. While countries qualify for access to the CCL on the basis of the criteria outlined above, they cannot draw the credit line until the IMF’s Executive Board conducts an “activation” review to determine that the country is severely affected by contagion and intends to adjust its policies as needed. What is probably more important than prequalification for combating a large, sudden capital outflow is a quick decision and up-front disbursement.

Both the SRF and the IMF’s $45 billion credit line from 25 creditor countries—the New Arrangement to Borrow (NAB) and the General Arrangement to Borrow (GAB)—have a “systemic threshold” for activation. That is, they can be drawn only when the situation is one that threatens the stability of the international monetary system. The CCL does not have that feature.

In our view, a systemic threshold is desirable in a contagion facility. Most episodes of contagion will not be systemic and can and should be handled by drawing on the country’s own reserves, by drawing on private-sector contingent credit lines, and by drawing on Fund resources within the normal access limits. To activate a backup contagion facility, the situation should be one in which adverse capital account and/or commodity price developments are affecting many emerging economies simultaneously, in which private creditors are not distinguishing appropriately between creditworthy and less creditworthy borrowers, and in which failure to go beyond the Fund’s normal access limits threatens a significant deterioration in the performance of the world economy. In other words, it ought to be reserved for systemic threats such as those prevailing in the fall of 1998 (the very early stages of the tequila effect of the Mexican crisis also might have qualified). Note that Argentina, Panama, and Venezuela were all able to issue 30-year bonds in September 1997, that Latin American borrowers maintained fairly decent access into the first half of 1998, and that Argentina and Mexico experienced only relatively brief loss of market access after the Brazilian crisis; they paid much more, but they retained access. Systemic ought to mean systemic. In country-crisis cases, the normal access limits should apply.

Turning to policy conditionality, we noted above that contagion sometimes operates via global shocks (changes in world primary commodity prices, changes in G-3 exchange rates, changes in risk premia to all emerging economies, etc.) that are largely beyond the control of individual emerging economies. Also, most episodes of contagion are short-lived. If contagion really is largely beyond the country’s control and is expected to be temporary, then it is hard to see the need for an IMF program or for upper-credit-tranche conditionality. We say “largely” because we should not expect perfect policy behavior by countries to qualify. On the other hand, if the country’s loss of market access is largely of its own making, then that country should not be permitted to draw on a contagion facility; it should instead approach the IMF for a normal standby program with normal Fund conditionality. Here too, the CCL tries to play it both ways. It is supposed to apply to contagion that is largely beyond the country’s control and consequent upon developments in other countries. But drawing the credit lines requires the country to accept policy conditionality similar to that in upper-credit-tranche Fund programs.

Finally, there is the issue of adequate funding for a contagion facility. The CCL is to be funded out of the IMF’s existing resources. One explanation for that decision is that the deterrent effect of the CCL (with the IMF’s “good housekeeping seal” of approval for qualified borrowers) will be so strong as to preclude the need for any new money. We have argued, too, that the IMF could (and should) become smaller—but only if it reduced its financial support for overvalued fixed exchange rates and if its member countries made greater use of private debt rescheduling under appropriate circumstances. We doubt that the deterrent effect of the CCL would make it unnecessary to prepare for rare but extremely serious cases of systemic contagion.

Nor are we convinced that the quota enlargement process will always go smoothly. Last year, for example, the IMF’s quota increase was blocked for a long time because of an acrimonious debate in the US Congress over IMF policies—and this at the same time that the global economic crisis was raging and the IMF’s coffers were running low. While many of the issues in that congressional debate were significant, we should not allow such national debates to jeopardize the international community’s ability to respond to true systemic threats. For this reason, we suggest in Section IV that funding for a systemic contagion facility might best be assured by a special allocation of IMF Special Drawing Rights. You should only take the big hook and ladder engine out of the firehouse for big fires—but when you take it out, it needs to be fueled up and in good working order.

When the time comes to review the CCL, in May 2000, it (along with the SRF) should be replaced with a new contagion facility that would be simpler to operate, would have a more systemic focus, and would be assured of adequate funding (along the lines laid out in Section IV).

(8) Threats to the effectiveness of and popular support for the IMF and the World Bank. There is good news and bad news about the Bretton Woods institutions and the global economic crisis. Before the crisis, many Americans knew or had heard little about what the IMF and the World Bank do. The good news is that this is no longer the case. The bad news is that much of what they have heard over the past two years is that the Bretton Woods twins, and particularly the IMF, have been more part of the problem than part of the solution. Since the IMF is a key crisis lender and manager in the international monetary system and since the World Bank has a central role in combating poverty and promoting sustainable economic development, it is important to consider threats to their effectiveness and to their popular support. We will devote more attention here to the activities of the IMF because it has been at the center of controversy.

One line of criticism is that the IMF is no longer needed. Those who support this position point out that when the IMF was created after World War II, it was put in charge of overseeing an international monetary system that presupposed fixed exchange rates. But today’s exchange rate system is predominantly one of floating exchange rates, and management of floating exchange rates should be left mainly to the private markets.

But the real purpose of the IMF is not defined in terms of a particular exchange rate regime. Rather, its job is to help countries overcome balance of payments problems in a way that is not destructive of national and international prosperity. As costly as the Asian crisis has been, no doubt we would have seen even deeper recessions, more competitive devaluations, more defaults, and more resort to trade restrictions if no financial support had been provided by the IMF to the crisis countries. As discussed earlier, there can be legitimate differences of view about IMF advice on fiscal and monetary policy in the crisis countries. Maybe fiscal policy should have been eased earlier. On interest rate policy, there was no easy solution. But we had a look in the 1930s at how serious global instability is handled without an IMF, and few would want to return to that world.

In a similar vein, we do not assume in the United States that financial crises can be solved exclusively by private markets. We have a lender of last resort—the Federal Reserve—that can deal with private market excesses and collective-action problems before they get out of hand and result in a cascade of liquidity problems. The decisive action taken by the Federal Reserve right after the 1987 stock market crash is a case in point. For much the same reason, we need an IMF at the international level to deal with relatively infrequent and short-lived but nonetheless serious episodes of investor panic. Markets are better than anything else is at allocating resources, but they are not infallible.

It is well to keep in mind, too, that US contributions to the IMF do not affect either our government budget position or government spending for other programs. They represent an exchange of one reserve asset of our government for another. 19 And if we are going to assist our trading partners in times of crisis, it is better for us to do so as part of a multilateral effort, in which we can share any financial burdens with other creditor nations, than to do it on our own.

For all of these reasons, the case for abolishing the IMF is a weak one. But this does not mean that there is no scope for improving how the IMF goes about its business. As argued above, we think incentives for crisis prevention and resolution would be strengthened if the IMF reduced the size of its loans for country crises and if it insisted on greater burden-sharing by private creditors when debt burdens of emerging economies become unsustainable. In other words, moral hazard can be controlled by reforming the IMF’s lending practices, without closing the Fund down.

The IMF also needs to become more transparent in its own operations. In this regard, we welcome the release and posting on the IMF’s website of letters of intent and policy framework papers, of summings-up of IMF Executive Board meetings (called public information notices, or PINs), of the Fund’s level of financial resources and its liquidity position, and of credit extended to and debt payments made by member countries. We understand that countries’ forthcoming obligations to the IMF soon will also be posted there.

But there is another thorny issue that merits immediate attention in light of the experience of the Asian crisis, namely, the scope of IMF policy conditionality and advice.

Unlike a traditional lender of last resort, which provides liquidity assistance against marketable collateral, the Fund relies on its preferred credit status and on its policy conditionality to get its money back. The traditional focus of Fund conditionality is monetary, fiscal, and exchange rate policy. But the Fund, usually in cooperation with the World Bank, has increasingly over the past 20 years asked also for structural policy changes in return for its financial assistance. The rationale for that broadening of conditionality has been that structural policy problems often lay behind balance of payments crises and that providing assistance without addressing them would only deal with the symptoms—and not with the underlying causes—of the crisis.

This issue came to a head in IMF programs with the Asian crisis countries. Because financial-sector weaknesses and corporate governance problems loomed large in the origins of the Asian crisis, the Fund made structural policy reforms a key part of its conditionality. Could Thailand escape from its crisis without closing insolvent finance companies and banks? Could South Korea hope to recover without changing the way governments, banks, and the chaebol conducted business with one another and without reducing the high debt-to-equity ratios of the chaebol? Could Indonesia regain investors’ confidence without providing some indication that it was prepared to curtail inefficient infrastructure projects and rein in the worst cases of “crony capitalism”? On top of this, the scope of conditionality was widened further because reformers in some of the crisis countries saw the crisis as a unique opportunity to tackle some long-standing structural problems, and because some large official creditors used the crisis to extract concessions on some tariff and market-opening issues.

Critics argue that the IMF overstepped its mandate and its expertise in demanding such wide-ranging structural reforms. They argue that when the IMF contemplates a policy reform it should ask itself: Is this reform necessary to restore the country’s access to international capital markets? And would the Fund ask the same policy changes of a major industrial country if it were the subject of a Fund program? If the answer to either question is “no,” then that policy should not be part of the Fund program. Acting otherwise will—so the argument goes—only encourage countries to delay coming to the IMF until they have no alternative, will paint the IMF as insensitive to the cultural and social differences among emerging economies (with adverse effects on its popular support in those countries), and will weaken its reputation for competent, apolitical economic advice. In other words, unless emerging economies are prepared to “take ownership” of structural reforms, the prospects for success will be dim.

Subsequent events in the Asian crisis countries have not settled this debate. Several of the crisis countries have started to grow again and have begun to reestablish market access. Clearly, they have not corrected all of their financial-sector and corporate governance problems. On the other hand, they have made a nontrivial down payment on those reforms (without which the return of confidence might have been much further delayed).

There is at least one structural policy area, however, that would be hard to exclude from IMF policy advice and conditionality—banking and financial-sector policies. This is because banking and financial-sector policies generally have higher complementarity and interdependence with macroeconomic and exchange rate dimensions of crisis prevention and management than do other structural policies. Not only do banking weaknesses often act as a precipitating factor in subsequent currency crises, but also bank restructuring cannot easily be separated from the fiscal and monetary policy recommendations made during a crisis.

For example, one of the reasons the Mexican authorities were so reluctant to raise interest rates in early 1994 despite a marked drop-off in capital inflows was that overdue loans in the banking system had become a serious problem, and they were worried that a large interest rate increase could push the banks over the edge. Because of the huge fiscal costs of bank restructuring/capitalization, interactions between financial-sector reforms and monetary policy (e.g., the effects of bank capital requirements on bank lending), and the need to make quick decisions on closing insolvent banks, banking and financial-sector policies cannot easily be separated from the Fund’s more traditional policy focus.

The appropriate scope of IMF policy advice also touches on two broader, related issues that need to be settled. One of them is the scope of the Fund’s overall mandate. The other is the proper division of responsibility between the IMF and the World Bank.

Comparative advantage ought to apply to public institutions as well as to countries. Surely, no single international financial institution (be it the Fund, the Bank, or any new institution) can be expected to excel across a range of tasks that extends from monitoring financial standards, to building of financial infrastructure, to surveillance over the whole gamut of economic policies, to facilitator of debt rescheduling, to crisis manager and lender, to designer of social safety nets, to provider of technical assistance, to protector of the environment, and on and on—policeman, fireman, banker, adviser, arbitrator, teacher. There must be some choices and priorities among these roles. This is particularly so since the traditional separation of responsibilities between the IMF and the World Bank—that is, the Fund concentrates on macroeconomic policies and the Bank on longer-term structural and social aspects of development—seems to have become more blurred during the Asian crisis. For example, the Bank not only gave emergency financial assistance to South Korea in early 1998 but also has been commenting publicly on the appropriate macroeconomic policy mix in the crisis countries; at the same time, Fund programs with the Asian crisis countries have been reaching more deeply into a wide range of structural and social policies. Both institutions need to refocus on a leaner agenda.

