Reforming the International Monetary and Financial System

8 Refocusing the Role of the International Monetary Fund

Alexander Swoboda, and Peter Kenen
Published Date:
December 2000
  • ShareShare
Show Summary Details

Despite the recent emerging market crisis, the forces of globalization continue inexorably to weave a web of interdependence and integration across the world. The precarious state of the world economy has tarnished the process of globalization, however, as doubts have arisen about whether the promise of globalization will be fulfilled. That promise is one where liberalization and integration, supported by the information revolution and the diffusion of technology, will power a significant, sustained increase in global growth. In turn, economic growth will produce a convergence in living standards between developing and industrial countries. Over the past two years we have seen crisis and uncertain recovery in Asia, collapse in Russia, strains across Latin America, and; apart from the United States, sluggish growth in the G-7. In April 1998, the IMF projected global growth at 2.3 percent for the year and growth in world trade at just 3.8 percent.1 The emerging market crisis has subsided, but the rekindling of interest in those markets on the part of international capital markets inspires worry as well as relief. That worry is whether the globalized economy will be stable enough to deliver on its promise.

Doubts about globalization coincide with doubts about the IMF. If the IMF calculated a global index of public and elite opinion about itself over its entire history, the graph would be in a trough. Of course, there have been peaks and troughs in the past. But while the IMF’s prominence in crisis management has boosted its name recognition, more people are more critical than ever. Interestingly, the criticism comes from both ends of the spectrum of views of capitalism. Some say that the IMF hinders capitalism, through both its advice (more taxes means more government) and its finance (more money means more moral hazard), and some of these people want the IMF closed. There are others who believe that the IMF promotes heartless capitalism, urging liberalization without regard to dislocation, labor impact, or environmental damage. IMF partisans can take heart that its opposition is divided. But, perhaps more than in a generation, the IMF lacks defenders outside of the official international financial community.

What has happened? The IMF’s problem is derivative of the problems arising from globalization. Remedies for the IMF must center on the institution’s contribution to redressing the problems of globalization. In particular, the proper role for IMF finance must be embedded in a strategy for the international financial architecture aimed at making the world safe for globalization.

The strains accompanying globalization stem, in part, from the speed in the spread of markets outstripping the response from the official community. That response includes policy and institutional development by national governments and the modernization of the international financial architecture by the international community. In short, the official community has fallen behind, and now part of the catch-up process entails reevaluating what governments can and should do and what must be left to the markets. National governments may be abandoning the commanding heights of their economies as they privatize, liberalize, and end government direction and other interferences. But that does not mean governments have no role or even that they are less important. Rather, it means new responsibilities and greater discipline from markets.

Governments must now preside over a process of strengthening the institutional and policy bases of their economies to make them hardy enough to withstand and then benefit from globalization. Part of that process involves destruction—tearing down the remnants of state or crony interferences in the economy; and part of it involves creation—building newer and stronger legal and regulatory frameworks; but a third part involves shifting the burden of responsibility for absorbing economic shocks from governments to the marketplace. The IMF can and should play a role in advancing every part of this process.

Those who see IMF finance as irrelevant or counterproductive to the functioning of markets are reacting to a role that the IMF has had to assume during this intense period of globalization. Rather than concluding that there is no proper role for the IMF in financing adjustment in its member countries, it is time to reexamine whether the world is asking too much when it asks the IMF to place its financial arsenal against the international markets. Perhaps refocusing the role of the IMF, in the new international financial architecture under discussion, can create a context in which the IMF can more successfully fulfill its role.

This paper presents an approach based on three proposals. First, the international community should end the large-scale IMF bailouts for member countries with balance of payments problems. That means retracting the implicit IMF guarantee that has evolved in recent years, and restoring a strong presumption of modest access limits for countries seeking IMF finance. Second, the burden for coping with economic problems should be shifted away from the IMF and to the markets and private sector participants. That means encouraging members to adopt (or, for recent converts, to maintain) flexible exchange rates in order to end the costly peg defenses we have seen. In this setting, policy adjustment and normal access to IMF finance should usually suffice. If not, the international community should encourage workouts with creditors. Third, while bailouts for countries should no longer be needed, the IMF should still be ready to respond if the health and integrity of the international financial system is endangered. That means creating a large pool of resources as a last line of defense, and creating a governance structure that allows for the use of these resources only at a very high threshold of systemic threat.

Origins of the Crisis

Where did the emerging markets crisis come from? What was the international community’s response? There are two camps of thought regarding the genesis of the crisis. One you might call “the IMF camp” argues that the crisis came from underlying problems in the affected economies, including deep flaws in the structure and regulation of domestic financial systems.2 After all, every crisis country, at least until the crisis spread to Brazil, saw a collapse of its banking system. In short, the problem was various forms of crony capitalism and the policy and economic consequences that resulted. The IMF camp calls for greater national prevention efforts, with the IMF catalyzing those improved efforts; international support for crisis management, including conditional IMF lending to provide breathing space for countries undertaking adjustment; and workouts involving private sector restructuring when problems are sufficiently severe.

