Current Developments in Monetary and Financial Law, Vol. 3

CHAPTER 11 Why Is There So Much Disagreement About the IMF and Reform of the International Financial Architecture?

International Monetary Fund
Published Date:
April 2005
  • ShareShare
Show Summary Details

The current debate over reform of the international financial architecture was stimulated largely by the rash of currency and financial crises in emerging markets that began with Mexico in 19941 and continued through Asia in 1997, Russia in 1998, Brazil in 1998–99, and Turkey and Argentina on several occasions, with the most dramatic being the crash of the Turkish lira in 2001 and the forced abandonment of Argentina’s currency board, also in 2001.

Such debates are not new, but in earlier manifestations they occurred under the rubric of reform of the international monetary system. The change in labeling is mostly semantic but it does capture what is one of the most important changes in the global landscape over recent decades—the growth of international financial interdependence. Some on the left have been quick to counterattack from the fall of the communist model of central planning by claiming that the recent crises are just another example of the inherent instability of capitalism and/or financial markets. Whatever the degree of limited truth such dictums may command, they offer little in the way of a productive agenda for reducing the frequency and severity of future crises. Much more insight is gained by viewing the most important causes of these crises as lying in the failure of national governments and international organizations to adjust sufficiently their thinking and policies along with their institutional structures in light of the realities of increased liberalization and financial interdependence. Accepting this framework still leaves considerable scope for controversy among mainstream economists. Just how well or badly did international financial markets perform? What were the policy mistakes that contributed to the crisis in Asia? They clearly weren’t the traditional ones of overly lax monetary and fiscal policies. Were the International Monetary Fund (IMF) loan packages too large or too small and did IMF policy advice help or hurt the situation? And what was the role of underlying problems with the international financial architecture? Should regulation of international capital flows be increased or reduced? Should the IMF be reformed to act more like a true international lender of last resort or, at the other extreme, should it be abolished? Does the IMF tend to be too easy or too tough on borrowers? Do exchange rates need to be made more flexible or more fixed? One can find leading experts who favor each of these propositions. This makes it very difficult for someone who does not follow these debates full time to get a reasonable idea of the point at issue.

This range of controversy is due in part to the complexity of many of the issues involved. On a number of these questions the only answer is that we do not really know. A number of experts, however, have discovered that acknowledgment of such uncertainties is not a good way to get public attention. Furthermore, many economists became highly committed to particular theories or models or views of the world and approached policy debates like true believers rather than serious scientists. Thus what so often goes on public display are extreme experts telling diametrically opposed stories rather than the two-handed economists who offer more balanced views.

This chapter offers a two-handed economist’s view of the state of debate on a number of aspects of reform of the IMF and the international financial architecture. A major theme is that when the extremes are eliminated there is much more consensus among experts than commonly meets the eye. It is the disagreements not the commonalities that attract attention. In the process I shall, of course, offer my own perspective on many of these issues, but within the context of distinguishing between personal views under uncertainty and generally accepted propositions. Throughout the chapter a number of the reasons for disagreements will be discussed. These include different objectives and interests, different mental models or theories, and different assumptions about political economy considerations.2

An Overview of the Debates3

What is really wrong with the international monetary and financial system and how should we fix it? A reader dipping into the literature on this subject quickly discovers that there is widespread agreement that the system needs reform. But there the consensus quickly ends. There is considerable disagreement about whether radical surgery or only more modest evolution is needed. Worse still—especially among those seeing the need for radical reforms—the directions proposed are often in sharp conflict. For example, many on the right think IMF lending needs to be radically scaled down to reduce moral hazard, while the conditions on loans should be made tougher. Many on the left think just the opposite. To them what is needed is more lending and less IMF austerity. While most academic experts from the left, right, and middle think that the rapid escalation of the areas of IMF conditionality over the past decade needs to be scaled back, such a move will face opposition both from mercantilist-oriented government officials in industrial countries, who want to use crises and IMF programs as levers to pry open business opportunities, and from representatives of labor groups and nongovernmental organizations who want to try to use IMF leverage to gain support for their particular objectives, be they labor reform, environmental policy, or human rights.

