Journal Issue

ASDB/World Bank/IMF conference: Policymakers face major challenges in tracking and limiting spread of financial contagion

International Monetary Fund. External Relations Dept.
Published Date:
January 2000
  • ShareShare
Show Summary Details

Over the past decade, the global economy has been rocked by major financial conflagrations that started out as local brush fires. Many of these fires proved uncontainable, spreading rapidly and destructively across boundaries and flaring up on continents half a world away, fueling fears of systemic failure and global meltdown. Policymakers at both the national and international level have been searching for ways to prevent the outbreak and spread of financial crisis or, at least, to limit its duration and reduce the severity of its effects. A recent conference cosponsored by the World Bank, the Asian Development Bank (ASDB), and the IMF, entitled “International Financial Contagion: How It Spreads and How It Can Be Stopped,” addressed these issues.

Theories of contagion

In a session chaired by Guillermo Perry of the World Bank, participants considered some of the conceptual and empirical issues related to contagion. In their analysis, Garry Schinasi of the IMF and Todd Smith of the University of Alberta noted that theoretical contagion models rely on market imperfections to explain why adverse shocks in one asset market might be associated with asset sales in unrelated markets. However, basic portfolio theory can explain contagion, without recourse to market imperfections.

Portfolio diversification and leverage, Schinasi said, may be sufficient to explain why investors would find it optimal to sell many higher-risk assets in response to a shock to one asset.

Also questioning the presence of contagion, which they defined as “a significant increase in cross-market linkages after a shock,” Kristin Forbes and Roberto Rigobon, both of the Massachusetts Institute of Technology Sloan School of Management, noted that recent empirical work has found little evidence of this phenomenon during crises. They concluded that many countries are highly interdependent, regardless of the current economic climate, and the strong cross-country linkages that exist after a crisis are not significantly different from those during more stable periods.

Building on the issues raised by Forbes and Rigobon, Matt Pritsker of the Board of Governors of the U.S. Federal Reserve System argued that the channels through which a shock in one country or market spreads to other countries or markets are not well understood. But this does not imply that markets are irrational. Rather, more channels for propagation need to be theoretically modeled and empirically tested. In his paper, he considered several channels—including a common lender and the interaction of financial institutions and markets—and found that none of them require irrationality for shocks to be transmitted. Nevertheless, excessive price movements in one market or country may arise as a result of contagion from elsewhere, due to market imperfections involving information asymmetries.

Roberto Chang of the Federal Reserve Bank of Atlanta and Giovanni Majnoni of the World Bank attempted to identify and evaluate the public policy implications of financial contagion. On the domestic level, Chang said, they emphasized the need for policies aimed at reducing financial fragility—for example, reducing unnecessary short-term debt commitments. At the international level, the authors stressed the need for improving financial standards.

Measuring contagion

In a session chaired by Yung Chul Park of Korea University, theoretical issues were explored further. Looking at the experience of 20 countries with three crises—debt (1982), Tequila (late 1994-95), and Asian “flu” (1997)—José De Gregorio of the Universidad de Chile (and Minister of the Economy for Chile in the new administration taking office on March 11) and Rodrigo O. Valdés of the Central Bank of Chile asked what the “propagation” channels of crises across countries are and whether there are useful policy instruments that countries can use to shield themselves from contagion. They concluded that although crises may be triggered by common shocks, transmission of their effects across countries depends on regional trade, but also on the countries’ macroeconomic characteristics—current account deficit, level of the exchange rate, and credit conditions. The increased financial integration in the world suggests that contagion is more frequent and pervasive now than during previous episodes of financial turbulence, but although the causes of the crises may have changed, the channels through which contagion spreads have not.

Considering the policy actions that countries might take to narrow the transmission channels in times of crisis, De Gregorio said they found no evidence that capital controls are effective. In contrast, exchange rate flexibility and a maturity structure heavily weighted toward long-term debt did have a limiting effect.

The literature on contagion, according to Graciela Kaminsky of George Washington University and Carmen Reinhart of the University of Maryland, has focused on attempts to document its presence and to discriminate among the possible transmission channels of financial disturbances. However, the global or regional consequences of a disturbance may depend on where the shock originates. Analyzing the daily behavior of financial indicators for 35 countries, Kaminsky and Reinhart reached a preliminary conclusion that the global or regional consequences of a disturbance depend importantly on whether the shock originates in the periphery or in the center. In the case of the Asian crisis, for example, Japanese banks’ exposure to Thailand played a prominent role in the spread of the crisis. If the shock never reaches the center, however, it is unlikely to become systemic, Kaminsky said. In the case of Brazil, for example, financial market participants at the center countries were largely positioned for a devaluation of the real by early 1999; as a result, this crisis was “more of a ripple in global capital markets than a tidal wave.”