The opposing camp, one you might call the “it’s a financial panic, stupid” camp, argues that the crisis was one of confidence, and that this is an inherent problem in modern capital markets.3 For that camp, technological advances in capital markets and poor international regulation were the villains in the saga. Concern about market psychology is not new to economics, but the modern version is updated to the new realities of the globalized economy. First, huge amounts of capital flowed into emerging markets—coming from a wide variety of new financial institutions, delivered by new technologies, abetted by regulators, and flowing with a force never seen before. Some argued that implicit guarantees of IMF support encouraged excessive risk-taking. Next, panic set in. And once panic set in, each investor was rational to get out if others were getting out.

The panic camp points out that just as each investor was smart to get out if others were getting out, if everyone were calmed, all would stay in. So this camp argues that in such a crisis no steps should be taken that undermine confidence or economic activity. When the IMF’s Managing Director pays a visit to a crisis country, his list of conditions has the effect of emphasizing weakness and impairing confidence. Hence, conditionality should wait. Instead, the international community needs to stop the panic by acting as a lender of last resort. The other change advocated by this camp is the regulation of capital inflows. Regulators in advanced countries should amend the Basel capital adequacy standards to stop encouraging short-term flows (via zero or low scoring of short-term loans against capital adequacy). And emerging market countries should discourage “undesirable” short-term inflows through prudential regulation or outright capital controls.

Well, was the crisis a flood of capital—first in, then out? Or were some countries “weaker swimmers” than others? The economics profession will be sustained for years discussing and writing about this question. In a speech on April 9, 1998, then-U.S. Treasury Secretary Robert Rubin concluded that weak policies and institutions in many developing countries, and an inadequate focus on risk on the part of banks and investors in industrial countries, combined together to produce vulnerabilities in these economies. It was this combination that ultimately led to the abrupt collapse in confidence that spread through Asian and other emerging economies from the summer of 1997 onward.4

The fact that there is truth in the views of both camps makes the response to the crisis complicated. The agenda for modernization of the international financial architecture that is emerging from all the meetings, conferences, and communiqués seems to acknowledge the duality of the problem. In brief, what has emerged as an agenda includes four basic parts.

  • First, industrial countries have said they will act, including using interest rate cuts, to keep their economies growing.
  • Second, the international community will promote crisis prevention. That means working with emerging market countries, including through new efforts to promote standards for good conduct, to improve the policy and institutional setting in their countries. And it means working in advanced markets to improve the regulatory framework within which lenders and investors operate.
  • Third, prevention will have a financial dimension. The IMF’s new Contingent Credit Line aims to help countries affected by financial contagion. The new facility is viewed by some as prevention and by others as a step in the direction of lender of last resort.
  • And fourth, there is very preliminary agreement that when a crisis is severe, the private sector will be expected to play a role in crisis resolution. But as many have acknowledged, in this world of Eurobonds and derivatives, creating new mechanisms to do an old job, and to do it without undermining healthy markets, will be a challenge.

Access to IMF Resources

The proposed agenda is broad and encompasses many of the key issues that need to be tackled. Implementing the agenda is likely to improve crisis prevention and management. One area that has been left indeterminate, however, is the scale of IMF finance for crisis management. This indeterminacy in the debate about access policy going forward surely derives from uncertainty about what the next crises will be, as well as a desire to maintain room for the international community to maneuver, and constructive ambiguity about IMF resource commitment vis-à-vis market participants. The issue is complex and the rationale for this indeterminacy is understandable. Nonetheless, the remainder of this paper argues for the need to put the genie of large IMF lending operations back into the bottle, and to restore a more rigid adherence to the use of access limits in approving members’ use of IMF resources.

This argument is put forward without suggesting that the international community was wrong to provide large-scale financing during the emerging market crisis. As fixed exchange rates collapsed and crisis spread, the IMF-led package was the tool available for combating crisis. The question, going forward, is whether crisis management can be improved by combining new approaches for the three most challenging items on the architecture agenda: IMF finance, exchange regimes, and private sector workouts. The international community is coming around to the view that greater exchange rate flexibility is desirable. It is also edging in the direction of incorporating private sector restructuring and workouts, into procedures for crisis management. If these two desirable steps are taken, they would go a long way toward permitting a return to limits on IMF lending. But more important, achieving those two desirable goals may depend, at least in part, on first limiting IMF lending.

In general, neither countries nor market participants should be led to presume that IMF resources will be available in large amounts. While country authorities have strong incentives to avoid crisis, limits on the availability of IMF resources could affect choices on the margin, helping to dissuade countries from using or excessively relying on exchange rate pegs. Market participants, in evaluating potential crisis outcomes, should have a clear idea of the limits of IMF support so that investment decisions are not subject to undue moral hazard. In a crisis, more rigid limits would provide clearer benchmarks for the adjustment and financing mix, as well as for the limits beyond which debt restructuring and workouts should be initiated. After all, part of the original rationale for limits was so that the IMF could catalyze private resources, instead of supplanting them.

Traditionally, IMF policy has been to limit financial support to a member country to a fixed percentage of quota per year and a fixed multiple of quota over a multiyear period. Before the most recent quota increase, the access limits were 100 percent of quota annually and a cumulative limit of 300 percent of quota. These limits have been adjusted from time to time as quotas changed and as members’ needs changed. There has long been recognition that in exceptional circumstances access could be granted above the limits, but before the Mexico crisis that was seldom done. Access was, broadly speaking, geared to creating an appropriate mix of adjustment and finance, where the scope of the balance of payments need came from current account difficulties and the need to maintain or restore central bank international reserves.