In this example it is easy to see that conflicting advocacy comes from conflicting objectives. This is, of course, one of the prime reasons for the wide variety of views on reform. Typically, even when actors have common objectives in some areas they will have conflicts of interest in others. Thus, for example, both developing country borrowers and industrial country lenders have a common interest in a smoothly functioning international financial system, but they will have quite different objectives with respect to the terms of loans or debt restructuring. And of course the shorter the time horizon adopted, the greater will tend to be the ratio of conflicting to common interests. While someone with strong particular interests at stake may make valuable arguments or proposals, it is wise to give such pronouncements special scrutiny.

Unfortunately, this cautionary note applies not only to those with special pecuniary interests, but also to those with strong ideological interests as well. Intellectuals from the left in the rich countries seldom have their material well-being substantially influenced by IMF policies, but they may have just as “committed” a view as the boldest industry lobbyist. While, on average, intellectuals on the right may have a higher correlation between personal gain and policy views than those on the left, neither side is fully immune to ideological blinders. This helps explain the curious phenomenon that on numerous occasions representatives from the far left and far right in the U.S. Congress have made common cause against increased U.S. funding for the IMF.

This situation of the two ends uniting against the middle is not explained by this being one of those rare cases where both ends agree on objectives. Their agreement rather is caused by a double disagreement—both normatively about values and positively about how the world works. Thus, just as multiplying one negative number by another yields a positive number, their double differences yield agreement. Where many on the left see the IMF as a capitalist lackey that imposes unduly harsh conditions on weak countries in order to improve business opportunities for the rich, many on the right see the IMF as an international bureaucracy run amuck that fosters socialism and creates moral hazard through pushing money at countries and winking when the promised policy reforms fail to be implemented.

Clearly both of these views cannot be correct as general rules. So how do such vastly different perceptions persist? I think a major part of the answer is that many individuals find strong ideological views to be quite comforting. Not only do they generate a feeling of solidarity with others of like view, but they greatly reduce the need for hard thought and careful analysis. Unless one has to personally bear the costs of the discovery that one has had a false view of reality—which has occurred for many, but far from all, international borrowers and lenders in the wake of the recent crises—then having a strong ideology or world view that one can hold onto makes life much easier. Of course the number who adopt ideological blunders so strongly that they are completely impervious to the facts is limited, but there are often many ways to interpret the same facts, and when one has strong a priori beliefs, then it takes only a few confirming examples to prove the case beyond doubt. Both the left and the right can point to real-world examples that fit with their beliefs. Each side can happily take the cases that fit as conforming to the general rule and assign those that conflict to the class of occasional exceptions. Neither side sees the need for undertaking the difficult task of seriously investigating relative frequencies, much less the conditions that make the outcomes go one way or the other.

A hallmark of debate about the IMF and international financial issues is that frequently those with the strongest views know the least about the subject. For example, it’s not unheard of for U.S. congressmen, or even some from the Washington-area policy think tanks, to fail to understand basic differences between the IMF and the World Bank. If a person’s ideology is powerful enough both organizations are virtually identical members of the community of international financial institutions.

Sadly, it must be admitted that academic experts are not entirely immune to the problems of personal interests and/or severe ideological blunders. The self-interest problems come primarily from the desire for visibility and making a name for oneself. The news media much prefer experts who offer strong, unqualified opinions as opposed to the economists’ typical (and intellectually honest) penchant for the more balanced view of “on the one hand and on the other.” While evenhandedness pays off in terms of reputation with most of an expert’s peers, it is the one-handed experts who make the papers (or television).

Ideological commitments also play a prominent role, although in the case of economists, the commitment is usually to some particular economic theory or school of thought rather than to conventional political ideologies. Such commitments to particular theories play a major role in the extent of criticism of IMF policy recommendations. Monetarists, Keynesians, new classical economists, supply-siders, and structuralists—to name only some of the most prominent schools—all have disagreements among each other about the most appropriate strategies for macroeconomic policy. To the eclectically inclined economist, the persistence of these different schools suggests that complicated issues are in dispute and a certain degree of humility in presenting analysis and policy prescriptions might be in order. But many of the adherents to the various schools have firmly convinced themselves that their school is fully right and everyone else is wrong.

To the uncommitted, the continual warring of these various economic tribes might suggest that modern economics is little more advanced than witch doctoring. That would be highly misleading. It is the controversies among economists that draw public attention. The broad areas of wide consensus tend to be much less visible outside of the profession.