Implications For Repairing The Architecture

As our survey of vulnerabilities to financial crises indicates, there are many areas of the existing architecture that are in need of repair. In this sense, we agree with former US Treasury Secretary Robert E. Rubin that there are no “magic wands” and that the task ahead is going to require a collection of actions over time. In the task force’s view, particular attention needs to be devoted to:

 

IV. Recommendations

In this section, we lay out the task force’s recommendations for improving the architecture. We have not tried to cover everything. Rather, we have concentrated on seven areas in which we think significant progress can be made over the next two to three years. Each of our recommendations revolves around a theme and encompasses several specific suggestions. After each recommendation, we offer a commentary that captures some of the task force’s deliberations on that issue. Following the recommendations is a summary of how our reform package differs from that proposed by the G-7 and the G-22.

The Task Force’s Reform Package

Recommendation 1: Greater Rewards for Countries That Join the “Good Housekeeping Club”

The aim is to improve the incentives for emerging economies to undertake greater crisis prevention efforts by forging a closer link between implementation of crisis prevention measures and the terms of emergency financial assistance from private capital markets and from the IMF.

  1. Much in the way that private insurance companies provide a better insurance deal to policyholders who reduce their risk (say, by installing smoke detectors), the IMF will henceforth relate the interest rate it charges member countries (that borrow from the Fund) to the strength of the country’s crisis prevention efforts.

  2. For this purpose, “crisis prevention efforts”—or “good housekeeping”—will be interpreted to mean sound macroeconomic policies; compliance with a set of international financial standards and good practices (i.e., the Fund’s Special Data Dissemination Standard, the Basle Committee’s Core Principles of Effective Banking Supervision, international accounting standards, the International Organization of Securities Commissions’ principles of securities regulation and international disclosure standards, deposit insurance reform along “structured early intervention and resolution” lines 20 ); maintenance of a “viable” currency regime; prudent debt management (particularly as regards the maturity and currency composition of external and domestic public debt); and efforts to put in place various (non-Fund) sources of liquidity support that could be activated in the event of a crisis (ranging from contingent credit lines from private financial institutions to a country’s holdings of international reserves).

  3. In order to encourage a larger private market payoff to crisis prevention efforts, the Fund will make public both a “standards report,” which assesses countries’ progress in meeting international financial standards, and the Article IV consultation report, which assesses countries’ overall economic policies and prospects, including macroeconomic policies and other elements of crisis prevention. For example, each IMF member country might be put into one of three categories: class A—currently complies with a core set of international financial standards; class B—is making good progress and has pledged to meet these standards within a three- to five-year period; class C—is in early stages of reform and/or has made no commitment to these standards. Other things being equal, class A countries would face lower Fund borrowing costs than class B countries, which in turn would face lower costs than class C countries.

  4. In revising the Basle Capital Accord for credit risk of banks and in setting bank provisioning guidelines, international and national financial regulators will take account of the Fund’s standards report and country classifications.

Commentary: Because private credit-rating agencies are probably subject to weaker political pressures than is the Fund, there would be an advantage in having private agencies incorporate compliance with international financial standards in their ratings. But we think there will need to be a transition period under which the IMF does the ratings. Some task force members argued that the incentive for greater crisis prevention efforts ought to be framed in terms of access levels to Fund resources rather than to the interest rate charged for Fund borrowing (since the latter could be deemed inconsistent with the IMF’s present charter). A few task force members urged going further by making countries that did not implement international financial standards ineligible for Fund loans. They felt that stronger penalties were necessary, in particular to bring public data disclosure at the international level up to the standards mandated for public issuance of securities in the United States. Most task force members, however, favored the interest rate channel.

There is also a question of whether an “insurance bonus” for good crisis prevention efforts should be a feature of all Fund lending windows or only of some of them. The new Contingency Credit Line (CCL) is supposed to take crisis prevention efforts into account as one of the factors determining eligibility. Drawings under the Fund’s regular lending windows do not take it into account. If the CCL were redesigned to eliminate prequalification and to respond only to systemic crises (as we suggest in recommendation 5), it would be better to consider crisis prevention efforts as a determinant of the borrowing cost for drawings in the regular lending windows. A number of task force members felt that IMF loans to low-income developing countries under the Enhanced Structural Adjustment Facility (ESAF) either should be subject to a smaller risk-based increase in Fund borrowing costs or should be exempt from such an increase.

Task force members emphasized that differences of view over methods should not be allowed to obscure the key principles that there needs to be a stronger link between crisis prevention efforts and the cost/availability of emergency financial assistance if the “good housekeeping club” of nations is to be expanded, and that greater progress on crisis prevention would reduce the need for official emergency assistance.

As regards the “standards report,” while the IMF may be the best vehicle for collecting and disseminating the information on compliance with international financial standards, it need not make all the assessments of compliance on its own; for example, whereas the Fund may be best placed to assess compliance with the data and banking standards, it might have to rely on other official or private professional groups to assess compliance with other standards (e.g., international accounting, securities regulation, etc.). Nevertheless, there would be merit in collecting such assessments in one place and publishing the results. If the results of the IMF’s assessments are not made public, one loses the key incentive mechanism for complying with the standards. Indeed, most task force members felt that the potential “market payoff” is probably larger than the potential payoff from lower IMF interest rates, larger access to IMF resources, or preferred risk weights for regulatory capital.

It should be acknowledged that there will certainly be “gray areas” in trying to reach judgments about compliance with financial standards and about the appropriate rating/classification category for particular countries. Several task force members thought that a standards report should be issued periodically (not necessarily annually) and that it should evaluate how close countries were to objectives specified earlier in a plan of action.

Several task force members, while not opposing the promulgation and monitoring of international financial standards, argued that they were likely to have only a small impact on strengthening emerging-market financial systems. They felt that foreign ownership of banks might have a greater impact on improving risk management but acknowledged that there were political limits to how far that process could go. Several members maintained that a lack of training and expertise remained a key bottleneck in implementing the standards, and that what was needed was less “lecturing” of emerging economies for their past errors and more training of personnel to prepare for the new challenges of globalization. Notwithstanding very useful efforts to improve training for banking supervision (e.g., by the BIS’s Financial Stability Institute) and to increase incentives for complying with financial standards, some members concluded that the emergence of resilient financial sectors in many emerging economies was going to take a long time.

Another view was that the push for international standards should not go so far as to prevent emerging economies from pursuing lending policies that served particular social goals, such as better income distribution for backward areas, assistance to certain minority groups, or infrastructure projects that support overall national development. A number of task force members, however, were concerned that this could easily become a rationale for reintroducing crisis-prone lending policies.

Task force members regarded the pilot programs for the publishing of IMF Article IV reports on a voluntary basis (that is, with the member country’s consent) and for issuing “transparency reports” as a step in the right direction. Most felt that publication of these reports should be converted to an obligation of IMF membership as soon as possible, and that transparency reports should be expanded into “standards reports” along the lines suggested above.

An advantage of private contingent credit lines is that countries might be willing to draw on them before they get into deep trouble. What is meant by a “viable” currency regime is discussed in recommendation 4, but here it is sufficient to say that an “adjustable peg” currency regime would be regarded as particularly crisis-prone for emerging economies.

Recommendation 2: Open Capital Markets, Yes, But Guard against “Too Much of a Good Thing”

The objective is to moderate the “boom-bust” cycle in private capital flows and to tilt the composition toward less crisis-prone flows, while still preserving most of the benefits of greater market access.

  1. The IMF should not merely permit holding-period taxes of the Chilean type on short-term capital inflows but should advise all emerging economies with fragile domestic financial sectors and weak prudential frameworks to implement such measures (until their ability to successfully intermediate such flows has become stronger).

  2. Where such penalties/taxes on short-term inflows are applied, they should be transparent and nondiscriminatory, should be price rather than quantity oriented (akin to tariffs rather than quantitative controls), should not be used as a substitute for undertaking corrective action on weak policy fundamentals, and should not impede the entry of foreign financial institutions into the financial services industry.

  3. The IMF should intensify its surveillance over countries’ public debt management, with a view toward discouraging heavy reliance on short-term, foreign-currency-denominated debt as well as the inclusion of put options in medium and long-term debt instruments.

  4. In revising the Basle Capital Adequacy Framework, financial regulators should avoid weighting schemes that provide incentives for short-maturity cross-border flows or for interbank lending.

  5. While ongoing efforts to strengthen “indirect” approaches to regulating highly leveraged institutions (HLIs) deserve a fair trial (e.g., tightening risk-management guidelines for the banks and securities houses that lend to hedge funds, publishing data on lenders’ exposure to HLIs, etc.), financial regulators should be prepared to consider more direct regulatory initiatives if those indirect approaches do not bear fruit. In particular, they should consider imposing a higher capital charge (risk weight) for bank loans that go to offshore financial centers that do not meet minimum international financial standards (including public disclosure guidelines).

Commentary: By imposing a higher tax/penalty on capital flows that stay in the country for less than a specified time period (say, one year), such a capital-flow regime not only discourages flows of short-term capital (and encourages longer-term ones) but also implicitly taxes at a higher rate those types of financial institutions (including many HLIs) that engage in a lot of “in-out” trading. In contrast, nonfinancial and financial institutions that rely more on foreign direct investment and other capital flows with a relatively long time horizon should not be adversely affected. Also, by specifying that these taxes on short-term inflows should apply only to countries with relatively fragile financial sectors and should be maintained only during the transition to stronger local banking systems, this proposal avoids a “one size fits all” approach. In this connection, such measures can be regarded as not prejudicing the longer-term drive for capital account liberalization (when more emerging economies are “ready” for it).

Some task force members argued that taxes on capital inflows should focus on short-term borrowing by financial institutions. Others, however, felt that if holding-period taxes were going to be effective, they would need to be imposed across the board (so as to limit substitution from the taxed to the untaxed categories of inflows). Still others were concerned that a comprehensive tax would hit some “good” short-term inflows (such as trade credit) that were complementary to longer-term flows (such as FDI).

A number of task force members took the view that holding-period taxes did not need to be the only game in town. Some would not exclude, for example, the use of outright position limits on short-term foreign-currency debt (rather than taxes). A few argued that one should not draw too strong a distinction between holding-period taxes on capital inflows and exit taxes on capital outflows. The latter would be market-oriented too, and would have the same effect on an investor anticipating a short “round-trip.”

Some task force members were opposed to any blanket recommendation for holding-period taxes on capital inflows. They argued that emerging economies should decide for themselves if the extra cost of borrowing associated with such taxes was worth the risk reduction. In addition, there was a danger that inflow taxes could be expanded beyond their original scope. They were easy to implement but difficult to take off. More fundamentally, some members felt that emerging economies would do better instead to follow Argentina’s lead and bolster their “liquidity” defenses against volatility (rather than to rely on inflow taxes to reduce that volatility). A few argued that capital-inflow taxes had in fact played only a minor role in Chile’s good performance.

At least one task force member took the view that rarely—but not never—controls on outward capital movements may be necessary. Capital-inflow restrictions and measures to induce greater private-sector burden-sharing addressed only one form of capital flight: that which involves foreign-currency debt. While this was a key aspect of the capital account reversal in the Asian crisis, debt flows are only one of many channels of capital movement. Massive capital flight could also occur as domestic investors sell local assets and convert them into dollars. If one takes the principle of burden-sharing seriously, one has to concede that there may be times when it would have to extend to all potential sources of flight capital, not just to holders of short-term debt; this, in turn, means that controls on outward capital movements could not always be ruled out-of-bounds.