The crises of the 1990s have posed a new financing challenge for the IMF, in part because of the predominance of capital market pressures. In earlier decades, the IMF saw its role as helping to finance current account imbalances, thereby limiting the extent to which aggregate demand would have to be reduced through policy adjustments. In the Mexican crisis and the emerging market crisis, the IMF helped countries finance capital outflows to a much greater extent than before. Of course, that distinction is not exact in all cases, but it exists and stems from the expansion of capital market access among emerging market economies.

For the most part, official financing of large-scale capital outflows occurred where pegged exchange rates were attacked, defended, and abandoned, and where intense capital outflows caused extreme exchange rate depreciation and threatened widespread financial and nonfinancial bankruptcy. In most of those cases, current account deficits disappeared, or even turned to surplus as domestic demand plummeted. Responding to the huge scale of capital outflows and, in time, the depth of capital market pessimism, the international community agreed to permit crisis countries exceptional access to IMF funds (as well as funding from the development banks and bilateral sources). That response was not wrong. Pegged rates were failing, contagion was spreading crisis, the consequences for the international financial system looked ominous, and the IMF was the bulwark against contagion.

Can the system be changed so that the IMF can restore adherence to access norms? And, if so, what is the role for the newly created facilities, the Supplemental Reserve Facility (SRF) and the Contingent Credit Lines (CCL)?

As this paper will argue in some greater detail below, a more widespread reliance on flexible exchange rates would, as a by-product, likely reduce the number of countries needing IMF funding and the scale of IMF funding for those in need. Over the past several decades, countries have come to the IMF for help with macroeconomic imbalances, structural problems, transition from communism, and debt difficulties. Often countries have had a combination of these problems. In the absence of peg defenses, there would be fewer cases where macroeconomic imbalances and international reserve inadequacy lead to crisis. Greater reliance on exchange rate adjustment and less on reserve intervention would help prevent capital market runs on central bank reserves. For the most part, even intense capital market pressures stemming from doubts about balance of payments prospects would be relieved in a timely and adequate manner by exchange rate depreciation. Foreign exchange sales on a modest scale, funded by IMF loans, can play a constructive cushioning role in supporting imports, much as was the case in earlier crises, when policy adjustment, exchange rate depreciation, and the availability of modest IMF support successfully stemmed capital outflows. Such sales would constitute a temporary departure from a floating exchange rate regime, but could be preannounced, fixed amounts, and for a limited duration.

Floating rate regimes do not rule out episodes of extreme depreciation resulting from problems of debt sustainability, banking system failure, or other forms of widespread insolvency. Debt crises, however, should be less frequent than crises of failed pegs. Difficulties with exchange rate pegs are virtually as frequent as fiscal and monetary difficulties. Debt difficulties require a buildup of fiscal or real economic difficulties over a longer period, or significant real shocks. In a world of more floating exchange rates, debt buildups will trigger exchange rate pressures that will impart discipline for addressing the debt issue, and real shocks can be abated, at least in part, by depreciation. In those instances where debt problems arise, it is preferable to respond with debt restructuring rather than massive official financing.

The SRF was created in the context of crisis resolution. The IMF had been called upon to lead the international community’s response to crisis by placing large financial resources in support of the maintenance or restoration of capital market confidence, and the SRF was a constructive innovation. Its shorter maturities and higher interest rates, compared with traditional IMF facilities, properly improved incentives for borrowers to return to capital markets sooner. What is debatable, however, is whether the IMF should continue to provide such large-scale support as has occurred in the few instances the SRF has been activated. If access is scaled back, use of the SRF would diminish. Even with a return to more rigid access norms, there could still be instances in which the terms of the SRF are more appropriate than the terms of Stand-By Arrangements. But determining a set of guidelines for the SRF that are within normal access limits, in particular when countries might be expected to forgo stand-by terms, requires further consideration.

The CCL, on the other hand, offers a potential tool for crisis prevention. Provided that the CCL is reserved for countries with strong institutional bases and strong records of policy management, an “A-list” of countries could be established, each of which could gain access through its qualification for membership. Membership on the “A-list” would have value if it translated into improved capital market access and financing terms. That could include the establishment of private contingent lending operations, as envisioned in the CCL facility. If national authorities also perceive stronger incentives to adhere to the membership terms—that is, through better policy management—the CCL will prove an effective prevention device.

One obvious problem with the CCL, which has been identified by others, including the IMF itself, is maintaining proper standards for qualification. The value of the CCL will be diminished if mediocre performers are admitted, or if once-stellar performers cannot be de-qualified once their performance slips. That problem will not be easily managed. A more important problem, however, is that potential access under the facility has not been rigidly limited. The IMF decided that there would be no general access limits for the CCL and has said it expects access in the range of 300 to 500 percent of quota.5 This decision allows much greater leeway and discretion than has been traditionally the case under upper credit tranche lending facilities. The potential exists then for the CCL to become the target for bargaining, with the private sector attempting to pressure the international community into “upping its ante” in the prevention sweepstakes. It is instead preferable to strenghten CCL access limits and to set them at levels consistent with regular conditional lending. Then let the markets determine the value of combined modest finance and IMF seal of approval.

