At their next summit in Halifax this June, the leaders of the Group of Seven (G-7) industrialized nations have set themselves the task of revising the Bretton Woods framework, which is widely credited as a foundation of the postwar economic prosperity among Western nations. High on their agenda should be reform of one of its main components, the International Monetary Fund.

Why reform the IMF? The current international economic environment is more hopeful than ever before. For the first time, the overwhelming majority of the world's population lives in some form of market economy. The industrial world enjoys a rare combination of growth and low inflation; the "Washington consensus," a model of economic development that emphasizes macroeconomic discipline and open markets, is being adopted by more countries. The IMF played crucial roles in the 1980s debt crisis and in the transformation of former communist economies. Radical change, many might argue, is neither necessary nor desirable.

Such complacency is misplaced. There is a gulf between the rhetoric and reality of the IMF's role, a gulf that has been emerging since the fixed exchange rate system broke down in the early 1970s but which is proving increasingly hazardous. The growth of capital markets has rendered the organization impotent in industrialized countries; the world's richest economies neither borrow from the IMF nor are they required to follow its policy advice. Its role in the developing world is worrisomely unclear. The Mexican crisis earlier this year is a case in point. The IMF proved wholly inadequate at crisis prevention (it did not foresee the Mexican debacle), and its attempts at crisis resolution were dangerously improvised (it pledged a disproportionate share of its liquid resources to Mexico, breaking all existing rules on the limits of financial support).

In the world's poorest countries it has effectively become a development institution with a narrow macroeconomic focus. This role is a far cry from the original notion of providing countries with temporary financial support and raises the question of overlap with the World Bank. The IMF is floundering as it looks for a role.


The goals of the founders of the IMF were breathtakingly ambitious: to create a system that would foster prosperity through stable exchange rates and free trade. The IMF was a key part of the Bretton Woods vision crafted just after World War II. It would oversee the fixed exchange rate system and provide short-term support to countries facing financial difficulties. Half a century later, the world economy has changed beyond recognition. It is almost a cliché to point out that a key characteristic of the global financial system in the 1990s is the size, complexity, and speed of international capital markets. The breakdown of the Bretton Woods system of fixed exchange rates in the early 1970s, the removal of capital controls and, most important, technological and financial innovation have transformed global finance. No longer are the world's savings segmented by country; investors can seek out the best returns in ever more markets and hedge their risks with increasingly sophisticated financial instruments.

The geographic scope of the global financial system is also extending. As developing countries open up their economies, cut their budget deficits, and privatize state assets, they attract foreign capital. Private financial flows to emerging economies soared to $160 billion in 1993 from less than $42 billion in 1989, with the increase dominated by foreign direct investment and portfolio investment. Much of the explosive growth in capital flowing to emerging markets in the early 1990s was related to low interest rates in industrialized countries. As a result the growth rate slowed sharply last year, when rising interest rates made the returns on traditional assets like U.S. Treasury bills more attractive. In the wake of the Mexico crisis there will doubtless be a temporary reversal of the flow of funds as the euphoria over "emerging markets" is tempered. But the long-term trend is clear. While developing countries will have to rely primarily on their own savings to finance investments, there is no doubt that freer capital flows are a natural corollary of free trade. The size of capital markets in industrialized countries is on the order of $35 trillion. Shifting even one-half of one percent of this portfolio to emerging economies would more than double their current annual flow of private foreign finance.


Exchange rate management among the world's major economies is almost impossible in the face of this private-sector financial muscle. As financial integration continues, and bond and equity markets become increasingly integrated across countries, macroeconomic management by an individual country will be increasingly constrained. Sweden and Italy, for example, were forced by the market to compensate for fiscal profligacy with high interest rates; Mexico's lax monetary policy was punished even more dramatically.

Another challenge is posed by the seemingly new ways in which the financial system can break down. The possibility that an isolated collapse can precipitate the breakdown of an entire system is inherent in all financial systems. The debt crisis of the 1980s posed such "systemic risk." The growth of bond and equity-based finance should, in theory, reduce systemic risk. If a country defaults on its bonds or if its stock markets collapse, individual investors face the loss. However, as Mexico showed, bondholder crises bring new problems. Markets are liable to panic.

In addition, there is the "tequila effect": the risk of contagion when a sudden loss of confidence in one market precipitates the collapse of other markets as investors try to recoup their initial losses by selling elsewhere. After the Mexico crisis, markets from Bangkok to Buenos Aires suffered as investors abandoned emerging markets and "fled to quality". The result of such contagion is apparently irrational market behavior, where countries in good economic shape suffer from crises elsewhere.