Several task force members, in analyzing the volatility of private capital flows to emerging economies, emphasized the funda-mental asymmetry (outlined in Section II) between the size of emerging-market financial systems and the positions that large, highly leveraged institutional investors could take in those markets. These economies had engaged in a “devil’s bargain” over the past decade or so. They wanted greater liquidity and a lower cost of borrowing, and they turned to the international capital markets—especially to large, institutional investors—to give it to them. They got it. But along with it came large foreign players with huge resources that gave them enough market power to significantly affect asset prices when they moved in and out as a group. It is the medium-sized and larger emerging economies that are most susceptible to this asymmetry. There is not enough liquidity in the smallest economies to attract the big players, while the largest financial markets (the yen/dollar market or the market for US government securities) still resemble the “atomistic” model of competition. 21 In controlled settings, such as government bond auctions in the major industrial countries, it is possible to ensure that no single player gets an undue share of the market. But this is much tougher to police in secondary markets and when there are a large number of players. Regulatory changes that affect the cost of borrowing for the HLIs will reduce leverage and this asymmetry a little but will not eliminate it. Some task force members emphasized that HLIs should be promptly disclosing balance-sheet and position information to their counterparties whether the HLIs were local or located offshore, and that the regulatory system should penalize HLIs that were not doing so.

A reason not to give short-term loans a very low risk weight in calculating regulatory capital requirements is that the liquidity advantages that are available to a single lender can evaporate in a crisis when the borrower has to roll over a large share of debt and when all short-term lenders want to exit simultaneously. A concern about encouraging interbank lending (with low risk weights) is that the “special” role of banks will make it harder to reschedule interbank loans during a crisis and this in turn will lead to excessive lender moral hazard.

Recommendation 3: The Private Sector: Promoting Fair Burden-Sharing and Market Discipline

The objective should be to increase the timeliness and orderliness of debt rescheduling and to reduce the moral hazard for private lenders associated with national and international financial rescue packages.

  1. To increase the orderliness and timeliness of debt rescheduling, all countries—including particularly the G-7 countries—should commit to including “collective action clauses” in their sovereign bond contracts and to requiring that such clauses be present in all new sovereign bonds issued and traded in their markets.?

  2. For similar reasons, the IMF and the G-7 countries should use their influence to encourage the formation of standing steering committees for holders of emerging-market bonds and bank loans and work with these committees during future debt rescheduling negotiations. The Fund should also encourage emerging economies to maintain a current and comprehensive register of their external creditors.

  3. To deal with moral hazard at the national level, the IMF should advise emerging economies to adopt a “structured early intervention and resolution” approach to deposit insurance reform in the banking system, and reward countries that do so (along the lines laid out in recommendation 1 above).

  4. To address moral hazard problems at the international level, the IMF should make it known that it will provide emergency financial assistance only when there is a good prospect of the recipient country’s achieving “balance of payments viability” in the medium term and that such viability requires a sustainable debt and debt-servicing profile.

  5. In extreme cases in which the existing debt profile is clearly unsustainable, the IMF would expect, as a condition for its support, debtors to engage in “good faith” discussions with their private creditors with the aim of reaching timely agreement on a more sustainable debt and debt-servicing profile. No category of debt would be presumed to be exempt from these discussions.

  6. In such cases, the IMF would recognize that orderly debt rescheduling may be facilitated by having the debtor declare a temporary payments standstill (say, of 30-60 days); the final decision to impose the standstill would rest with the debtor country.

  7. When such a payments standstill is imposed, the IMF would encourage debtors to seek an agreement that is nondiscriminatory between foreign and domestic holders of debt and to provide creditor banks with timely and reliable information on interbank exposures to the country. Likewise, the Fund would encourage creditor banks to maintain interbank lines and to refrain from any legal challenges during the period of the standstill.

Commentary: As argued earlier (in Section III), the way to avoid a negative signal being associated with collective-action clauses is for the most creditworthy countries (the G-7 countries) to include them in their own sovereign bond contracts. It is hard to imagine, for example, that such a decision would raise the US government’s cost of borrowing. But still, no US secretary of the treasury relishes the thought of doing it on his watch—even if it would assist debt rescheduling in the developing world.

A few task force members argued that even if collective-action clauses were included in all sovereign bond contracts, they could still increase the cost of borrowing for emerging economies. Others disputed that view. Such clauses provide greater certainty, and in the long run this would reduce spreads. It was hoped that greater use of collective-action clauses in sovereign bond contracts would spread to private debt instruments.

Some task force members noted that organizing standing steering committees was apt to be more difficult in the bond market than for bank loans. Who would call the meeting? Who would take the temperature of the investors? Would the committee members be only those who owned the bonds when the problems occur? Perhaps the original underwriter could be saddled with that responsibility.

A large majority of task force members supported the IMF’s “lending into arrears” policy but felt that the time had come to move beyond it to obtain better private-sector burden-sharing. Under the policy proposed above, the IMF would ask as a condition for its support (in extreme debt overhang cases) that the country agree to a debt rescheduling with private creditors—much in the same spirit that official creditors in the Paris Club require “comparable” concessions from private creditors (as a burden-sharing quid pro quo) for their own debt rescheduling concessions. While the precise terms and modalities of that rescheduling would be left to the debtor and creditor to work out, the final outcome would have to fall in a range that was consistent with the Fund’s view of a “sustainable” debt and debt-service profile (if the debtor wanted to get the IMF loan). This proviso is important because otherwise private creditors and official debtors can “game” the system by agreeing to rescheduling terms that are good for them but not for official creditors (or for the debtor in the long run).

Most task force members argued that private-sector burden-sharing should be thought of in terms of accepting their fair share of the losses associated with resolving the crisis. For banks, maintaining loan exposures or adding new loans could be consistent with such burden-sharing, but only if the terms of those commitments translated (in present discounted value terms) into equivalent losses. It should also be clear that burden-sharing would apply to private-sector debt as well as to sovereign debt.

If one wanted to give the IMF the legal authority to sanction (rather than just to recommend) a temporary stay on creditor litigation, it would probably be necessary to amend Article VIII, Section 2(b) of the IMF’s Articles of Agreement. However, given the time-consuming process of amending the Articles, it would be preferable to give the informal approach a trial. The intent should not be to permanently affect creditors’ rights but rather to provide a “breathing space” to bring more order into the negotiation process. It was acknowledged that there are many thorny operational questions associated with debt rescheduling (for example, how to treat outstanding derivative contracts when the underlying instrument is the subject of a standstill), but most of these would have to be handled on a case-by-case basis.

The statement that no category of debt would be automatically exempt from rescheduling is meant to apply to bank loans and bonds—not to foreign direct investment or equities.

A few task force members thought that while these proposals (and those contained in recommendation 2) would reduce lender moral hazard and help to moderate the “boom” in private capital flows, they would not deal with the “bust” part of the cycle. They saw a real danger in the period ahead that spontaneous private capital flows to emerging economies would be too small and/or that these flows would only be provided at rates that were too high to allow producers in emerging economies to compete effectively in the global market.

One view was that the best way to deal with this problem would be to grant either an exemption from debt rescheduling or a partial guarantee to sovereign borrowers that are judged by the IMF to have qualified for membership in the “good housekeeping club” (as defined in recommendation 1 above). This would keep the volume of financing to well-behaving emerging markets from dipping too low (and the cost from rising too high), and it would also provide a bigger “carrot” for crisis prevention than the other incentives mentioned earlier. Most task force members, however, were opposed to having the official sector offer new guarantees or exemptions from rescheduling on moral hazard grounds. Moreover, they thought that there would be enough “searching for yield” and enough profitable investment opportunities to generate (without any guarantees) good-sized capital flows on reasonable terms to emerging economies the years ahead.

Another view was that the task force’s recommendations overemphasized the role of lender moral hazard, incorrectly singled out commercial banks for inadequate burden-sharing, exaggerated the need for mandatory changes in bond contracts, and underplayed the importance of maintaining market access for emerging economies. Reflecting the mass securitization of emerging-market debt, new players such as mutual funds and pension funds have increased in importance. The IMF should only support temporary halts in debt payments within the context of overall restructuring or refinancing. Debt rescheduling should be seen as a cooperative exercise—not as a zero-sum game. Collective-action clauses should be included in bond contracts only on a voluntary, case-by-case basis; otherwise, their forced inclusion could reduce the demand for these instruments at a time when it was desirable to encourage capital flows to emerging economies. If the IMF changed its lending policy along the lines suggested above, developing countries would be the losers, because they would face a lower supply and higher cost of external finance.

Most task force members, however, felt that the balance between limiting systemic risk and encouraging market discipline had tilted in recent years too far away from market discipline. Unless that balance was restored, we would not be successful either in deterring future crises or in garnering popular support for official rescue packages. The IMF could and should help in facilitating adjustment to balance of payments problems and in stepping in when there is widespread investor panic, but it should not be in the business of guaranteeing de facto short-term debt. And developing countries would be better off in the long term if they recognized the need to bolster their defenses against a volatile international capital market.

Recommendation 4: The Currency Regime: Just Say No to Supporting Pegged Exchange Rates

The objective should be to reduce crisis vulnerability and improve overall economic performance by making better exchange rate policy choices in emerging economies.

  1. In the advice that the IMF offers emerging economies, both in its Article IV missions and during the negotiation of IMF programs, the Fund should counsel against adopting a currency regime based on an adjustable peg, and should place strict limits on the financial support it extends to defend fixed exchange rates that are arguably overvalued.

  2. In most circumstances, the IMF should encourage emerging economies to adopt and to maintain a currency regime of “managed floating” (where managed floating runs from floating with “leaning against the wind” intervention to regimes with crawling exchange rate bands).

  3. The Fund should also be willing to support the establishment and maintenance of currency boards in those unusual circumstances where alternative currency arrangements are unlikely to restore monetary policy discipline and where the preconditions for a currency board’s effectiveness and sustainability are met.

  4. Over the longer term, emerging economies should consider the benefits and costs of reducing their crisis vulnerability by adopting one of the major reserve currencies (e.g., dollarization), but at this point the factors that have led to agreements on monetary union (as in the euro zone) do not appear to be present elsewhere.

Commentary: There is no single currency regime that will perform best for all emerging economies in all circumstances. On the other hand, having the IMF take a neutral position on the choice of currency regime in developing countries (i.e., let a thousand flowers bloom) would seem to ignore the lessons of the Mexican, Brazilian, and Russian crises (among many others). The position sketched out above—that managed floating should be the mainline recommendation, that adjustable pegs should be avoided, that currency boards should be reserved for particular situations, and that managed floating does not prejudice a move to single currencies later on, if certain conditions are met—is an attempt to select the lesser of many evils.

Implicit in the recommendation for a managed floating regime in emerging economies is the assumption that greater recognition of currency risk will lead private-sector participants to make better use of hedging instruments to reduce their exposure to such risk. A few task force members felt that recommending managed floating to emerging economies was not sufficient. The IMF also should become a strong advocate of inflation targeting and should give member countries specific advice on how to make it work.

Some task force members advocated a harder line on managed floating; specifically, they argued that “crawling band” regimes are not much less crisis prone than adjustable peg regimes (since strong speculative pressures would quickly move the former to the edge of the band) and therefore that the class of regimes referred to as “managed floating” ought to be reserved for cases in which there is no publicly announced exchange rate commitment.

A few task force members argued that currency boards are living on borrowed time. The first time a country under stress abandons its currency board, the message will get through that currency boards are not so hard to “undo” after all, and then all of them will become vulnerable.

Several members took the position that the world of two or three large currency blocks is closer than most think. The political factor is not the crucial one. The unacceptable volatility and uncertainty associated with floating rates is increasingly pushing emerging economies to look for alternatives, and a single currency regime—even if it does not come with a seat on the Open Market Committee—is better than the alternatives. How much monetary independence do emerging economies with floating rates have anyway?