Exchange Rate Regimes

Until quite recently, the role of exchange rate regimes was curiously absent from the discussion of international financial architecture. The subject has been an uncomfortable one for the international community. The G-22 Working Group report on architecture avoided it. Emerging market participants tended to defend their particular regimes, and G-7 representatives did not want to open the door to a discussion of target zones for their currencies. The IMF view, if there is one, has been that exchange regimes must be tailored to countries’ individual circumstances. The academic community has many views—if you ask five economists, you get six views. Almost everyone agrees that the emerging market crisis shows the dangers of fixing exchange rates without complete policy commitment.6 Many have concluded that emerging market countries should go to one or the other end of the spectrum, either float freely or adopt very hard pegs, such as in currency boards, and avoid the murky middle of discretionary crawling pegs and bands.7

Former U.S. Treasury Secretary Robert Rubin weighed in on the subject, adopting the “ends of the spectrum” position, and proposed putting some muscle behind it.8 He suggested the international community should not “provide exceptional large-scale official finance to countries intervening heavily to defend an exchange rate peg.” He offered an important caveat, however, by admitting an exception “where the peg is judged sustainable and certain exceptional conditions have been met, such as when the necessary disciplines have been institutionalized or when an immediate shift away from a fixed exchange rate is judged to pose systemic risks.” That approach seems to point in favor of a new presumption against peg defenses, with a large dose of “optionality” for the international financial community, which will strike some as valuable and others as dangerous.

This new debate is helpful, because the question of what is the best exchange regime ought to be considered, as it has been historically, in conjunction with the question of what is the appropriate international financial architecture.9

The history of the international financial system has seen exchange rate regimes predominate for a period and then fall victim to stresses, strains, and crisis, as political and economic forces change and make the regime impractical or not useful. Bretton Woods, which was established by international convention, was an exception to the more general rule that regimes develop, gain adherents, last for some years, come under pressure, and then collapse in an evolutionary process. More often the leadership of a key country or countries, pledging themselves to a regime, was followed as others joined in consistent behavior.10 We are now at an interesting and fluid moment in the evolution of exchange rate practices.

The heterogeneity of practices in exchange rate regimes, in recent years, has followed from the logic that regimes should be selected on a case-by-case basis. The trade-offs involved are encapsulated in the often-cited device of the holy trinity of objectives: exchange rate stability to facilitate commerce and finance, monetary independence to manage domestic demand, and capital account openness to facilitate inflows. The trinity argument is that only two of the three objectives can be achieved by any exchange rate regime—fixed exchange rates sacrifice monetary independence, floating exchange rates sacrifice stability, and capital controls sacrifice openness. The argument also implies that no single exchange rate regime is optimal and the choice must be made on a case-by-case basis suited to the circumstances and policy imperatives of the country involved.

In the advanced countries, we have seen a period of floating among the major currencies and infrequent exchange market intervention. At one level, policymakers have presumed that foreign exchange intervention alone would have a limited impact given the awesome size of major currency markets. At another level, policymakers see that markets are extremely sensitive to signals of their policy intentions. What is really at play is that policymakers in advanced countries consider their main objective to be creating the conditions for strong domestic economies, and they are convinced that achieving sustained growth with low inflation is the best recipe for healthy external conditions. A corollary of this is reluctance toward any serious subordination of domestic instruments to external objectives. Europe is hardly likely to raise interest rates soon because of the weakening euro. Japan is desperately trying to figure out how to conduct an expansionary monetary policy, even though it might weaken the yen. And the United States, despite its trade imbalance, often repeats the refrain that it will not use the exchange rate as an instrument of trade policy.

In emerging market countries, a period of experimentation with adjustable pegged exchange rates has seen success in some cases, but in others stress, strain, and failure.

From the late 1980s onward, a number of countries had pegged their exchange rates to combat high inflation and instability, and some of those countries switched to crawling pegs and crawling bands as their stabilization efforts progressed. Other countries adopted adjustable pegs to gain policy discipline and to signal policy commitment. Since Mexico’s crisis began in late 1994, a large number of countries with adjustable pegs have suffered currency crises. In the prominent crisis cases, in Asia, Russia, and Brazil, but elsewhere as well, countries have altered their exchange rate regimes to build in more flexibility. In fact, since the Mexican crisis, about 15 percent of the non-G-10 membership of the IMF has abandoned pegged arrangements and adopted floating regimes.

What these countries have experienced is not something that is incorporated into the holy trinity framework—the costs of regime failure. Empirical analysis has confirmed that pegs impart and convey discipline and thereby contribute to price stability.11 Now, it is also clear that when pegs break down, the costs are high. Whatever the cause of the breakdown—macroeconomic policy errors, adverse shocks, or swings in capital market sentiment—once foreign exchange reserves are depleted it becomes very difficult for countries to reestablish exchange rate stability.12 Restoring confidence is extremely difficult without both credible policies and international liquidity. The resulting extreme exchange rate depreciation raises the local currency burden of foreign currency debt, and puts stress on banks, companies, and governments. Currency crisis converted into national bankruptcy dooms a country to a complicated process of domestic and international financial workouts that can be lengthy and debilitating. Worse, it has proven nearly impossible for some of the crisis countries such as Indonesia to organize workouts, given their political, legal, and institutional weaknesses.

The combination of the globalization of finance and the widespread adoption of convertibility in emerging markets in the last decade created the circumstance where defense of an exchange rate peg or its eventual failure could involve extremely large capital outflows and reserve decumulation. Capital flows to emerging markets rose dramatically in the period up through 1996. The breadth of exchange rate convertibility and the liberalization of capital account transactions both facilitated these inflows and permitted both domestic and foreign investors to engage in capital outflows as well. The implicit guarantee embedded in exchange rate pegs lowered the perception of risk for investors. And the depletion of reserves, especially in light of short-term debts and foreign currency obligations, created a combustible mixture.

