Dealing with Currency Crises


Ramon Moreno

FRBSF Economic Letter 1999-11 | April 2, 1999

The currency crises of the 1990s—the European Union’s in 1992, Mexico’s in 1994, East Asia’s in 1997, and Russia’s and Brazil’s more recently—raise concerns for a number of reasons. They are not only hard on the countries experiencing them, but if they spread widely they also may disrupt the international flow of credit, hindering trade, investment, and GDP growth in the world economy.

Pacific Basin Notes. This series appears on an occasional basis. It is prepared under the auspices of the Center for Pacific Basin Monetary and Economic Studies within the FRBSF’s Economic Research Department.

The currency crises of the 1990s—the European Union’s in 1992, Mexico’s in 1994, East Asia’s in 1997, and Russia’s and Brazil’s more recently—raise concerns for a number of reasons. They are not only hard on the countries experiencing them, but if they spread widely they also may disrupt the international flow of credit, hindering trade, investment, and GDP growth in the world economy.

Currency crises reflect actions by investors that are very similar to bank runs or classic financial panics. They differ, however, in that currency crises are common, while bank runs have been relatively rare during the postwar period, particularly in industrialized economies. This Economic Letter discusses the similarities and differences between bank runs and currency crises as well as their implications for efforts to prevent future currency crises.

Why bank runs are rare

Banks do not fully back their deposits with reserves, which makes them vulnerable to financial panics or bank runs. During a bank run, each depositor fears that other depositors will withdraw their funds; since the bank does not have enough reserves to give all depositors their money, depositors race to be first to withdraw their money before the reserves are depleted. This process can drive an otherwise solvent bank to bankruptcy.

Bank runs have been limited by two institutional arrangements. First, the central bank serves as a lender of last resort; that is, it can increase the money supply and also lend funds to individual (typically solvent) banks that may be experiencing runs. A sufficiently large increase in liquidity can end the incentive to flee the banking system. The second arrangement is deposit insurance, which ensures that depositors’ funds are safe even if their bank fails. In the United States, deposit insurance explicitly covers $100,000 for each depositor in a given bank. In addition, implicit insurance operates in many countries where some banks may be considered “too big to fail;” thus, depositors (but not necessarily equity holders) are implicitly fully insured against any losses.

While lender of last resort and deposit insurance facilities explain why bank runs in industrial countries have been rare in the postwar era, they also create an incentive for banks to take on more risk than they otherwise would—this is known as moral hazard—which can contribute to future panics. To curb moral hazard, banks face extensive regulation and supervision. Regulators directly assess the riskiness of loan portfolios and require that financial institutions insure themselves against failure by holding adequate amounts of capital.

Why currency crises are frequent

With the growing integration of financial markets, investors from developed market economies have acquired large claims in developing countries (for convenience I will use this last term also to refer to emerging markets). A central bank that pegs its exchange rate to a “hard” currency (one that is widely traded internationally) implicitly guarantees that any investors who want to exchange their local currency assets for that hard currency can do so at the prevailing exchange rate, that is, without experiencing a capital loss from a devaluation of the local currency. If investors fear a devaluation and think that the central bank’s foreign reserves are not adequate to make good on this guarantee, they may flee the currency, depleting hard currency reserves and forcing the devaluation they fear. Investors may fear devaluation for many reasons; for example, a rise in bad loans may create doubts about the government’s ability to raise interest rates in order to maintain a peg. Such weaknesses in the financial system often play an important role (Moreno 1998).

Thus, just as commercial banks are vulnerable to panics because they can run out of cash reserves, central banks that peg the exchange rate are vulnerable to panics because they can run out of foreign exchange reserves. However, in contrast to commercial banks from industrial countries, central banks in developing countries have no automatic access to lender of last resort facilities in “hard” currencies. This is one reason currency crises have been frequent, while bank runs in industrial countries have not.

International lender of last resort?

In principle, industrial countries that issue hard currencies could prevent currency crises in developing countries by acting as lenders of last resort. They could adopt expansionary monetary policies or provide sufficiently generous access to hard currencies (directly or through international organizations like the International Monetary Fund (IMF)) to countries experiencing speculation against their currencies (Calvo 1994, Fischer 1999). However, industrial countries have had difficulty in fully assuming such a role for at least three reasons.

First, the monetary policies of industrial countries ultimately must be geared to their own national policy objectives. At times (for example, 1998, when industrial countries’ monetary policies eased), industrial countries’ policy objectives may help emerging markets experiencing liquidity pressures. However, conditions change, and the policy objectives of industrial countries and developing countries may not always coincide.

Second, developing countries cannot be required to adopt policies that may prevent currency crises. In contrast to commercial banks, which must submit to domestic regulation and supervision—in extreme cases they may be taken over by their regulators—developing countries are sovereign entities that cannot legally be required by foreign countries or international organizations to hold adequate reserves or to adopt macroeconomic and prudential measures that would make them less vulnerable to runs on their currencies. As a result, the access of developing countries to hard currency financing around times of currency crises has not been automatic, but has followed negotiations in which they agree to implement adjustment programs intended to facilitate recovery or prevent future currency crises. Developing countries typically accept adjustment programs after currency crises have broken out, but they have had little or no incentive to do so during good times.

Third, distributing the burden of adjustment in the wake of a crisis is very difficult. In deciding whether or not to provide hard currency support in the wake of a crisis, industrial countries and the international organizations they fund face a dilemma. On the one hand, such support can create moral hazard—that is, incentives for risky behavior on the part of developing countries and international investors. On the other hand, withholding support may mean delayed recovery in developing countries or, in extreme cases, the spread of crises to other countries.