A large group of task force members (comprising more than one-third of the group) felt that there could be no serious currency reform—indeed, no serious overall architectural reform—without reform of the G-3 currency regime (the dollar, the euro, and the yen). While this report focuses on emerging economies, they argued that much of the impact of the global economy on emerging economies is driven by swings among the G-3 currencies—swings that have been enormous, volatile, and unrelated to underlying economic fundamentals. For example, the dollar rose from 79 against the yen in 1995 to 145 in 1998. In their view, there is no interpretation of the Japanese and US economies that can explain such instability on the basis of the economic fundamentals. Early experience with the new euro, recent protectionist pressures in the United States associated with the sharp rise in the US current account deficit, and recent Japanese currency market intervention (to keep the yen from rising) added to these concerns. The reform of the G-3 currency regime that they preferred was a system of rather broad target zones or reference ranges (along the lines laid out in Section III).

Most task force members, while sharing a desire for more exchange rate stability among the G-3 currencies, were opposed to the target zone proposal. They doubted that G-3 monetary authorities would be willing to increase interest rates during a recession, or to reduce them during a bout of strong inflationary pressures, for the sake of defending a target zone—and with good reason. Furthermore, given the size and power of today’s global capital markets, they doubted that exchange market intervention could have much lasting effect on G-3 exchange rates; intervention therefore would not resolve this potential conflict for interest rate policy. They also did not believe that the announcement of wide and changing target zones would induce much stabilizing speculation. They thought the proponents of target zones had overplayed the role of G-3 currency fluctuations—relative to weaknesses in the domestic financial systems of the crisis countries—in motivating the Asian financial crisis. And they regarded the initial weakness of the euro as illustrating the potential pitfalls of target zones—not their attraction (since a target zone for the euro would likely have prevented European monetary authorities from reducing interest rates to counter weak domestic economic activity).

Given that the majority of the task force was not in favor of “loud” (that is, publicly announced) target zones, we also considered whether a regime of “quiet” target zones would have some appeal. Some task force members thought that quiet zones might induce authorities to identify and to act against large exchange rate misalignments at an earlier stage—without simultaneously tying too tightly the hands of the monetary authorities. Under such a regime, national authorities, with the assistance of the IMF, would agree on a reference (central) rate for each currency, along with, say, a zone of 15 percent on each side of that rate. National authorities also would agree to use coordinated exchange rate intervention along with monetary policy to keep market exchange rates from breaching the zones. However, neither the reference rate nor the zones would be announced to the public, so monetary authorities would retain the option of altering or overriding the zones (without publicly losing credibility) in those (hopefully unusual) cases where orienting monetary policy toward the exchange rate threatened to generate “pro-cyclical” effects on economic growth and/or inflation.

A few supporters of target zones thought that a prudent trial (say, for three years) with quiet zones would be worthwhile; some other supporters felt the quiet-zones variant would be too timid to achieve much (since one could not induce stabilizing speculation if the zones were kept quiet). Some opponents of target zones were concerned that quiet zones could not really be kept quiet, and this concern about leaks to the market would—in their view—prevent G-3 finance and monetary officials from supporting the idea. Others argued that if the zones were really kept quiet, there would be no test of whether they succeeded or failed.

Recommendation 5: Reforming the IMF’s Lending Policy: Less Can Be More

The objective should be to draw a sharper distinction between “country crises” and “systemic crises” and to treat the two differently in IMF emergency lending operations.

  1. The IMF should adhere consistently to normal access limits (100 percent of Fund quotas on an annual basis and 300 percent cumulatively) for country crises, that is, for crises that do not threaten the functioning of the international monetary system or the performance of the world economy.

  2. In the unusual case in which there appears to be a systemic crisis (that is, a multicountry crisis where failure to intervene threatens the performance of the world economy and where there is widespread failure in the ability of private capital markets to distinguish creditworthy from less creditworthy borrowers), the IMF would turn to its “systemic” backup facilities—either the existing NAB/GAB or a newly created “contagion facility.”

  3. The NAB/GAB would be used when the country’s problems are largely of its own making, when an IMF program (with upper credit tranche conditionality) is needed to correct the country’s problems, and when larger-than-normal amounts of assistance are needed. The contagion facility would be used for victims of contagion, that is, for countries in which the deterioration in the balance of payments reflects wider developments “largely beyond their control” and expected to be temporary (e.g., a large fall in global primary commodity prices, a large increase in the risk premium for many emerging-market borrowers, a sharp decline in private capital flows to emerging economies associated with a “flight to quality”). The contagion facility would not require a Fund program. Countries’ access could be no greater than the estimated effects of contagion on their balance of payments.

  4. The new contagion facility would replace both the Supplementary Reserve Facility and the Contingency Credit Line. Countries would not prequalify for the contagion facility, but eligibility decisions would be made expeditiously and disbursements would be heavily front-loaded. Loans would have short repayment maturities (no more than, say, 18 months) and, like the present SRF, they would carry interest rates higher than the rate for normal Fund lending operations.

  5. Like the NAB/GAB, activation of the contagion facility would require a supermajority of the creditor countries contributing to it to agree that this was indeed a systemic crisis. Nonsystemic episodes of contagion would be handled (within the normal access limits) by the Fund’s regular lending windows

    .
  6. The facility could be funded by a onetime allocation of IMF Special Drawing Rights (SDRs), in which all Fund members would agree to donate their share of the allocation to the facility. 22 There would also be agreement that only developing countries could draw on the facility. While the facility would be housed in the IMF, it would be regarded as lending “in association with the Fund”—not as the Fund’s facility. That is, the contributors to the facility (and not the Fund) would be taking the credit risk.

Commentary: Task force members recognized that making a distinction between “country crises” and “systemic crises”—especially in the heat of a crisis—was easier said than done, and that inevitably there would be cases in which the international repercussions of a crisis turned out to be either larger (for example, Thailand and Russia) or smaller (for example, Brazil) than anticipated. Still, the large majority of task force members took the position that efforts to draw such a distinction and to set a higher threshold for systemic crises are crucial if better market discipline is to be restored.

A number of concerns and questions were voiced about the potential implications of smaller IMF-led rescue packages for country crises. One worry was that smaller packages could have the unintended effect of inducing crisis-stricken countries to impose comprehensive capital controls. IMF quotas are much smaller today—compared to the volume and variability of trade and capital flows—than they were in the Bretton Woods era. Furthermore, quotas had not been reapportioned in favor of emerging economies that have liberalized capital inflows and thus were more vulnerable to capital outflows. If these countries run low on reserves and cannot intervene on the scale required to prevent precipitous depreciations, they may feel compelled to impose comprehensive capital controls, with undesirable spillover effects. Malaysia’s recent experience with capital controls was seen by some task force members as a case in point. Recourse to sensible debt restructuring might avert this outcome, but this cannot be assured in a world where governments often procrastinate in taking needed actions.

A second concern was that efforts to strengthen emerging economies’ financial systems would take considerable time and that a shift to smaller Fund lending before that strengthening took place could leave them in a vulnerable position. In this connection, some task force members argued that if certain emerging economies did not receive adequate protection against external shocks over the years during which they were building up their resilience, they might never evolve into strong market economies.

Yet a third concern was that smaller IMF rescue packages might put a disproportionate burden of crisis adjustment on the poorest and most vulnerable groups in the crisis countries.

Some task force members noted that the IMF had stayed well within normal access limits for the great majority of its lending operations. It was only in a small group of highly visible special cases that these limits were greatly exceeded.

These concerns notwithstanding, most task force members concluded that many emerging economies will not build the crisis prevention framework that is critical to greater resilience until they believe that they are more “on their own” in country crises and that smaller IMF loans are a necessary part of sending that message. Also, by making better currency regime choices, by making greater use of private debt rescheduling under appropriate circumstances, and by putting good social safety nets in place before crises strike, it should be possible to square smaller Fund packages with effective crisis prevention and management and with protection of the most vulnerable groups. They also noted that “smaller” was defined relative to Fund quotas, and that quotas themselves should be adjusted over time to reflect changes and growth in the world economy.

Turning to the systemic facilities, most task force members felt that the existence of such a contagion facility would give the international community a special instrument to deal with unusually widespread and serious episodes of contagion (that is, it would be like “catastrophic” insurance). It would be less subject to lender moral hazard than either the SRF and the CCL because of the “systemic” threshold for activation and because contributing countries would have their own money at stake when loans from the facility were extended to developing countries. It should not involve any borrower moral hazard, since it would be triggered by wider developments largely beyond the borrower’s control and since borrowing from it would carry market interest rates and short maturities. It would be simpler to operate than the newly established CCL, since it avoids prequalification and since no IMF program (shadow or otherwise) is required. It would have a different focus than the NAB/GAB.

An important advantage of giving a contagion facility permanent funding via an SDR allocation is that there would be no need to go to national legislatures for IMF funding while a crisis was raging.

Some task force members thought that an allocation of SDRs to fund a contagion facility might well be consistent with the requirement in the IMF’s Articles to meet the “long-term global need for reserves.” Others believed that an amendment to the Articles would be required.

While SDR allocations do not require congressional authorization or appropriations in the United States, most task force members took the view that extensive consultation with the Congress should be part of the decision process. If the United States contributed its share of an SDR allocation to a contagion facility, those funds would come from the US Treasury’s Exchange Stabilization Fund. For example, if the contagion facility had the same size as the NAB/GAB—that is, about $45 billion—then the US share would be about $8 billion. If a larger contagion facility were contemplated, say $100 billion, then the US share would be in the range of $18 billion.

A few task force members thought that the need for a systemic contagion facility has not yet been demonstrated—the events of the fall of 1998 notwithstanding. Several others took the contrasting view that if the systemic threshold for activating the facility were set too high, contagion could already be in high gear before any funds were forthcoming. They also argued that even if contagion of asset prices did not last long, the effects on the countries concerned could be serious and protracted. A few members came to the conclusion that if a systemic contagion facility was to be useful it would need to be quite large—much larger than the NAB. Some also questioned the interpretation of events “largely beyond their control”; for example, a country with low reserves would be more likely to be attacked.

Most task force members thought that distinctions among the Fund’s various lending windows were becoming blurred and that there were advantages in simplifying the lending structure. In the task force’s recommendation, that structure was straightforward: all nonsystemic country crises would be funded through the IMF’s regular lending window with normal access limits, whereas systemic crises would be funded either through the NAB/GAB (if the crisis was largely of the country’s own making and required an IMF program to correct it) or through the contagion facility (if the crisis was the result of contagion largely beyond the country’s control).

Several task force members were more supportive of the prequalification feature of the CCL as a deterrent to speculative attack. In their view, while the success of such an official contingent credit line cannot be assured, it is a worthy initiative. Private contingent credit lines have their own problems, including the risk that private lenders would withdraw their exposure from other loans if the contingent line was activated (resulting in no change in net credit to the crisis country). A few members regarded substitution of the task force’s proposed contagion facility for the CCL as a retrograde step. A number of task force members were concerned that even with the safeguards proposed by the task force, politically important troubled countries might still be able to obtain large bailouts even if there was no systemic risk. They argued therefore that the governance structure of systemic facilities should be made more conservative than the existing NAB. Some others argued that setting the systemic threshold very high would produce inaction in the face of true threats to the world economy and/or work to the disadvantage of smaller economies.

Several task force members also cautioned against viewing the IMF as the sole defense against systemic threats. In particular, and as illustrated by the events of the fall and winter of 1998-99, cuts in G-7 interest rates could be mobilized to counter systemic liquidity threats. Some task force members also highlighted the importance of maintaining an adequate level of global demand as a bulwark against widespread contagion of financial crises and underscored the responsibility of the larger industrial countries in this regard.

Recommendation 6: Refocusing the IMF and the World Bank: Back to Basics

The objective is to refocus the IMF and the World Bank to make their mandates more compatible with the needs of today’s global economy.