Worse yet, the failure of pegged exchange rates has proven costly for the international financial institutions and the international community. As countries defended pegs, or attempted to restore exchange rate stability after depreciation of the exchange rate, they rightly asked the IMF for help. In Asia, the international community found out how expensive it was to help countries replenish depleted foreign reserve holdings and restore a reserve cushion that would be viewed as sufficient in light of short-term debts. Even the big packages were not immediately convincing, because the dire financial position in crisis countries left little time for national authorities to undertake needed policy and institutional changes. And the types of problems that needed fixing were not just macroeconomic imbalances, but deeper structural and institutional problems. The IMF faced two daunting tasks: designing extensive and intrusive economic adjustment measures without adequate information, and completing programs so quickly that there was inadequate time for deliberation. IMF programs, which in an earlier era might have been negotiated in weeks, were negotiated in days and approved swiftly under the Emergency Financing Mechanism. Despite the daunting sums—western support for the Asian crisis countries, Russia, and Brazil totaled $180 billion—depreciation was extreme and, in most cases, not reversed quickly, and recessions have proven to be deep.

With floating exchange rate regimes, it is less likely that we would see such explosive crises. Provided that rates truly floated—that reserves were not depleted in an effort to target a particular rate—several factors would mitigate crises. First, even if central bank credit and interest rate policies were partially directed at stabilizing exchange rates or at maintaining exchange rates in “desired quiet zones,” the absence of explicit pegs or bands would greatly reduce the implicit guarantee extended to market participants. Thus, policy short-comings and exogenous shocks would lead to fluctuation of exchange rates, and with them, asset price valuations and real terms of trade, which would diffuse the real impact of adjustment broadly in the domestic and global marketplace. As a result, routine exchange rate variability would lead market participants to enter more broadly into hedging contracts, redirecting risk from central banks to the markets, and probably from banking systems, to a broader range of market participants.

With the creation of the euro and talk of dollarization in some parts of Latin America, it may seem incongruous to speak of the benefits of floating exchange rates. Monetary union brings with it the surrender of monetary policy in the interests of lower capital costs, greater capital market integration, and hence higher economic growth. It is possible that the world will see greater monetary integration or even the creation of large currency blocs in the long run. Without entering into the complicated subject of optimal currency areas, monetary union will be workable for some countries where economic factors make it desirable, and where political and institutional factors make the commitment to live with the strictures possible. But in most cases, even if monetary union is a desirable long-term goal, it will make sense to approach the goal in a nonlinear way. That is, begin with a floating exchange rate, then sufficiently strengthen financial and legal institutions and regulatory practices and the conduct of macroeconomic policy, and then jump to a monetary union. This was one of the lessons of the European Monetary System crisis in 1992, following which bands were broadened for the remaining run-up to the euro.13 Even Poland, which aspires to adopt the euro, seems to have accepted this lesson by widening its band and introducing greater flexibility for its transition period.

The IMF’s reluctance to develop any systematic preference regarding exchange rate regimes is understandable. The IMF faces a difficult trade-off vis-à-vis its members, akin to the statisticians’ trade-off between type one and type two errors. The IMF errs if it fails to support a government committed to addressing its problems, intent on using a nominal anchor to convey its commitment, and willing to live with the implied constraints. Alternatively, the IMF errs if it supports a government proclaiming those ambitions, but unable or unwilling actually to do so. Hence, the challenge for the IMF in balancing these two risks is to support committed and able governments without being susceptible to costly defenses of less committed governments.

Given the costs to the international community when exchange regimes fail, the IMF has a clear interest in the choice of national exchange regimes and their management. After all, that was part of the logic for the creation of the IMF. How might the IMF proceed? First, the IMF should, through its surveillance, dissuade countries from adopting pegs, unless special circumstances make the use of a nominal anchor highly valuable. Those circumstances might include an effort to end a rapid inflation, where the local currency’s role as a nominal anchor has been demolished, or the case of very small, very open economies, with little need for monetary policy. Whenever those circumstances do not exist, the IMF should urge countries to adopt a high degree of exchange rate flexibility. Second, as former Treasury Secretary Rubin has suggested, the international community should refrain from supporting exceptional access to IMF resources in funding large financial packages to defend pegged exchange rates. But to keep Rubin’s optionality caveat from becoming a loophole for undesirable bailouts, all members should be restricted again to only ordinary access. Even committed governments fixing their exchange rates can receive IMF support, but not exceptional access. Denying extraordinary access limits the risk of costly, failed peg defenses, and limits the incentive for governments to opt for the peg (or to sustain an embattled peg) in order to gain access to large amounts of official finance.


A second key step toward making our global capital markets function better is to develop and implement an approach to the restructuring of private claims when crises are severe. The international community has spent several years discussing the need for orderly workouts (sometimes called bail-ins). Many have recognized this as perhaps the most difficult of the international financial architecture issues, given the difficult balance between not undermining debt service discipline and not undermining the market discipline of creditors in their evaluation of risks.14 Restructuring has occurred in Korea and Ukraine, and a preference has emerged for a case-by-case consideration of restructuring needs. So far, little progress has been made in developing a systematic approach to the subject under which cases could be dispatched.