The resulting difficulties in achieving smooth responses to crises are illustrated by the experience with the debt crisis of the 1980s. After this crisis broke out in 1982, the debt burdens of developing countries—already unsustainable—initially increased instead of declining. According to Dooley (1995), this reflected the unwillingness of bank lenders to forgive any portion of the debt and the unwillingness of the governments of industrial countries to assume any part of it. Ultimately, market forces and U.S. government initiatives that facilitated the securitization of developing country debt and its sale at a discount in the open market brought about reductions in debt burdens, the resumption of capital flows to indebted countries, and the resumption of growth in these countries in the 1990s. However, the process took years. Latin America experienced a “lost decade” in which living standards declined and the trade and investment opportunities offered by the region were very limited.

New approaches

The experience with currency crises in the 1990s has prompted a search for new approaches for dealing with such crises. So far, the emphasis has been on the timely mobilization of large amounts of money and on crisis prevention.

More money. The financial rescue packages put together in response to crises in the 1990s are very large by historical standards, and they have helped end instability in currency markets relatively quickly. For example, the rescue package for Korea in the last quarter of 1997 totaled $58.4 billion (the IMF share was $21.1 billion), and an additional $22 billion in short- term debt (about 36% of the total short-term debt) was almost immediately rescheduled. While the package did not prevent a painful economic contraction, it stabilized Korea’s exchange rate a few months after the crisis broke out and set the stage for an expected recovery in the Korean economy in 1999.

Financial packages are now being geared to encourage the adoption of policies that could prevent crises in selected developing countries. Backed by a recent IMF quota increase of about $90 billion (including $17.9 billion from the U.S.), an enhanced IMF Facility is being considered to provide a contingent short-term line of credit that would be available before a crisis breaks out, but only if a country adopts certain policies that would limit its vulnerability. The line of credit is expected to be of short maturity and to charge interest rates above market rates to discourage misuse.

Another important feature of the new financial packages is that they do not rely exclusively on IMF funds. Governments and other international organizations are to make financing available along with the IMF, and private investors are encouraged to participate in providing financial relief. While the precise details are still being worked out, this may include arrangements which ensure that private investors or lenders do not flee financial markets during periods of uncertainty and instead voluntarily reschedule their loans to prevent or attenuate crises. As such arrangements would impose costs on private creditors in the aftermath of a crisis, they are also seen as a way of curbing moral hazard and preventing future crises. Recent financial rescue packages—including those for Korea in 1998 and Brazil in 1999, already contain some of these elements.

Crisis prevention. While large money packages can prevent or end panics in the short run, they also can create moral hazard, increasing the vulnerability to future crises. To avoid such an outcome, recent rescue programs have called for the adoption of structural reforms in the financial and corporate sectors, as well as improved prudential supervision. The idea is to improve risk management as well as the overall operation of the financial sector, thus reducing the chances of the type of panic that can trigger currency crises. In line with this, legal reforms in some countries that have experienced crises now allow more efficient and well-capitalized banks or firms to take over bankrupt institutions. Furthermore, sectors previously closed to foreign investment have been opened. Reforms also aim to enhance transparency in government, the financial sector, and the corporate sector in order to increase accountability and facilitate the assessment of risk.

Some actions are also contemplated at the international level. German Bundesbank President Hans Tietmeyer recently proposed enhancing the surveillance of the global financial system to reduce the chances of future crises. The enhancements include improving the information provided by financial institutions as well as national authorities, strengthening the coordination of national and international regulatory bodies, and assessing whether currently unregulated financial institutions should be regulated. Tietmeyer also proposed the creation of a G-7 (Canada, France, Germany, Italy, Japan, U.K., and U.S.) Financial Stability Forum that would identify and oversee actions to prevent crises.


While it is too early to assess their effectiveness fully, recent measures to strengthen the international response to currency crises appear to be facilitating the early recovery of countries affected by such crises and may also contribute to their prevention. Nevertheless, some of the obstacles to the smooth provision of international lender of last resort support that were cited earlier will remain. Under these conditions, developing countries may be well advised to give the highest priority to making their economic systems less vulnerable to crises. Experience with recent currency crises (Moreno 1998) suggests that countries may reduce their vulnerability by developing robust and well-supervised financial systems, with well-capitalized financial institutions that are more able to absorb shocks. Eliminating the government guarantee implied by a fixed exchange rate, by allowing greater exchange rate flexibility, may also reduce the likelihood of future currency crises.

Ramon Moreno
Senior Economist


Calvo, Guillermo. 1994. “Mexico: Stabilization, Reform and No Growth. Comments.” Brookings Papers on Economic Activity, 0(1), pp. 298-303.

Fischer, Stanley. 1999. “On the Need for an International Lender of Last Resort.” Manuscript.

Dooley, Michael P. 1995. “A Retrospective on the Debt Crisis.” In Understanding Interdependence: The Macroeconomics of the Open Economy, ed., Peter Kenen. Princeton: Princeton University Press.

Moreno, Ramon. 1998. “What Caused East Asia’s Financial Crisis?” FRBSF Economic Letter 98-24 (August 7).

Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Anita Todd and Karen Barnes. Permission to reprint portions of articles or whole articles must be obtained in writing. Please send editorial comments and requests for reprint permission to