  1. Neither the IMF nor the World Bank should be abolished. The IMF is still needed to see that balance of payments problems, be they under fixed or flexible exchange rates, are resolved in ways that do not rely on excessive deflation, competitive devaluation, and imposition of trade restrictions, and to respond to liquidity crises when neither private capital markets nor national governments can handle those problems well on their own. Moreover, the IMF has an increasingly important role to play as a crisis manager or convener (to facilitate orderly debt rescheduling) and as a monitor of compliance with international financial standards. Yes, there is a moral hazard problem associated with large IMF financial rescues, but that moral hazard can be reduced significantly by altering the IMF’s lending policies (as we propose in recommendations 3 and 5 above). Likewise, the World Bank’s mission to combat poverty and to promote sustainable economic development remains essential, and it should continue to take a leading role in responding to the structural, human, and physical needs of its developing-country members.

  2. While restricting IMF membership to countries that have implemented a set of international financial standards and other crisis prevention measures goes too far, a closer link should be established between countries’ crisis prevention efforts and their access to IMF emergency financing, as emphasized in recommendation 1 above.

  3. Notwithstanding the continuing need for an IMF, there are some signs that the IMF is losing its focus and reducing its effectiveness by doing too much. Specifically, the IMF should limit the scope of its conditionality to monetary, fiscal, exchange rate, and financial-sector policies.

  4. In a similar vein, the World Bank would benefit from a refocusing of its mandate. The Bank should concentrate on the longer-term structural and social aspects of economic development. It should expand its work on social safety nets. But it should not be involved in crisis management, in emergency lending, or in macroeconomic policy advice.

  5. If the IMF and the World Bank were to refocus their mandates in this way and to collaborate effectively with one another, their partnership would be preferable to a merging of the two institutions.

Commentary: Several task force members were critical of the IMF’s policy advice to the Asian crisis countries—particularly the initial tightening of fiscal policy in Thailand and the removal of subsidies on food and fuel in Indonesia—and of its failure to recognize earlier the scope and implications of the Asian crisis.

A few argued that the IMF should have done more to implement “early warning” indicators of currency and banking crises and to develop a set of graduated “early action” measures to encourage a stronger, corrective crisis response on the part of emerging economies. Some members also felt that the IMF had shown a tendency to prescribe the same policy medicine in crisis situations that called for a more differentiated response. In contrast, some task force members thought that the Fund had been unfairly castigated for its interest rate policy recommendations in the crisis countries (since with reserves and currencies under strong downward pressure there was little alternative but to raise interest rates). Most task force members took the view that while (with hindsight) the Fund’s monetary and fiscal policy recommendations could have been better, these were judgment calls that did not lead to specific reform proposals.

Financial-sector surveillance and crisis management are included in the Fund’s mandate, because banking and financial-sector problems are much more connected than are other structural policy areas to the prevention, management, and resolution of financial crises. If the Fund were to function more effectively in this financial-sector sphere, it would need to bolster its expertise and resources in that specialty. Some members cautioned against assigning the IMF a larger role in the monitoring of financial standards and in surveillance of international capital markets without giving it the staff resources to do that job effectively; even if the Fund were to go to smaller lending packages, the staff resources problem would still be there. A few members maintained that oversight of financial sectors would have to involve auditing and accounting, bankruptcy and insolvency, and corporate governance. Thus care should be taken not to define the Fund’s oversight role too narrowly.

Some task force members argued that narrowing the scope of the Fund’s mandate need not imply that other important areas of economic policy would be ignored. Those areas would fall instead to other agencies with the requisite mandates and expertise. For example, core labor standards should be the business of the International Labor Organization (ILO), environmental problems should be taken up by the World Bank, trade policy issues should be addressed by the World Trade Organization (WTO), and so on. It was conceded, however, that so long as the IMF has the financial leverage that some other international organizations do not, there will be a temptation to use it to advance other objectives.

A few task force members argued that it is a “cop-out” to look to other international organizations to secure labor, environmental, and social protections (since these other organizations have no enforcement power). They also emphasized that financial stability is a means to broadly shared prosperity and not an end unto itself; that competition should be encouraged mainly via improvements in value added (and not via suppressing wages and working conditions); that human rights and democratic freedoms are not only key concerns in their own right but also reduce corruption and cronyism; that the Fund and the Bank will not be able to maintain the support of labor and human rights groups unless they pay greater attention to core labor standards; and that failure to pay such attention will only force advocates to pursue unilateral approaches through their own governments.

One view was that the best way to meet these labor and social concerns is to establish a joint IMF/World Bank/ILO working group that would look into linking core labor standards with international financial standards. Most task force members, however, while sympathetic with these concerns, were against “overloading” the policy conditionality of the IMF and World Bank with new objectives. A few members expressed strong reservations about the wisdom of trying to reshape social policy during a financial crisis.

A number of task force members emphasized that the design of social safety nets should be a crucial part of the Bank’s mandate for the social aspects of development. These should ideally be in place before a crisis strikes. Without an adequate social safety net, many crisis-management policies would simply not be sustainable. In this connection, they stressed that the Asian crisis was first and foremost a human tragedy for the inhabitants of the crisis countries and that other technical components of architectural reform should not be allowed to obscure this. If IMF loans for country crises will be smaller in the future, concern for the human aspects of crisis resolution would imply that care must be taken to see that the poor do not bear an unduly heavy part of the adjustment burdens.

One reason the World Bank got into emergency lending in the Asian crisis was that the IMF was running out of money. This reinforces the case for ensuring that a contagion facility be adequately funded outside of the quota replenishment exercise.

Several task force members argued that both the Fund and the Bank ought to add specialists who have expertise on social, political, and cultural factors in developing countries. To accommodate this, the “team approach” to Fund and Bank missions may need to be expanded.

A few task force members cautioned against a rigid interpretation of avoiding duplication of effort by international financial organizations. They noted that lending by regional development banks—or even new regional institutions—might overlap with lending by global financial institutions, but the former could be informed by deeper knowledge of the region. Some members urged the regional development banks to become more active in the crisis prevention area. The point was also made that an absolute prohibition on crisis lending by the World Bank could be unwise: while rare, systemic threats could arise in which it would be necessary to order “all hands on deck.”

Recommendation 7: Fostering Political Commitment and Ownership for Architectural Reform among Emerging Economies

The objective is to find a vehicle for strengthening the political commitment to reform and for promoting ownership of reforms—particularly among emerging economies.

  1. The IMF Interim Committee, the Basle Financial Stability Forum, and the presidents of the regional development banks together should convene as soon as possible a special global conference of finance ministers. 23

  2. The aim of that conference would be to reach a consensus on priorities and timetables for domestic actions that countries will need to take to strengthen national financial systems against financial crises.

Commentary: As argued earlier, many of the cracks in the architecture reflect weaknesses at the national level, and these in turn can be fixed only if national governments muster the political will to do so. Inevitably, there will be vested interests trying to block needed reforms. Reform-minded governments would find it easier to confront these obstacles if they were part of a larger group or “club” that was publicly committed to implementing reforms. In addition to applying peer pressure in the right direction, a public declaration should induce the right kind of market payoffs: those signing on would be rewarded by markets with a lower borrowing cost, while those opting out or reneging on earlier commitments should see their borrowing costs increase.

Task force members argued that reform programs were most successful when the countries most affected participated directly in their design and when they “took ownership” of them. Several task force members noted that it would not be surprising if some crisis countries resented calls for wholesale changes in their domestic financial systems when they were flat on their backs during a crisis. A global conference would allow many emerging economies to have a greater input into desired changes in the architecture away from the confines of an IMF program. The decision to commit to a joint declaration would be completely voluntary, although it would have market repercussions. Many task force members emphasized that if emerging economies did not see themselves as full partners in the reform exercise, it would not work in the end. While some other fora (for example, the IMF, the G-22 working groups, and soon the Financial Stability Forum) had emerging-market participation, it should include more than just the largest emerging economies.

A few task force members thought that participation of the regional development banks in such a global conference could be useful in bringing regional perspectives to bear on architectural reform; this could help to guard against a “one size fits all” approach to standards and could increase support for reform.

A number of task force members favored such a conference for another reason. It has been nearly five years since official discussions of architectural reform intensified. Also, prospects for recovery from the Asian/global crisis have brightened over the past nine months or so. There is therefore a real danger that “fatigue” and “complacency” will combine to stall the momentum for reform—before most of the recommendations set out in this report can be implemented. Some way must be found to regain that momentum, and perhaps a global conference that would “round up” individual reform codes could produce a wider consensus for implementation.

A number of task force members preferred that the special global conference be a summit held by heads of state and governments (rather than by ministers of finance). They argued that major international economic initiatives have invariably been decided by heads of state, that finance ministers have not been able to deliver sufficiently sweeping architectural reform, and that a global summit offers the best chance of reinvigorating the reform process. This summit could be convened jointly by G-7 leaders and the managing director of the IMF.

Not all task force members shared these views. Some doubted that another conference of finance ministers (or even a summit of heads of state) discussing reform of the architecture would add much to previous meetings and communiqués—even if participation from the developing world was somewhat wider. Many of the larger emerging economies had already publicly committed to the main elements of reform of the architecture in the G-22 working groups and elsewhere. Some worried that discussion of the details of such a global conference (for example, who would convene it, how many smaller emerging economies could participate, how sweeping the agenda would be, whether there would be a series of such conferences, etc.) could detract attention from substantive discussion of the issues, or could raise unrealistic expectations about what could be accomplished.

In the end, however, most task force members thought this initiative was worthy of their support.

Where We Differ From The Official Sector’s Recommendations

Many of the themes emphasized in this report have also been part of the official sector’s plans and suggestions for the future architecture. In this sense, there is a substantial area of agreement between us and them on what is “broken” in the existing architecture and on the broad direction of reform. That said, there are at least seven areas in which our recommendations differ from theirs:

 

Concluding Remarks: Moderate versus Radical Reform Plans

Within the already large and rapidly expanding universe of architectural reform plans, the set of recommendations outlined in this report probably is best characterized as “moderate” or “moderate plus.” We did consider many more radical proposals—ranging from more comprehensive controls on private capital flows, to single currency regimes, to a much larger and more powerful IMF, to no IMF, to the creation of new supranational regulatory institutions, to various kinds of insurance mechanisms, to much larger penalties/restrictions on large, highly leveraged traders. Yet, as wide and as deep as the Asian/global financial crisis was (is), most of us were not convinced that we should choose the more radical alternatives.

Some of them seemed impractical. Some seemed undesirable. And perhaps most important, we thought that the package of moderate recommendations put forward by the task force would make a significant difference to crisis prevention and resolution if implemented and if given a chance to work. Only time will tell if that was the right call.

 

Dissenting Views

On Target Zones For The G-3 Currencies

As indicated in the report, a “group of task force members . . . felt that there could be no serious currency reform . . . without reform of the G-3 currency regime (the dollar, the euro, and the yen).” We constitute that group.

Indeed, we would go further. We believe there can be no serious reform of the overall financial architecture without fundamental reform of the way in which the G-3 manage the relationships among their exchange rates. We therefore advocate working toward the development of arrangements to achieve greater stability among the dollar, euro, and yen. Preferably, we contemplate arrangements along the lines of rather broad target zones or reference ranges. The current debate over international monetary reform, and the report of the task force, have understandably focused on the emerging-market economies (including their exchange rate regimes, to which we return below). But the impact of the global economy on emerging countries is driven significantly by swings among the currencies of the three major economic powers. In recent years these swings have been enormous, volatile, and frequently unrelated to underlying economic fundamentals.

For example, the dollar soared from its record low of 79 against the yen in 1995 to 145 in 1998—a rise of over 80 percent in three years. No conceivable interpretation of the economies of Japan and the United States could explain such instability on the basis of “the fundamentals.” This has had at least three major effects on the Asian/global crisis:

Early experience with the new euro heightens concerns over these currency problems. Its weakness has limited the scope for reductions in European interest rates at a time of weak growth there and, if extended, could add significantly to the US trade deficit and its attendant complications.

These are only a few of the most recent and most critical manifestations of the absence of effective G-3 currency arrangements. We are disappointed that the G-3 governments have continued to bypass this central issue even while seeking to reform the international financial architecture. We believe that the task force report should have included forceful recommendations for improving G-3 currency management.