Fears of such volatility will doubtless have some beneficial effect: they will prod developing countries into increasing their own rates of saving. But countries may see too many costs in the "Washington consensus" of economic reform and be tempted to return to less open or even autarkic economic systems. Fostering global integration means ensuring that reformist countries continue on the liberalizing path.

Finally, many countries are still excluded from the new world of private finance. More than 80 percent of the private capital that has surged into developing countries since 1989 has gone to only 20 countries. Many countries, largely those in Africa, have yet to benefit from the private-sector development revolution. Political instability, war, and bad economic policy have prevented them from attracting foreign funds. Heavily indebted, they still rely on a near-stagnant pool of foreign aid. As industrialized countries lose the enthusiasm and the resources for providing foreign aid, these countries risk getting left out altogether.


The Bretton Woods vision - and its attendant role for the IMF - bears scant resemblance to today's world. John Maynard Keynes and his colleagues had a profound mistrust of market-based international financial coordination. The institution they founded was designed to manage a system of fixed exchange rates predicated on a world of low capital mobility. The IMF initially had three basic functions: overseeing a system of pegged (but adjustable) exchange rates, promoting currency convertibility to foster international trade, and acting as lender of last resort for countries facing short term balance-of-payments crises.

Some of these functions have changed, but the original goals remain surprisingly unaltered. Certainly the IMF no longer presides over a system of fixed exchange rates, since no such system exists. But it still aims to monitor the world's monetary system and provide financial support (with policy conditions attached) to needy economies.

For the world's richest countries, however, the IMF has lost most of its relevance. It collates useful statistics, undertakes useful economic research, and occasionally recommends economic policies that influence domestic debates. But the IMF largely lacks the power to constrain. No major industrial country has borrowed from it since the mid-1970s. In annually reviewing the health of domestic economies, its opinions are regarded more as henpecking than hectoring. Michel Camdessus, current head of the IMF, did little more than irritate the Germans with his public criticisms of the Bundesbank's policy of high interest rates.

Policy coordination among the world's major economies, to the extent that it occurs at all, is informal. During the 1980s, the G-7 made some attempts to formalize exchange-rate management -- the Plaza and Louvre agreements, for example. Tellingly, the IMF was barely involved. Nor did it have any role in the creation (or humiliation) of Europe's Exchange Rate Mechanism (ERM). Ironically, the IMF has become a victim of the success of its policies. Once a developing country gains (or regains) access to large amounts of private capital, it no longer requires IMF financial assistance. Private rating agencies then become the barometer of a country's economic health.

The IMF's focus has shifted to the developing world. The debt crisis of the early 1980s and later the collapse of communism provided new rationales for the IMF's existence and an impetus for adopting new methods. Though it still professed to be a source of short-term financial support, the IMF created longer-term lending programs in the wake of the debt crisis, recognizing that the financial difficulties of countries like Brazil, Argentina, and even Mexico could not be solved overnight. It forged for itself a central coordinating role_cajoling banks toward providing insolvent countries with new money in return for improved economic policies. It became both an police officer of economic policy and a mediator between debtors and creditors.

When a balance-of-payments crisis in an emerging market is caused by a massive outflow of short-term foreign capital (as it was in Mexico), the IMF's limitations are underscored: it may not have the speed or resources to cope. In the aftermath of December's bungled devaluation of the peso, Mexico reluctantly turned to the IMF. Negotiations for short-term balance-of-payments assistance were begun, but the markets were unimpressed. The peso continued to fall. As the specter of default loomed, the Clinton administration came up with a $40 billion package of loan guarantees. This eclipsed the IMF entirely. After the package failed to muster congressional support and the Americans drew up an alternative plan, Camdessus saw the chance to reestablish the organization's relevance, and the IMF pitched in $17.8 billion to the second rescue plan: a pledge far higher than its rules would permit, and one impossible to repeat for many other countries, at least with its current resources.

Camdessus' quest for relevance in the Mexican crisis wholly undermined the supposedly rule-based nature of the IMF, and it has been widely interpreted as an example of the IMF's increasing politicization. "Saving" Mexico is, after all, predominantly a U.S. concern: it shares the lengthy border along the Rio Grande, and American funds held most of the outstanding short-term Mexican debt. Nonetheless, Mexico is probably less an example of direct political pressure than of managerial opportunism.