Barry Eichengreen and Galina Hale, both of the University of California, Berkeley, and Ashoka Mody of the World Bank looked at how negative shifts in market sentiment are felt in the bond market and the country characteristics associated with the damage. They found a role for changes in fundamentals and market sentiment following three crises of the 1990s (Mexico, Asia, and Russia). However, Eichengreen said, although there is ample evidence of changes in market sentiment, the dominant effect is within the region in which the crisis originates and on prices rather than quantities or maturities. Only in East Asia following the 1997 crisis did changes in market sentiment have a large impact on bond issues. The observed decline in bond flotations following the outbreak of the Mexican crisis can be fully ascribed to deteriorating fundamentals, according to the authors, as can, “to a surprising extent,” the increase in spreads in Asia. The authors do not, therefore, deny a role for the “flight to quality” as a channel for contagion, Eichengreen said, but suggest, rather, that its impact may be more limited than often assumed.

Role of investors

The role investors play in spreading contagion was explored in the session chaired by Jung Soo Lee of the Asian Development Bank. The Asian crisis has highlighted the importance of strong domestic financial systems, according to Linda Goldberg, presenting the paper she wrote with Gerard Dages and Daniel Kinney, all of the Federal Reserve Bank of New York. There is less agreement, however, on the role of foreign banks in contributing to strengthened financial systems. Both Argentina and Mexico had opened up their financial systems to direct foreign participation through the ownership of local financial institutions, she said. The experiences of both countries suggest that foreign ownership of domestic financial institutions contributes positively to the overall level and stability of domestic credit.

International banks are a major source of financing for emerging economies, but they are also among the most volatile, according to Caroline Van Rijckeghem of the IMF and Beatrice Weder of the University of Basel. Banking centers may add to financial contagion, Weder said, when two countries that depend on a common bank lender are vulnerable to spillovers through this linkage. Spillovers through such “common bank lenders” were important in transmitting the Thai currency crisis, and possibly the Mexican and Russian currency crises as well. In the Russian crisis, however, the withdrawal of funds seems to have been more generalized.

It is often argued that foreign outflows lead to price overreaction and contagion, but many financial economists contend that international investors do not create or exacerbate crises; their trading behavior simply reflects their assessment of underlying fundamentals. To test this contention, Kenneth A. Froot of Harvard University, Paul G.J. O’Connell of FDO Partners, LLC, and Mark S. Seasholes of Harvard University explored the behavior of daily international portfolio flows into and out of 44 countries from 1994 through 1998—encompassing almost 4 million trades by large institutional investors. Among their findings, O’Connell said, was that international investors engage in positive feedback or “trend chasing”—that is, an increase in today’s returns leads to an increase in future flows. In addition, O’Connell said, inflows have a positive forecasting power for future emerging market returns. Interestingly, their data revealed that international investors did not abandon emerging markets during the Asian crisis, despite the steep decline that took place in emerging market equity prices.

Analyzing the behavior of another important group of investors, Graciela Kaminsky, Richard Lyons of the University of California, Berkeley, and Sergio Schmukler of the World Bank examined the role of mutual funds in spreading crises. They found that mutual fund managers take account of the features of a country’s economy and financial markets when deciding whether to adjust their exposure in that country. Economic fragility was not the only factor triggering withdrawals from emerging market economies—the decision also hinges on liquidity.

Case studies

The experience with contagion of individual countries, groups of countries, emerging market economies, and transition economies was considered during two sessions devoted to case studies, chaired by Michael Mussa, Director of the IMF’s Research Department, and Gerard Caprio, Director of Financial Strategy and Policy, World Bank.

  • R. Gaston Gelos and Ratna Sahay, both of the IMF, looked at financial spillover patterns since 1993 in the transition economies of Central and Eastern Europe, Russia, and the Baltics and concluded that with increased financial market integration, the financial markets of the more advanced transition economies were behaving more and more like their Asian and Latin American counterparts.

  • Carlo A. Favero and Francesco Giavazzi, both of Bocconi University, Milan, found evidence of contagion in the 1992 exchange rate mechanism crisis (ERM) following the June 1992 referendum in which Danish voters rejected the Maastricht treaty.