There is no need for the G-3 countries to return to pegged or fixed exchange rates. But freely floating rates have produced the costly consequences cited above (and even worse results in the 1980s, when the dollar soared by more than 50 percent and converted the United States from the world’s largest creditor to the world’s largest debtor country). In recognition of these costs, the G-3 authorities have in fact intervened in the currency markets from time to time.

The practical issue for the G-3 countries is how to manage a regime of flexible exchange rates among their currencies. The current G-3 authorities intervene on a totally ad hoc and episodic basis, without any clear sense of a sustainable equilibrium. Such intervention typically comes too late to prevent severe currency misalignments. These imbalances in turn trigger major economic distortions, protectionist trade pressures, and inevitable sharp currency reversals that generate a second round of large costs.

The concept of wide target zones, ranging from 10 to 15 percent on either side of agreed midpoints, pursues a very simple objective: avoidance of such large and prolonged misalignments, while permitting substantial flexibility for the conduct of national monetary policies and adjusting to changes in economic circumstances. Specifically, we believe that:

No system of currency management, including target zones, is perfect. We can envisage situations in which implementation of such a regime could run counter to the short-term requirements of one or more G-3 economies. But the current non-system of ad hoc managed floating also occasionally precipitates strong actions to counter exchange rate declines that threaten domestic objectives, typically after damage has been done, both for the United States and for the world economy as a whole. We believe that a system of target zones should have been recommended by the task force and should be adopted by the G-3 countries.

For somewhat similar reasons, we believe that many emerging-market economies might also properly adopt intermediate currency regimes like target zones (or frequently, in their cases, “crawling bands,” since the level of their zones may need to be adjusted periodically to offset differentials between their inflation rates and those of their main trading partners). It is true that some smaller countries that are highly exposed to the world economy (e.g., Hong Kong) cannot accept the high costs of any significant degree of rate flexibility and should thus seek a credibly fixed rate, perhaps through a currency board or complete dollarization (or euroization). Similar solutions may also make sense for countries with long histories of rapid inflation (e.g., Argentina), which therefore have an overriding need to find a credible anchor for price stability. Most emerging-market economies fit neither of these categories, however, and we would thus underline the task force’s recommendation that the IMF encourage such economies to pursue regimes of “managed floating,” including those based on crawling exchange rate bands. What we miss in that recommendation is emphasis on the complications for the conduct of any sensible exchange rate policy by emerging-market economies when the exchange rates among their major trading partners move so erratically.

Our point is that “reforming the international financial architecture” without reforming the currency regime is like watching Hamlet without the Prince. The international monetary system will continue to be ineffective and crisis prone until that crucial centerpiece of its operation is thoroughly revamped. We urge the G-3 countries to adopt some variant of target zones in the near future.

Paul Allaire
C. Fred Bergsten
George David
Maurice Greenberg
Lee Hamilton
John Heimann
Ray Marshall
James Schlesinger
George Soros
Ezra Vogel
Paul Volcker

On Regulation Of Highly Leveraged Institutions

The report concludes that “indirect” approaches to regulating hedge funds and other highly leveraged institutions (HLIs), such as tightening risk-management guidelines for banks that lend to hedge funds and publishing data on lenders’ exposure to HLIs, “deserve a fair trial.” We believe that this is inadequate and that more direct regulatory initiatives should be imposed immediately. A sensible first step is the imposition of higher risk weights for bank loans going to offshore financial centers that do not meet international financial standards, the locus of most HLI activity.

As the report points out, there is a fundamental asymmetry between financial markets in many emerging economies and the positions that large, highly leveraged institutional investors can trade in those markets. The market power of the HLIs can overwhelm the countries, particularly in a crisis atmosphere. The emerging- market economies are properly being asked to adhere to more rigorous financial standards, and those on the other side of the market should be required to do likewise. Hence direct regulation is needed now.

C. Fred Bergsten
Maurice Greenberg
Carla Hills
Laura Tyson

On A Global Summit

Recommendation 7 of the report calls for a special global conference of finance ministers to strengthen the political commitment to reform and promote ownership of reforms—particularly among emerging economies. We support the objective but believe that the proposed response is too weak.

The “special global conference” should be held by heads of state and governments rather than ministers of finance. Only the political leaders of the key countries can provide the necessary push for meaningful reform of the international financial architecture. Every major international economic initiative in the postwar period, including the creation of the IMF itself and such historic recent events as the creation of the euro, was decided by heads of state—often over the opposition of their ministers of finance.

The special conference could be called by the G-7 heads of state and government, perhaps in conjunction with the IMF. To ensure full and balanced representation of all relevant points of view, it should include all industrial and emerging-market economies that play a significant role in the world economy.

The ministers of finance and their deputies have been meeting continuously since the eruption of the latest crisis in the middle of 1997, and indeed since the Mexican crisis at the end of 1994. They have been unable, however, to agree on sufficient reforms to protect the world against future and perhaps even deeper disruptions—as the report makes clear.

Even the limited improvements that the ministers have adopted were driven importantly by initiatives from their political leaders. At the G-7 summit at Naples in 1994, the leaders presciently realized that the monetary system needed reform—a judgment that was reinforced by the Mexican crisis barely six months later—and instructed their ministers to address the topic. As a result, the leaders were unable to adopt some systemic improvements at their next G-7 summit in Halifax in 1995. The finance ministers failed to launch any new systemic efforts after the outbreak of the Asian crisis until the approach of the Asia Pacific Economic Cooperation (APEC) summit in November 1997 galvanized most members of that group to create the Manila Framework, which developed modifications in IMF arrangements that became crucial elements in the support program for Korea a month later. The task force that produced the present report was created at the suggestion of President Clinton, who obviously saw a need for ideas beyond those that were being offered by ministers of finance and their aides.

It would appear futile to expect the ministers of finance to adopt major systemic reforms without a new political impulse. This is especially true in light of signs that the latest crisis is ebbing, which will reinforce the tendencies of the ministers to retreat to the status quo ante. The only successful method to reinvigorate the process would be a summit meeting of the heads of state and government of the 30 or so key countries, both industrial and developing, that have the largest impact on the global monetary system. Such an initiative will be required to obtain early implementation of the recommendations of the report.

C. Fred Bergsten
Carla Hills
Peter Peterson
Laura Tyson

On Incentives For Crisis Prevention And An Early Warning System

The report stresses the need for better crisis prevention, especially in its first reform recommendation on joining the “good housekeeping club.” In our view, however, stronger language is called for in mandating accurate and timely data disclosure by IMF member countries. In addition, the IMF should publicize its own diagnoses in unambiguous terms, escalating the strength of its commentary in consonance with the severity of mounting crises.

The currency crisis phenomenon is well understood and such crises are largely predictable. For example, the IMF understood the problem in Thailand and tried to persuade that country to act for many months prior to July 1997. The situations were identical subsequently with Russia and Brazil, as they have been in countless other cases. The difficulty is that lenders are attracted by high yields, at the same time that borrowers profess commitments to fixed exchange rates—until the day that exchange rates are no longer fixed. Timely and accurate information revealing actual and potential weaknesses is often not available, prolonging and ultimately making more severe the inevitable crises.

The analogies to the public securities markets in the United States and elsewhere are extensive and persuasive. We have learned through repeated experiences the value of accurate information, timely analyzed and widely disseminated. The fact is that securities markets, when supported by reliable and widely known information, are themselves the best disciplinarians of excesses.

The comparison of data disclosures mandated for public issuance of securities in the United States, and for balance of payments and related data issued by governments, is adverse. The latter information is rarely timely, often subject to revision, and from time to time simply misrepresented. When the SEC mandates restatements of financial statements in the United States, it is an event of profound importance and, fortunately, of comparative rarity. The same cannot be said of financial information issued by any government.

The IMF must be the first line of defense in mandating accurate and timely balance-of-payments and related information. The newly agreed international data standards are a start, as is closer supervision of banking institutions.

But the only way the quality of information issued by governments will begin to approach that of information issued by private borrowers is if such information is a precondition of IMF emergency support (if ever needed) in the first place. The IMF should accordingly publish minimum standards for data disclosure and for financial-sector regulations; it should then regularly publish reports on individual countries’ compliance with these standards, and it should make availability of crisis lending conditional on such compliance.

The IMF should more regularly and more forcefully disclose its diagnoses of impending crises. Such disclosures will have an impact in the marketplace, encouraging (and properly so) corrective actions by both borrowers and lenders. To the concern expressed by some that IMF statements will precipitate crises, we respond that the crises will happen eventually anyway, and will almost always be the more severe for the delays.

Information and data shared openly and rapidly are the best devices to curb excesses, to limit exposures, and to cause corrective actions before impending crises become actual crises. The IMF must set the ground rules, materially expanded from today, and member countries must comply as a precondition for crisis lending facilities should contagion or other factors ever require these facilities to be drawn down.

Rapid and reliable information, greater transparency, and more aggressive IMF whistle-blowing in turn provide a foundation for the development of a much more effective early warning and early action system. As already noted, early warning is an analytical exercise that requires reliable anticipation of pending crises. Early action is a political exercise that seeks to convince governments that face looming crises to take preemptive measures. New intergovernmental mechanisms to pursue both should also be part of any effective reform of the international financial architecture.

Early warning can now be based on sophisticated and persuasive explanations of past crises. Many models of past banking and currency crises have been developed by scholars and market practitioners, and yield reasonably robust statistical indicators of future disruptions. Governments should study, adopt, and utilize these newly available tools to mount much more serious efforts to anticipate “the next crisis.”

To be sure, no set of indicators will be totally accurate. Some crises will be missed. Some “predicted crises” might not occur. But the system has failed in this respect in recent years, and it would be foolish not to try to improve the process, especially with far better analytic tools now available to support it.

Even with effective crisis anticipation, however, a country headed for trouble must still be persuaded to take preventive action. In light of national sovereignty and governments’ proclivity for self-delusion, this is the most difficult part of the entire process. As noted above, the management of the International Monetary Fund (and perhaps individual G-7 officials) will thus need to call public attention to a country that is clearly en route to a crisis but seems unwilling to adopt corrective measures. A series of gradually escalating expressions of official concern after private entreaties have proved fruitless, and carefully calibrated to the circumstances, would alert markets and others to reconsider the creditworthiness. Market pressure would thus be increasingly brought to bear for adjustment actions before the crisis stage was reached.

In addition, regional institutions should be created to help foster “early action.” Neighbors are in the best position to exert peer pressure on a recalcitrant country, both because they are likely to know the situation better and because they have more legitimate concerns over the costs of inaction—as revealed so dramatically in the Asian crisis and in Latin America from both the “tequila effect” in 1995 and the more recent spillover from the Brazilian crisis. Regional institutions could in fact come to support the IMF’s multilateral surveillance process in the same way that the regional development banks support some of the World Bank’s global development programs.

Aggressive reform of the early warning and early action systems should play a central role in rebuilding the international financial architecture, and the report errs in understating this element of the process.

C. Fred Bergsten
George David

On The Need For A More “structural” Approach To Architectural Reform

One cannot fail to be impressed by the wide range of issues covered by this report and the balanced analysis of the options. Coming out at a time when the sense of immediate crisis is receding, it is an important reminder of the simple fact that recurrent crises in the international financial system have become a fact of life. The report correctly emphasizes important questions for the United States as well as for smaller countries.

For all those merits, what we find absent is sufficient emphasis on structural characteristics of what has rapidly become a truly globalized economy that have made it prone to crises. Rather, the report takes a narrower focus: “primary responsibility for crisis avoidance and resolution in emerging economies ... belongs ... on emerging economies themselves and on their private creditors.” Given that approach, there is a clear consensus on some points: better banking supervision and regulation, stronger auditing, greater transparency, and more prudent lending. There is also sympathy for restraints on inflow of short-term capital.