On the face of it, the international community is attracted to the idea that workouts should occur. The spreads earned by creditors on financial instruments issued by emerging market countries raise the question, what risk other than restructuring or default risk could explain such prices? It is just not consistent for the international community to base its rescue packages on a presumption that all debt service to private creditors must always be paid. Markets, to some degree, are assuming that some mechanism for private sector involvement will be incorporated into workouts when economies have severe debt-servicing difficulties. Moreover, both the academic and official communities see the development of workouts as instrumental to the process of damping the excesses to which capital markets seem to be prone. The decline in emerging market spreads in the first half of 1999, and the resurgence of interest in emerging market bond issues, strengthens the logic of Treasury Secretary Rubin’s call for inducements “for creditors and investors to weigh risks more appropriately in decision making.”15

Moreover, history is replete with examples of debt restructuring. In each era of capitalism, there have been sovereign defaults and restructuring. Of course, while some have been orderly, most have been disorderly, with adverse consequences for borrowers and creditors alike. It would be historic if in the present era of globalization, with the sophistication of markets and the financial instruments they spawned, restructuring never took place.

Of course, the lack of progress in developing a broad approach results from a number of inherent obstacles. Many have expressed the concern that payments discipline not be undermined, that creditor moral hazard not be traded off for government moral hazard with the effect of retarding needed capital flows. That point illustrates the larger concern that we do not yet have a broad conceptual basis for distinguishing when workouts should and should not occur and what their terms should be. Economists have long analyzed bankruptcy in the domestic setting, but no “general theory of international workouts” has been enunciated.

The alarm over moral hazard, on the one hand, and the alarm over the sanctity of contracts, on the other, should in time give way to an analysis of the balance of factors at play. Beyond that problem are operational issues, such as who decides when a workout will occur and how workout terms are determined or negotiated. And there is a vexing set of issues surrounding the legal basis for securing the participation of creditors. Last, there is a concern that if workouts are institutionalized, their prospect or imminence could impart even greater instability in capital markets, by propelling a rush toward the exits by creditors and investors. The important search for a clear conceptual framework for workouts is sure to continue.16 Several aspects of such a framework are now emerging.

There should be an effort to map out a broad approach to conducting workouts. It will be important for borrowers and capital market participants to understand the system in which they are operating as investment decisions are made. That is not to say that simple rules can be prescribed and applied in individual cases. The history of the recent crisis shows that patterns of indebtedness and credit holdings as well as issues of timing will require solutions to be crafted on a case-by-case basis. But this could be within an approach that has been spelled out in broad terms. That would constitute a compromise between the U.S. and European views that emerged recently.

Creditors should be assured that workouts are not triggered whenever balance of payments problems arise. Broader use of flexible exchange rates should prevent some instances of monetary policy difficulties from turning into debt problems by permitting markets to absorb a greater degree of the burden of adjustment. When a country comes to the IMF, an effort should be made to find a combination of IMF access that is within normal limits and economic adjustment by that country that addresses the problem.

When debt sustainability is in question, however, that will prove difficult to accomplish. Typically, the balance of payments will show a prospective gap deriving from debt service (including interest, amortization of debt, and the execution of derivative contracts). In such a case, all major creditor classes should be expected to restructure their claims with comparable terms. The terms of the restructuring will need to depend on the severity of the country’s debt difficulties. In some cases, it will be clear that debt reduction is required to restore the sustainability of the balance of payments. In other cases, where it is unclear, debt can be rescheduled and, provided that the breathing space offered is sufficient, eventually serviced in full. The use of a cutoff date for rescheduling can effectively create a class of senior, post-cutoff-date claims that will permit countries to earn reentry into capital markets (akin to what happens in domestic bankruptcy situations). If further mismanagement or bad luck befalls the country, the rescheduled claims can be revisited and debt reduction on previously restructured claims can occur.

While that was not the design of the restructuring of commercial bank claims in Latin America in the 80s, that was the sequence of what evolved. That “lost decade” was notable for the slow pace of both restructuring and adjustment of economic policies. It would be an advance on that evolutionary process to develop a broad and accepted approach along more modern lines to be followed in the event of severe crisis.

Implementation of such an approach would leave the national authorities in the debtor country formally in control of the decision to suspend payments and to approach creditors for restructuring. While the creditors’ response would be voluntary, their options would be shaped by the context in which they were called upon to respond. The international community needs to structure its financial support in a way that facilitates private sector participation in workouts. First and foremost, the IMF needs to draw the line on the amount of finance that it will provide, so that a bargaining process doesn’t obstruct achieving private sector participation. In that sense, getting to workouts is integrally linked to restoring access limits. Second, the IMF can prevent its finance from boosting debt service payments, as occurred in its extended arrangement with Ukraine in 1998, by requiring that international reserve accumulation not fall below specified levels.17 Beyond that, the Paris Club, in turn, will condition official debt restructuring on the IMF program and a commitment from the debtor country to pursue comparability of treatment of significant creditors. And, if the debtor country, facing those constraints, limits debt service in the form of a standstill or partial standstill, the IMF’s lending into arrears further shapes the context for creditor response.