In other cases, however, IMF politicization has been obvious. To help the former communist countries, the IMF lent money to them with unusually weak conditions. The rationale, quite reasonably, was that these countries faced problems of economic stabilization and structural reform that were qualitatively different from those of other IMF members. The reality, however, is that the IMF and the World Bank were tapped by the G-7 as the best vehicles for Western assistance. Tight budgets (and lack of political will) precluded a new Marshall Plan. Shoring up reformers has become a political imperative that overrides many of the technical rules. It would be naive to expect any multilateral institution not to reflect the political concerns of its major shareholders, and it would be churlish to suggest that fostering a transition to capitalism is not a crucial Western responsibility. But it would be equally foolish to pretend that this role fits neatly under the IMF's original mandate.

Another area where the IMF's practice differs from its rhetoric is in the world's poorest countries, those that rely on the drip feed of foreign aid. Here the IMF has become an integral part of the aid machine, its presence permanent and its stamp of approval a prerequisite for much foreign assistance. In these countries the IMF's role (like the World Bank's) is to promote "structural adjustment," the ubiquitous phrase that describes the policy package of liberalization and privatization that is designed to help poor countries integrate into the world economy. These issues are the bread and butter of development economics; their complexity and long-term nature, however, sit uneasily with the IMF's purported mandate of addressing balance-of-payments problems in the short term. The result is that many criticize the IMF for having an excessively short-term horizon and concentrating too much on cutting budget deficits and controlling credit while neglecting the core issues that affect a country's capacity to produce goods and services.


The mixed IMF record of success, ad hoc improvisations, and adaptations has led to calls for reform. But of what sort? Combining the work of the IMF and the World Bank, as some have suggested, might be more efficient, but it would probably result in a loss of focus. The relatively small, disciplined IMF would become entangled in the bureaucratic sprawl of the World Bank.

Another suggestion is to reorient the IMF toward a more central position in economic policy coordination among major economies. The Bretton Woods Commission, a independent review panel of financial experts chaired by Paul Volcker, former chairman of the U.S. Federal Reserve Board, stressed the need for greater economic convergence and stability among major economies and called for the IMF to focus again on international monetary issues, its original mandate. In particular, the commission felt the IMF should become a locus for coordinating exchange rate management, leading to a more formalized system of exchange rates.

This is misguided on several counts. First, the global prosperity that accompanied the Bretton Woods system of fixed exchange rates has become erroneously identified with the Bretton Woods institutions themselves. The world did have both record growth rates and fixed exchange rates between 1950 and 1970. But the Bretton Woods system was held up by America's willingness to underwrite it; the IMF was little more than a facilitator. The system broke down in the early 1970s because the United States was no longer willing to subordinate domestic policy concerns to international monetary stability and because the growth of private capital flows made a fixed rate system untenable. The financing of the Vietnam War and the growth of the Euromarkets forced the demise of Bretton Woods.

It is increasingly difficult to see how a system of fixed, quasi-fixed, or openly targeted exchange rates among the world's biggest economies could work. Many economists have argued that global capital markets are pushing the world toward one of two extremes: a system of floating rates or a single unified currency. Quasi-fixed regimes are simply no longer possible. Even if a less extreme view is taken_that greater policy coordination would yield greater exchange rate stability_it is not clear that there is currently much political appetite for this among the IMF's major shareholders, the G-7 countries.

Since the Mexican crisis, calls for reform have come from a new direction. The IMF traditionally provided limited and conditional financial support to countries with no other outside source of finance. During the debt crisis of the 1980s this role broadened to include acting as a coordinator between banks and the insolvent developing countries they were exposed to, providing limited catalytic financing, and helping to correct economic policies that lay at the heart of the debtor countries' crises. The policy measures, rather than the amount of money, were key.

In Mexico, however, the goal was different. The American rescue package sought to prevent default by shoring up confidence in the market. In effect, the aim was to act as lender of last resort ex ante: prevent a financial collapse by lending large amounts of money up front. Such an approach poses difficulties. One is the amount of money required. As more countries become integrated into the global capital market, it is harder to foresee how much money would be required to underwrite such liquidity crises. A broad concept of lender of last resort creates an obvious moral hazard. The United States and the IMF in effect bailed out bondholders in the Mexican crisis. If such an approach were systematized, emerging economies would be tempted to pursue bad policies and international investors to fund them, both secure in the knowledge that they would be bailed out.