We are in favor of those approaches. But we are also certain that American-style regulation, supervision, and auditing will not protect small emerging economies from the volatility of international capital flows. The potential for controls in a country that wishes to participate fully in the world economy is also limited. In sum, these are useful and important points, but they are not really fundamental—they are more a matter of interior decoration than of basic architecture.

We do not know whether this latest crisis is the worst in 50 years or not. Certainly we have had others that threatened, at least as seriously, the large international banks and even US financial stability. What we do sense is that the crises of the 1990s provide a crucial test of the consistency of free and open global capital markets with the interests of individual nations, particularly small emerging economies.

Fortunately, the instinctive response to the crisis among those emerging nations has been mainly to integrate their economies further. Most obviously, in the search for stability they are permitting and even encouraging foreign ownership of banks and other financial institutions, matters that not so long ago were considered a question of vital national sovereignty. Direct investment has been relatively well maintained and the forms of equity investment are encouraged.

Most radically, while economists talk about flexibility, practical business interests in a number of countries are exploring the merits of “dollarization” or “euroization” in the search for stability and predictability.

All of this is recognition of the fact that the globalization of markets means that autonomy for “domestic” monetary policy, or for domestic “macro-policy” generally, is fading, certainly for smaller, inherently more internationally exposed nations.

The useful debate about the role of the IMF touches upon these matters. But, as reflected in a dissent in which one of us has joined with others, more attention needs to be directed toward means of practically achieving greater stability in exchange rates, including those of major countries.

Laura Tyson
Paul Volcker

On Core Labor Standards

The task force report quite properly acknowledges that financial stability is a means to the end of broadly shared global prosperity, not an end unto itself. The task force also urges the World Bank to do more in the design of social safety nets. The task force should, however, also place financial policy in a larger conceptual framework and recommend that the US administration: (1) propose that the World Bank, the IMF, and the ILO form a working group to explore linking core labor standards to international financial transactions; and (2) strengthen its efforts to get a working party on international labor standards in the WTO.

International financial institutions should encourage global competition mainly by increasing value added (productivity and quality), not by suppressing wages and working conditions. It can be demonstrated that high value added competition is more beneficial to most people in all countries than a “race to the bottom,” which would result from competition based on reducing or suppressing wages and working conditions. Human rights and democratic freedom are legitimate concerns of international financial institutions as ways to increase long-run economic sustainability as well as to prevent corruption and cronyism, which undermine financial stability and efficiency.

Social safety nets and the implementation of core labor standards are necessary for humanitarian as well as political and economic reasons. The increased poverty, unemployment, misery, and despair resulting from financial crises will undermine support for open economic and financial markets as well as for democratic institutions. We should not create the impression that investors will be protected while millions of ordinary people will see their futures and living conditions decimated. We should therefore endorse the G-7 countries’ request that the World Bank, in consultation with related institutions, develop principles for good social practice and recommend the inclusion of core labor standards in international agreements. We should also encourage cooperation between the ILO, the World Bank, the United Nations, and other international institutions, none of which individually has the expertise required for a comprehensive, broadly shared development strategy.

Core labor standards that should be part of the rules for international transactions include the rights to free association, collective bargaining, strike, workplace safety, minimum wages, prohibition on forced labor, and limits on child labor. These core standards were included in all major US trade legislation of the 1980s, and in 1995 Congress required the US Treasury to instruct the US directors to the World Bank and the IMF to use their “voice and vote” to persuade these organization and their borrowing member countries to respect core labor rights. In approving the 1998 US quota increase for the IMF, Congress mandated that respect for these core labor rights not be undermined by measures the IMF might require to increase labor market flexibility, a term often used to justify weakening workers’ fundamental rights to organize and bargain collectively.

Several developments during the 1990s raise doubts about the will of political and economic institutions to give more than lip service to core labor rights. The North American Free Trade Agreement (NAFTA) would not have passed without the Labor Side Agreement (LSA), which President Clinton and many members of Congress required as a condition for their support. However, at the urging of Mexico and US business interests, President Clinton deleted the provision for trade sanctions and monetary penalties. Without penalties, the LSA has been ineffective in ensuring that the Mexican government allow workers to exercise their constitutional rights to collective bargaining through independent unions of their own choosing. Violations of these rights in Mexico have been documented by the Labor Department’s US National Administrative Office (USNAO). The Clinton administration also attempted to include core labor standards in the WTO and the Free Trade Area of the Americas, but both organizations refused to form a working group even to begin discussions on how labor standards might be incorporated into the emerging trade and investment rules. In 1996 the WTO did refer this matter to the ILO, which announced its intention to strengthen its procedures to investigate labor rights violations. The problem, of course, is that while the ILO has well-established mechanisms to investigate and report on violations, it has no enforcement power. Combining the ILO’s investigatory processes with the trade and investment leverage of the WTO, the IMF, and the World Bank therefore could provide more effective enforcement.

Core labor standards are important because global economic integration has caused the basic economic and equity rationales for labor standards to apply to international as well as domestic markets. The equity rationale is to protect workers, especially the most vulnerable, from exploitation and the most adverse consequences of competitive markets. The economic efficiency rationale is to cause competition to improve efficiency, not to depress basic labor standards. As US Treasury Secretary Larry Summers noted:

Dani Rodrik [in Has Globalization Gone Too Far? (Institute for International Economics, 1997)] has raised the right and difficult question. United States laws prevent workers from being driven out of their jobs by other American workers willing to work 12 hours a day, accept subminimum wages or forgo the basic right to organize. How should they react to not being protected from foreign workers who are willing to accept these things?
(L. H. Summers, “Distinguished Lecture in Government: Reflecting on Managing Global Integration,” Journal of Economic Perspectives, Spring 1999, pp. 9-10)

The second reason for core labor standards in international financial and economic transactions is political. The promise of labor and environmental guarantees made NAFTA possible; disappointment with the enforcement of these guarantees and the resistance to core labor standards by the WTO and other international economic institutions was a major reason for Congress’s failure to grant the administration fast track authority in 1997 and for the inclusion of support for such standards in the 1995 and 1998 IMF legislation.

Acceptance of these recommendations might ensure broader public support for the task force’s recommendations, as well as for a more open and expanding global economy.

Ray Marshall

On Capital Controls, Private-sector Burden-sharing, And Collective-action Clauses

Although I do agree with the general thrust of the task force report, I take issue with certain of its recommendations and perspectives. I disagree with the report’s encouragement, in Section IV, of the use of capital controls in emerging economies. Capital controls are easy to implement, but difficult to remove. Furthermore, it is questionable how much they actually contribute to overall sustained economic growth. For example, in Chile capital controls played only a minor part in the success story of that country’s economy. The bottom line is that there is no substitution for the consistent implementation of sound economic policies over time.

One of the shortcomings of the report is that although it points out the change in the composition of private capital flows from the 1980s to the 1990s, i.e., a decrease in bank loans and an increase in other flows such as bonds, it nevertheless singles out commercial banks with regard to burden-sharing, and thus gives the false impression that the commercial banks are still the major holders of emerging market debt, as they were in the debt crisis of the early 1980s. Since that time there has been a mass securitization of emerging-market debt, such as Brady bonds, and the entrance of new players, such as mutual funds and pension funds. That is why any recommendations put forward on crisis prevention and crisis management and solution must take into account the current diversity of the holders of debt.

Also in the discussion in Section IV of the IMF’s role in fair burden-sharing and market discipline, the IMF should support temporary halts in debt payments only in the context of an overall restructuring or refinancing, so as not to discourage private financing and to improve market sentiment.

With regard to the new financial architecture, it is essential to ensure that capital flows return in a prudent and stable manner; however, it is important that this not be done in such a way as to discourage flows or close off certain markets. More specifically, we have to be careful about forcing the insertion of certain types of bankruptcy clauses into sovereign bond issues, because it could limit the demand for these instruments at a time when it would be desirable to encourage flows back into the emerging markets. The implementation of such clauses must be done case by case, on a voluntary—i.e., market—basis, on the part of both the issuers and the takers of these instruments. The object should be to avoid fighting yesterday’s war and weakening our ability to respond to new circumstances.

William R. Rhodes

On Measures To Encourage Sound Long-term Lending To Emerging Economies

This report has sought to build a consensus among people of widely divergent views and interests, and it has succeeded in presenting a coherent and well-informed analysis and some useful recommendations. I consider the exercise worthwhile and the product valuable; therefore, I shall keep my dissent to a minimum. I feel obliged, however, to point out a bias that permeates the report. The people who participated in the task force, myself included, occupy positions at the center of the global capitalist system. This colors their views and interests, and the report reflects it. The system is tilted in favor of the center, namely, the owners and providers of capital, and the economies at the periphery are at a disadvantage. The global financial crisis has exacerbated the difference. The report does not give sufficient weight to the need to create a more level playing field.

The report correctly points out that the IMF has encouraged excessive lending by coming to the rescue of heavily indebted countries in case of trouble. It seeks to prevent a recurrence by imposing a heavier burden on the lenders: bailing them in instead of bailing them out. Well and good. But the lenders are liable to charge for the risks they are taking. The various reform measures already undertaken and even more radically advocated by the report create a danger that the opposite extreme will occur: lending to the periphery will remain scarce and expensive. It will tilt the playing field even more in favor of companies operating from the center. Direct investment has the merit of being more stable, but it has other drawbacks, and it is liable to run into political opposition, especially if it takes place on an uneven playing field.

What is missing from the report is any measure to encourage sound, long-term lending. It could easily be introduced by tying the kind of assistance the IMF is willing to provide to the standards and performance of the borrowing countries. This is already happening with the introduction of the Contingency Credit Lines. It needs to be reinforced by an additional step.

The IMF should declare that in the case of those countries that meet the required standards, IMF programs would not involve debt restructuring, so that bondholders need not fear that the collective-action clauses would be invoked except in the case of individual companies failing. This would enable the countries concerned to borrow in the markets at cheaper rates. The IMF assurance would be confined to publicly issued bonds, and it would exclude bank lines. Providing banks with implicit guarantees has been a major source of the trouble in the recent crisis. In the case of the banks, the leverage that the IMF needs in order to prevent crises from developing could be provided by varying the capital requirements under the Basle Accords, as suggested by the report.

These measures taken together would provide both the sticks and the carrots the IMF needs to become an effective institution for crisis prevention. Moreover, the carrots would encourage long-term lending, and the sticks discourage short-term lending.

The proposal makes eminent sense, yet it was rejected on grounds of moral hazard. Some element of moral hazard is inherent in any lender of last resort or insurance activity. Moral hazard has become a code word for resisting any interference with the market that would have the effect of creating a more level playing field. It is a bias that should be resisted.

As it stands now, the report is all sticks, no carrots. Its recommendations would have helped to prevent the last crisis, but not the next one, just as the Maginot Line, built on the experience of the First World War, did not protect France in the Second. The danger facing us is a dearth of international lending, not an excess.

It so happens that the additional step I am advocating could be easily incorporated in recommendation 1 by making the following small modifications: In paragraph (a), replace “the interest rate it charges” with “the kind of assistance the IMF provides.” At the end of paragraph (c), before “Other things being equal, ...” insert: “In the case of class A countries the Fund would not countenance and in the case of class B countries the Fund would not require any modifications in the terms of bond contracts.”

I agree with many of the other recommendations—or do not disagree sufficiently to voice dissent—with the exception of recommendation 5, which is an ill-conceived attempt to placate market fundamentalists. It is ill-conceived because by distinguishing between “country crisis” and “systemic crisis” it leaves every country, and particularly small countries that are not supposed to pose a systemic threat, to the vagaries of inherently unstable financial markets. Who would have thought that a country like Thailand could unleash a global crisis? Replacing the Contingency Credit Line with a contagion facility would be a retrograde step.

The only valid point in this section is the need for a onetime allocation of SDRs in which all Fund members would donate their share of the allocation to the Contingency Credit Line.