Combatting Systemic Threats

This paper has argued that the IMF should set limits on the finance it will provide to help a country in trouble. It should, at the same time, have a significant pool of finance to respond to system-threatening situations. For most of its history, the financing provided to individual countries has been to assist those countries cope with balance of payments problems in situations where the health and integrity of the international financial system has not been in question. Decisions regarding IMF support were made with the circumstances of an individual member as the paramount concern. Those decisions, of course, were always constrained by the usual need to ensure uniformity of treatment of members in the application of conditionality and the provision of finance so that undesirable precedents were not set.

There have been instances, however, when difficulties have become so widespread, or when the position of financial institutions in major markets has been weakened, and concern has arisen about whether crisis might envelop the international financial system. While the IMF has always seen its ordinary resources as the main reservoir for crisis response, the General Arrangements to Borrow (GAB) and now the New Arrangements to Borrow (NAB) are explicitly designed to supplement ordinary resources, if need be, when system-threatening situations arise. It was apparent in the recent crisis, however, that the potential financial requirements to stem a global crisis are very large and that circumstances could arise in which all of the resources presently available to the IMF could be inadequate to stem the spread of crisis. For a month or so following the collapse of the Russian ruble, there seemed to be a potential for the crisis to spread to a number of large developing countries and to have a significant and dangerous impact on major markets as well. During this event, the easing of interest rates and the IMF-supported program for Brazil led to a reversal of capital market sentiment and an easing of crisis concerns. Had the crisis intensified and spread, however, and had the international community wanted to act to boost reserves in many large countries at the same time, it might have required hundreds of billions of dollars to do so convincingly. In that case, the IMF’s resources might well have proven inadequate.

The combined special circumstances that posed the danger of a truly global crisis are no longer present. And as innovations in the international financial architecture are made, especially the wider use of flexible exchange rates, these circumstances are less likely to be repeated. The innovations should help prevent crisis and its spread. But, just as very few people anticipated the onset and severity of the recent crisis, it is difficult to know what changes in the globalizing economy, innovations in capital markets, or other factors might precipitate crisis in the future. The lesson from the fall of 1997 is that the international community should be equipped to react forcefully if some constellation of factors threatens the international financial system. Surely, as we saw in the fall, monetary policy management in the advanced countries is potent and may prove decisive in combating dangerous conditions. But it is also possible that there could be a recurrence of conditions in which capital markets judge that international reserves in a wide range of countries are inadequate or when there is a need for a large number of countries to be spending international reserves simultaneously. While the probability of such a need is low, the costs for the international community of not having a large pool of resources ready could be large.

The provision for allocations of Special Drawing Rights (SDRs) was created to respond to global illiquidity. A general allocation can be decided by an 85 percent vote of the Executive Board and, thus, in principle, be achieved quickly in crisis conditions. A drawback in the present arrangements for SDRs, however, is that SDRs are allocated in accordance with quotas, thus preventing greater financial support to the particular countries that might need it. One approach to giving the IMF a more flexible response capability would be to seek a change in the Articles of Agreement to permit alternative formulas to be used for SDR allocations. Such a change, however, is likely to be difficult because political action is needed by members to change the Articles of Agreement. Small countries, unlikely to be recipients in a crisis, may see the change as favoring large countries, while some large countries may fear the start of a slippery slope toward indiscipline if it gives the IMF Executive Board greater leeway to create liquidity as it sees fit.

An alternative approach would be a two-step procedure that combines a large general allocation of SDRs with a trust fund into which a select group of large countries would pool their portion of the allocation and which could be set aside for use to defend against a crisis. The SDR allocation would have to be large, so that the participating countries’ portion would be significant. A $300 billion allocation, for example, would provide about $142 billion for the G-7 and even more for a larger group such as the GAB ($162 billion), or the NAB participants ($205 billion). A trust fund established by the participants could be used as a last line of defense to defend the international financial system in times of dire threat. Many mechanisms could be devised for participation and governance, and important details would have to be worked out by participants. Principles could be established for the creation of such a fund, however, along the following lines: (1) include significant countries with a strong interest in the international financial system, (2) limit the number of participants to a size that will permit quick activation in times of crisis, and (3) devise a governance structure with decision-making power that permits activation to defend the financial system but prevents activation to help individual, favored countries.

While there are probably several ways to satisfy these principles, one proposal would be as follows. The participant group would include the NAB participants. The trust fund charter would establish a threshold for activation with a requirement that there be a significant and probable threat to the stability and integrity of the international financial system. The activation decision would require a supermajority vote. Participants could agree to consider activation upon a request from IMF management, as in the case of the GAB and NAB. A further step to fostering conservative management of the trust fund would be for participants to be the direct creditors to borrowing countries, rather than acquiring claims on the IMF as is the case with the GAB and NAB. If structured in this manner, the fund (and financing from the fund) would not be an IMF facility, but rather a backstop to the international financial system and operated in the interests of the international community.

Under this proposal, countries not participating in the trust fund would still be SDR recipients. Thus, the desirability of the approach would depend on whether the distribution of SDRs was itself desirable. At present, an SDR allocation could be beneficial. Reserve adequacy varies from place to place, but a boost in liquidity across the developing countries would be beneficial where there are lingering concerns about the balance between reserve adequacy and debt positions. Moreover, the use of SDRs by recipients might well have net beneficial effects, at least as of summer 1999, given that growth remains slow in many developing countries, trade growth is also slow, and there is slack in Europe and Japan, which would export to users of SDRs.