  • The analysis of Ilan Goldfajn of Pontificia Uni-versidade Catolica, Rio de Janeiro, and Taimur Baig of the IMF indicates that in the aftermath of the 1998 Russian crisis, Brazil’s residents led the flight of capital from their country and were subsequently followed by foreign investors, resulting in the collapse of the Brazilian real. The authors did not find support for the hypothesis that the contagion was triggered by some investors reducing their exposure to Brazil to balance their portfolio affected by losses from the Russian crisis.

  • Analyzing the contagion of Mexico’s national markets from the crises of the past three years, Santiago Bazdresch and Alejandro M. Werner, both of the Bank of Mexico, said that the quick policy action taken by the Mexican government in the face of the adverse external environment was sufficient to restore market confidence and was responsible for Mexico’s healthy economic performance in 1998.

  • Yung Chul Park and Chi-Young Song of Kook-min University noted that Korea’s initially strong fundamentals before the Asian crisis and its rapid post-crisis recovery suggested that herding behavior of foreign investors panicking from the financial crises in Taiwan Province of China and Hong Kong SAR sparked the exodus of institutional investors from Korea that precipitated the crisis.

  • Tilak Abeysinghe of the National University of Singapore investigated how the recessionary impulses generated by the Asian crisis were transmitted across the borders of the eight crisis-affected countries of East and Southeast Asia. Although transmission plays an important role in the medium run, the immediate economic contractions are largely a result of direct shocks attributable to pure contagion, he noted. The postcrisis period, therefore, is likely to see intense competition among the crisis-hit economies for export markets and foreign direct investment.

  • Michael D.Bordo and Antu P. Murshid, both of Rutgers University, examined financial crises over the past 120 years for evidence of transmission channels of contagion. They concluded that the emphasis on contagion effects in the recent literature is “clearly overblown.”

Roundtable on international financial reform

In a discussion chaired by Stefan Ingves, Director of the IMF’s Monetary and Exchange Affairs Department, panel members considered ways countries could shore up their defenses against the fallout from financial crises in other countries, as well as strengthen their domestic financial systems and macroeconomic policy actions to protect against crisis in their own economies. Rudiger Dornbusch of the Massachusetts Institute of Technology said ministers of finance needed to understand that, in the “new world out there,” everything is correlated. If something big happens, it happens everywhere, especially in the immediate neighborhood. Any government that operates a “sleaze house” is going to get hit hard. The only way to avoid the worst effects of contagion, he said, is to “clean up your house.” Countries that do not want to take any risks should get out of the domestic banking industry altogether, sell the banks to foreign countries, or offer offshore guarantees. The best way to avoid currency crises, Dornbusch said, is for a country to divest itself of its national currency altogether by establishing a currency board. An economy without a central bank and a domestic banking system, Dornbusch said, may not be able to avoid contagion altogether, but at least the effects would be less harmful.

Morris Goldstein of the Institute for International Economics offered a shopping list of actions countries and the international community could take. IMF lending rates for countries seeking financial assistance should be “risk-based.” Countries should get rid of unsustainable exchange rates, provide evidence of sound macroeconomic policy, and adopt prudent standards and regulations. When a country in distress comes to the IMF for assistance, the IMF should impose normal access limits if the crisis is not systemic. Any restructuring or debt workout must involve the private sector, and no category of debt (like bonds) should be excluded from the new contracts. If the crisis is systemic, Goldstein said the IMF should apply a rough rule of thumb: if the crisis is the fault of the country, then the country has to fix it (that is, enter into a program with the IMF). In the case of a temporary crisis that threatens the international system, Goldstein recommended the establishment of a new facility (to replace the IMF’s current Contingent Credit Lines), funded with SDRs, available only to developing countries, and activated by a majority vote of the IMF’s shareholders.

The advice countries are given is often very difficult to implement, according to Yung Chul Park. For instance, the dictum to avoid balance sheet problems is hard to carry out because the problems—like maturity mismatches and exchange rate pressures—don’t surface until the crisis hits. Emerging markets are enjoined to conform to Western monetary and fiscal standards, Park said, but foreign investors do not seem to care so long as the potential returns are promising. If emerging markets become too concerned about shoring up against crisis—say, by tightening policy whenever a current account deficit threatens—they may not be able to invest more than they are saving.

Prudent oversight, of both the domestic and the international financial system, can help make an economy more resistant to contagious financial shocks, according to Philip Turner of the Bank for International Settlements. However, regulators and supervisors face several dilemmas in framing their policies and determining their actions in practice. Setting universal standards is probably impossible because degrees of sophistication in the banking system and depth and variety in the financial markets vary widely among countries. As a result, supervision will, in practice, remain a responsibility of the individual country, Turner said.

The papers presented at the conference on international financial contagion are available on the World Bank’s website at

Other Resources Citing This Publication