George Soros

 

Members of the Task Force

Commission on the Future International Financial Architecture Paul A. Allaire is Chairman and Chief Executive Officer of Xerox Corporation. Mr. Allaire joined Xerox in 1966, serving in a variety of senior posts, and was elected President and a member of the Board in 1986.

C. Fred Bergsten has been Director of the Institute for International Economics since its creation in 1981. Dr. Bergsten served at the Treasury Department as Assistant Secretary for International Affairs from 1977-81, and at the National Security Council as Assistant for International Economic Affairs from 1969-71.

Kenneth W. Dam is Max Pam Professor of American and Foreign Law at the University of Chicago. Mr. Dam was Deputy Secretary of State from 1982-85.

George David is Chairman and Chief Executive Officer of United Technologies Corporation. Mr. David joined Otis Elevator Company in 1975, just prior to the merger of Otis into United Technologies.

Jorge I. Dom’nguez is Director of the Weatherhead Center for International Affairs and Clarence Dillon Professor of International Affairs at Harvard University. Dr. Dom’nguez is a former President of the Latin American Studies Association.

Kenneth M. Duberstein is Chairman and Chief Executive Officer of The Duberstein Group, an independent strategic planning and consulting company. Mr. Duberstein served as Chief of Staff to President Reagan from 1988-89.

Barry Eichengreen is John L. Simpson Professor of Economics and Political Science at the University of California, Berkeley, where he has taught since 1987. Dr. Eichengreen was Senior Policy Advisor at the International Monetary Fund from 1997-98.

Martin Feldstein is George F. Baker Professor of Economics at Harvard University and President of the National Bureau of Economic Research. Dr. Feldstein was Chairman of the Council of Economic Advisers and President Reagan’s chief economic adviser from 1982-84.

Morris Goldstein is Dennis Weatherstone Senior Fellow in International Finance at the Institute for International Economics. Dr. Goldstein served as a staff member of the International Monetary Fund for twenty-five years, the last eight as Deputy Director of the Fund’s Research Department. Maurice R. Greenberg is Chairman and Chief Executive Officer of American International Group, Inc., and Vice Chairman of the Council on Foreign Relations. Mr. Greenberg is a past Chairman, Deputy Chairman, and Director of the Federal Reserve Bank of New York.

Lee H. Hamilton is Director of the Woodrow Wilson International Center for Scholars. Mr. Hamilton served as a member of Congress from the ninth district of Indiana from 1965-99 and chaired the House Foreign Affairs Committee.

John G. Heimann is Chairman of the Financial Stability Institute, a subsidiary of the Bank of International Settlements. Mr. Heimann was Chairman of Global Financial Institutions at Merrill Lynch & Co., Inc.

Carla A. Hills is Chairman and Chief Executive Officer of Hills & Company, an international consulting firm. Mrs. Hills was United States Trade Representative from 1989-93.

Peter B. Kenen is Walker Professor of Economics and International Finance at Princeton University. He taught at Columbia University from 1957-71, where he was Chairman of the Department of Economics, and Provost of the University from 1969-70.

Paul R. Krugman is Ford International Professor of Economics at the Massachusetts Institute of Technology. Dr. Krugman was International Policy Economist for the Council of Economic Advisers from 1982-83.

Nicholas R. Lardy is Senior Fellow in Foreign Policy Studies at the Brookings Institution and Frederick Frank Adjunct Professor of International Trade and Finance at Yale University’s School of Management. Dr. Lardy was Director of the Henry M. Jackson School of International Studies at the University of Washington from 1991-95.

David A. Lipton is Senior Associate at the Carnegie Endowment for International Peace’s Global Policy Program. Mr. Lipton served in several senior positions at the Treasury Department from 1993-98, most recently as Under Secretary for International Affairs.

Ray Marshall is Audre and Bernard Rapoport Centennial Chair in Economics and Public Affairs at the University of Texas at Austin. Dr. Marshall was Secretary of Labor from 1977-81.

Norman J. Ornstein is Resident Scholar at the American Enterprise Institute for Public Policy Research and election analyst for CBS News. In addition, Dr. Ornstein writes regularly for USA Today as a member of its Board of Contributors and writes a column called “Congress Inside Out” for Roll Call.

Peter G. Peterson is Chairman of the Blackstone Group and Chairman of the Council on Foreign Relations. A former Secretary of Commerce, Mr. Peterson is Vice Chairman of the Federal Reserve Bank of New York, founding Chairman of the Institute for International Economics, and founding President of The Concord Coalition.

William R. Rhodes is Vice Chairman of Citigroup/Citicorp/Citibank. He is also Senior International Officer for Citigroup. Mr. Rhodes joined Citibank in 1957 and served in many posts there, including as Senior Executive-International, before being named Director and Vice Chairman in 1991.

Stephen S. Roach is Managing Director of Morgan Stanley Dean Witter & Co. and Chief Economist and Director of Global Economic Analysis. Prior to joining Morgan Stanley in 1982, he was Vice President for the Morgan Guaranty Trust Company.

Henry B. Schacht is Director and Senior Advisor of Lucent Technologies. Mr. Schacht served as the company’s first Chairman and Chief Executive Officer from 1995-97. Mr. Schacht was Chairman and Chief Executive Officer of Cummins Engine Company from 1977-95.

James R. Schlesinger is Senior Advisor at Lehman Brothers and Counselor to the Center for Strategic and International Studies. Mr. Schlesinger was Director of Central Intelligence in 1973, Secretary of Defense from 1973-75, and the first US Secretary of Energy from 1977-79.

George Soros is Chairman of Soros Fund Management LCC. Mr. Soros funds a network of foundations dedicated to building and maintaining the infrastructure and institutions of an open society.

Laura D’Andrea Tyson is BankAmerica Dean, Walter A. Haas School of Business, the University of California, Berkeley. Dr. Tyson was Chair of the Council of Economic Advisers and National Economic Adviser to President Clinton. She served on the President’s National Security and Domestic Policy Councils.

Ezra F. Vogel is Henry Ford II Professor of the Social Sciences at Harvard, Director of the Fairbank Center for East Asian Research at Harvard, and Director of the Harvard University Asia Center.

Paul A. Volcker is Henry Kaufman Visiting Professor at New York University’s Stern School of Business. He retired as Chairman and Chief Executive Officer of James D. Wolfensohn & Co., Inc., upon the merger of that firm in 1996 with the Bankers Trust company, of which Mr. Volcker became a director. Mr. Volcker was Chairman of the Board of Governors of the Federal Reserve System from 1979-87.

Vin Weber is a partner at Clark & Weinstock. Mr. Weber was a Co-Founder and Co-Director of Empower America, a non-profit organization that deals with economic, social welfare, and educational problems. He represented Minnesota in the United States House of Representatives from 1980-92.

 


Endnotes

Note 1: By “moral hazard,” we mean situations in which the availability of insurance from the official sector weakens investors’ and borrowers’ sense of responsibility for their own actions.  Back.

Note 2: A basis point is equal to one-hundredth of a percent; for example, the difference between a 10 percent and 11 percent interest rate is 100 basis points  Back.

Note 3: Special Drawing Rights (SDRs) are a reserve asset that the IMF can create by bookkeeping entries. Since the SDR was created to promote better management over the global stock of international reserves, it is an appropriate instrument for dealing with systemic liquidity problems.  Back.

Note 4: These plans are laid out in three working group reports issued in October 1998 by the Group of 22. The reports deal with enhancing transparency and accountability, with strengthening national financial systems, and with managing international financial crises.  Back.

Note 5: By “moral hazard,” we mean the provision of insurance by the official sector that weakens investors’ and borrowers’ sense of responsibility for their own actions.  Back.

Note 6: Some independent analysts have argued that the official statistics exaggerate China’s economic growth over the past two years.  Back.

Note 7: There are other ways of measuring the international integration of financial markets, including “law of one price” comparisons for similar assets, savings and investment correlations, and departures from optimally diversified portfolios. In general, these more sophisticated techniques also point to increasing integration over the past few decades.  Back.

Note 8: The last published data are for 1996.  Back.

Note 9: See Institute of International Finance, Report of the Working Group on Financial Crises in Emerging Markets (Washington, January 1999). By “potential” losses, we mean losses that would have materialized if lenders/investors had “sold” their claims at prevailing market prices as of a certain date (say, July or December 1998).  Back.

Note 10: ERM refers to the Exchange Rate Mechanism of the European Monetary System (prior to the adoption of the euro).  Back.

Note 11: The “yen carry trade” refers to a trading strategy of borrowing money short-term in Japan and investing it in higher-yielding instruments either in emerging economies or in mature markets.  Back.

Note 12: The IMF has just begun (with two pilot country cases) to experiment with “transparency reports,” which discuss a country’s compliance with international financial standards. There is no agreement as yet on what form future transparency reports should take, although there is apparently a reluctance to classify countries’ compliance into several ratings-type groups.  Back.

Note 13: By “overvaluation,” we mean a situation in which the currency is too strong to maintain the competitiveness of the country’s exports and import-competing goods.  Back.

Note 14: Another potentially contentious issue is whether the issuer of the single currency should be willing to compensate an emerging economy for some of the government revenue losses associated with replacing the national currency (losses in “seigniorage").  Back.

Note 15: As argued earlier, it is still premature to declare the Asian crisis “over.” Thus, the losses suffered by investors and other private creditors in the crisis could go higher. The recent difficulties associated with Daewoo’s restructuring illustrate the risks involved.  Back.

Note 16: One reason why the South Korean rescue package was so big relative to Fund quota was that South Korea’s quota in the Fund was widely regarded as much too low for its economic importance.  Back.

Note 17: Specifically, there is a requirement to hold an unremunerated fixed-term (usually one year) reserve at the Central Bank of Chile equivalent to a fraction of capital inflows of selected categories. The implicit tax rate declines with the permanence or maturity of the capital inflow. The tax rate has been altered over the 1991-98 period. It was set at 20 percent in June 1991; it was raised to 30 percent in 1992 and maintained there through June 1998. It was reduced to 10 percent in July 1998 and reduced to zero in September 1998, when there was a shortfall in capital inflows.  Back.

Note 18: Under the proposed new capital adequacy framework, loans either to banks with low ratings (from major credit-rating agencies) or to banks located in countries with low credit ratings would receive less favorable risk weights than banks with better credit ratings. This means that (unlike the existing framework) lending to poorly rated banks would not carry a regulatory capital advantage over lending to poorly rated corporates or sovereigns. At the same time, the Basle Committee still proposes in the new framework that loans to banks with an original maturity of six months or less receive more favorable risk weights than longer-term loans to banks. Hence, there still appears to be a bias toward short-term borrowing.  Back.

Note 19: When the IMF draws on the US credit line by lending US dollars to other countries, the United States receives an interest-bearing, liquid claim on the IMF (which is included in our international reserves).  Back.

Note 20: The International Organization of Securities Commissions (IOSCO) is a forum for international cooperation among securities regulators. It is the analogue to the Basle Committee on Banking Supervision but in the securities area.  Back.

Note 21: By the “atomistic” model of competition, we mean a market in which there are many buyers and many sellers, none of which alone has the ability to affect the price.  Back.

Note 22: The SDR is the IMF’s artificial currency. The SDR is a reserve asset that the IMF can create by bookkeeping entries. The IMF was given the right to create SDRs to give it control over the total stock of international reserves; it may create (or cancel) SDRs when its member governments concur in a judgment by the IMF’s chief executive (the managing director) that there is a need to adjust the global stock of reserves. Few SDRs have been created thus far, and they account for only a small fraction of international reserves. The value of the SDR can be obtained from the market values of a basket of key currencies (dollars, euros, yen, and pounds).  Back.

Note 23: At the Cologne Economic Summit in June 1999, G-7 finance ministers recommended that the IMF Interim Committee be given permanent standing as the “International Financial and Monetary Committee.”  Back.