Also fraught with difficulty is the idea that the IMF should be able to lend large amounts of money on short notice to a country suffering an "unjustified" speculative attack. Camdessus proposed such a role after the debacle involving Europe's Exchange Rate Mechanism; he has pursued it with renewed vigor after the Mexico crisis. Though conceptually appealing, such a notion assumes that the IMF understands a country's economic fundamentals better than the market. It would try to prevent the natural reappraisal of risk that follows any financial crisis. The practicality of such an approach is prohibitive. How much money would be required to instill confidence in the markets? How much money would it take to convince George Soros and his ilk that speculating on a particular currency was not worthwhile?

These reform proposals are trying to fulfill the IMF's mandate in a profoundly changed world. The modern financial system cannot be "managed" by an international institution. In a world of high capital mobility, the traditional emphasis on exchange rate management and the provision of liquidity are both less feasible and less appropriate.


Acknowledging that the global financial system is overwhelmingly market-driven does not reduce the IMF to irrelevance. It has important roles to play in two main areas: mitigating the risks and failures of financial markets and integrating new players.

Reducing risk requires better monitoring of capital flows and economic policy. This would allow the IMF to play a far greater role in crisis prevention, particularly in countries whose access to capital markets is more recent and at greater risk. The IMF should promote the timely publishing of key economic indicators and rate countries on the quality of their economic statistics. Its judgments about economic policy should be made public. In today's world, greater openness by the IMF would provide more reassurance to markets than secret negotiations about policy advice. In essence, the IMF should become more like a public rating service, providing financial and economic analyses of stability and risk.

Mexico has shown dramatically how severely policy mistakes are punished by the market. In such situations the IMF should fulfill two objectives. It should first try to minimize market overreaction through helping a country announce rapid corrective policy measures. The second goal should be to concentrate on finding ways to develop private sector liquidity. In Mexico, for example, short-term debt could have been rolled over into longer-term maturities backed by oil reserves. Had there been no prospect of an official bailout, investors would have found such a solution preferable to default. In any market panic, it is in the interest of every individual investor to exit as quickly as possible. Providing a bailout merely enables them to do so at others' expense.

In extreme cases, where a country faces insolvency rather than illiquidity, the IMF should play a central role in coordinating a debt restructuring. An analogy first proposed by Harvard economist Jeffrey Sachs is that of a bankruptcy court judge: the IMF would able to announce a debt moratorium, declare fresh "working capital" a higher priority than obligations to previous creditors, and organize the restructuring of existing obligations. Such a role would give the IMF increased power in relation to both official creditors and private markets, but, if operated according to strict rules, it would improve the efficiency of debt workouts. The beginning of this kind of role was seen in the debt crises of the 1980s.

A fund that concentrates on promoting prudent macroeconomic policy and bankruptcy adjudication will retain a distinct role in integrating countries that still do not have access to private capital. Countries remain outside the private financial system because of poor economic policies, because heavy debt burdens render them uncreditworthy, and because they lack the legal and institutional infrastructure private capital requires. In former communist countries the greatest challenge is reorienting centrally planned economies. In Africa, political instability and corruption are at the root of the problem. In all cases a government's commitment to reform is crucial; the IMF's role should be to help it achieve macroeconomic stability and, where necessary, play a central role in working out debt.

Even here, though, the IMF's financial role should decrease. Given the declining Western appetite for foreign aid, this may seem an odd conclusion. Surely the world's poorest countries will need access to official funds for some time. The lesson of the last 20 years, however, is that balance-of-payments support to provide short-term relief is not the answer. More often than not the money is misspent, resulting in higher debt burdens, which inhibit growth and integration. The long-term funding that the poorest countries need should be targeted to specific areas like education, health, and so forth. These cheap loans should be provided by the World Bank and accompanied by specific reforms in these sectors. Along with other bilateral foreign assistance and debt relief, they should be contingent on ongoing IMF approval of a country's reform package. The IMF should be the police officer promoting policies that allow rapid integration into private capital markets; the goal should be economic integration rather than direct poverty alleviation.

An IMF that develops along these lines may seem a pale shadow of the Keynesian vision. It would neither manage the world's monetary system nor be a major lending institution. However, it would address the challenges faced by today's financial system. Far from becoming irrelevant, it would have a key role in fostering integration and mitigating risk. In the modern world these will be crucial functions in the promotion of global prosperity. The spirit of the Keynesian vision remains; its interpretation should be updated.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Zanny Minton-Beddoes writes on international economics for The Economist. She was on the staff of the IMF from 1992 to 1994.
  • More By Zanny Minton Beddoes