The operation of such a trust fund would have both similarities and differences with a lender of last resort. The fund would differ from the recently established contingent credit line. It would not lend to countries merely because they had been affected by contagion, even if they had a strong policy record and their difficulties were intense. The systemic threat criterion would prevent that. However, the fund would lend freely and at a penalty rate, as called for in the operation of lenders of last resort. Moreover, the management of the fund should retain constructive ambiguity about the circumstances in which it might be activated, in order to limit moral hazard on the part of governments and market participants. That ambiguity would include country eligibility, financing amounts, and the role of conditionality. Each would have to be determined on the basis of the circumstances of the crisis being confronted.


During the emerging market crisis, the IMF was asked to take on exceptional responsibilities under uncertain conditions. Some are harshly critical of the IMF’s performance, and, while no one is fully satisfied with its performance, others believe the IMF played a crucial role in helping the world through a difficult and dangerous episode. As future events unfold, the international community will be well served to have the IMF ready again to enter into the financial fray as needed. Part of maintaining the IMF’s readiness is to refocus its role so that its assigned tasks are the right tasks, and so that its effectiveness is not impaired by a lack of legitimacy and public acceptance. This paper has put forward a strategy for redrawing the limits on the use of IMF financial resources by shifting more of the burden of dealing with economic problems to the markets and private sector participants. This may prove difficult in the near future. If crisis reemerges before changes are made in the international financial architecture, the IMF will be asked to continue in its present firefighting mode. If Asian recovery continues and the emerging market crisis recedes, there may be little besides the memory of the crisis to propel further changes in the international financial architecture.


    CamdessusMichel1999“Sustaining Asia’s Recovery from Crisis,”speech at the South East Asian Central Banks Governors’ ConferenceSeoulMay20.

    • Search Google Scholar
    • Export Citation

    EichengreenBarry1994International Monetary Arrangements for the 21st Century (Washington: The Brookings Institution).

    EichengreenBarry1996Globalizing Capital: A History of the International Monetary System (Princeton, New Jersey: Princeton University Press).

    • Search Google Scholar
    • Export Citation

    EichengreenBarry1999Toward a New International Financial Architecture: A Practical Post-Asia Agenda (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    FischerStanley1999a“On the Need for an International Lender of Last Resort,”Journal of Economic PerspectivesVol. 13 (Fall) pp. 85104.

    • Search Google Scholar
    • Export Citation

    FischerStanley1999b“The Financial Crisis in Emerging Markets: Lessons for Eastern Europe and Asia,”speech at the East West InstituteNew YorkApril23.

    • Search Google Scholar
    • Export Citation

    FrankelJeffrey1999“No Single Currency Regime is Right for All Countries,”testimony before the Committee on Banking and Financial Services U.S. House of RepresentativesWashingtonMay21.

    • Search Google Scholar
    • Export Citation

    GhoshAtish and others1995“Does the Nominal Exchange Rate Regime Matter?”IMF Working Paper 95/121 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    Group of Twenty-Two1998Report of the Working Group on International Financial Crises (Washington: Group of Twenty-Two).

    HankeSteve1998“The Case for a Russian Currency Board System,”Cato Foreign Policy Briefing No. 49October.

    HenningRandall1999“The Exchange Stabilization Fund: Slush Money or War Chest?” (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation

    Institute of International Finance1999Involving the Private Sector in Crises in Emerging Markets (Washington: Institute of International Finance).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund1999World Economic Outlook April 1999 (Washington: IMF).

    KrugmanPaul1998“Will Asia Bounce Back?,”speech at Credit Suisse First BostonHong KongMarch.

    RubinRobert1999“Remarks in Reform of the International Financial Architecture to the School of Advanced International Studies,”U.S. Department of the Treasury press release RR-3093 Treasury NewsApril21.

    • Search Google Scholar
    • Export Citation

    SachsJeffrey1998“Proposals for Reform of the Global Financial Architecture” (unpublished paper; Washington).

    SachsJeffrey1999“Creditor Panics: Causes and Remedies,”Cato JournalVol 18 (Winter) pp. 37790.

    World Bank Group1999Global Development Finance 1999 Analysis and Summary TablesVol. 2 (Washington: World Bank Group).

*In preparing this paper, I have benefited from discussions with many people, both in and out of the official international financial community. In particular, my views have been shaped by my participation in discussions with members of the Council on Foreign Relations Task Force on International Financial Architecture, especially Barry Eichengreen and Morris Goldstein. The views expressed here, of course, are my own.
1See IMF (1999b).
2The IMF camp acknowledges a role for capital markets in the genesis of the crisis, but emphasizes national policy management. See Fischer (1999b).
3See Sachs (1998a) and the discussion in Krugman (1998).
5See IMF (1999a).
6Except for some devotees of currency boards, who argue that the institutional change will force the necessary disciplines (Hanke, 1998).
7Resisting that tide, Jeff Frankel has recently reasserted the traditional case-by-case argument (Frankel, 1999).
9IMF Managing Director Camdessus has called for greater discussion and consideration of this subject; see Camdessus (1999).
12See Sachs (1998a) for a discussion of the circumstances of failed pegs.
13See Eichengreen (1996), pp. 171–81, for a discussion of this episode.
14See IMF (1999b).
16See the paper by Eichengreen in this volume.
17See Eichengreen’s paper in this volume.

    Other Resources Citing This Publication