This suggests that we should judge the quality of IMF policy advice not by the amount of criticism that it draws, but by how well it corresponds to middle-of-the-road professional best practice. On this criterion the IMF scores quite well. Strong Keynesians tend to think that the IMF is too monetarist and strong monetarists tend to think it’s too Keynesian. That is just where the IMF should be positioned.

It is generally agreed that the IMF made some important mistakes in diagnosis and forecasting during the Asian crisis and that, as a result, some of its initial policy recommendations were inappropriate. But the frequent charge of critics that the errors were due to the IMF’s rigid conformity to a narrow theoretical approach and the inability of Fund staff to discern that the low-inflation, fiscally prudent Asian countries were different from the stereotypical flaccid-budget, high-inflation Latin American countries is disingenuous.

While my observations have convinced me that there is a strong negative correlation between the vehemence of the criticisms of the IMF and knowledge about its operations, the correlation is not perfect. Some of the strong critics of the IMF do have solid professional credentials. Furthermore, while a substantial majority of the outside experts knowledgeable about IMF and reform issues reject many of the arguments of the far left and right, this doesn’t mean that most of the experts in the middle think that there are no problems with the IMF.

The Exchange Rate Issue

Financial liberalization, along with high levels of international financial interdependence, is hardly an entirely new phenomenon. Indeed, on some measures, interdependence was greater in some periods of the nineteenth century than it is today. To a substantial degree the recent rash of currency crises was due to a failure of governments and private borrowers and lenders to learn one of the major lessons from the collapse of the adjustably pegged exchange rate regime of the Bretton Woods international members’ monetary system in the early 1970s. Lest some think that I am displaying snobbery toward the intellectual capacities of those from developing countries, let me hasten to add that it was this same failure to learn (or to remember) that was a major cause of the currency crises in Western Europe in the early 1990s.

To be sure, all of these crises and collapses had a variety of contributing factors that influenced both their timing and severity, but they all had an important element in common. Commitments to some form of pegged or sticky exchange rates combined with various shocks and/or policy trends create seriously over- (or occasionally under-) valued exchange rates. This gave speculators marvelous opportunities to break the bank by betting against governments. Since the direction of any major changes in exchange rates was obvious, this gave speculators a prospect of large gains balanced by the risk of only small losses. To keep such situations from guaranteeing genuine one-way speculative gambles, governments of weak currencies had to increase interest rates, sometimes to astronomical levels. Worse still, while motivations might differ, the behavior of prudent businessmen and businesswomen seeking to hedge positions and protect themselves and their companies from the risk of a large depreciation generates the same types of exchange market pressures as the outright speculations of hedge funds and the “Gnomes of Zurich.”

Understandably, governments’ and central banks’ ability and/or willingness to maintain sky-high interest rates is limited. In such contests between the markets and governments, markets usually win. One of the best known examples is the almost one billion dollars that George Soros’ hedge fund took from British taxpayers in 1992 because of the government’s ill-conceived efforts to maintain an overvalued pound.

What this illustrates is a basic principle of international monetary theory, which comes under various names, perhaps the most common being the unholy trinity. This principle argues that in a world of high capital mobility it is not possible to have simultaneously pegged exchange rates, freedom from controls, and independent monetary policy. Something must give. It does not take a rocket scientist to understand this principle, so why is it apparently so hard for many policymakers to learn? I believe that a major part of the answer is that the unholy trinity theorem is true only in the medium or long term.

In the short term, luck and international financing can allow all three objectives to be pursued simultaneously. Until the pegged or sticky exchange rate starts to become seriously under- or overvalued, there is no problem. In some cases this happy situation can last for a number of years—for several of the Western European countries, 5 or more, and for Thailand, 14.

A second important part of the answer is that the modern theory of optimum currency areas shows that the traditional economists’ debate over fixed versus flexible exchange rates was, to a substantial extent, displaced. Despite some vocal enthusiasts for floating rates for everyone, or, more recently, advocates of currency boards or dollarization for everyone, economic theory shows that there is no one exchange rate regime that is best for all countries. Every regime has both costs and benefits and their ratios will vary systematically across countries, depending on a number of factors, such as their size and openness, the flexibility of their domestic labor markets, and the patterns of shocks that they face. Thus, for example, small, highly open economies, such as Estonia (population two million), are likely to secure net benefits from a fixed exchange rate currency board, while the United States, with its minimally managed float, made quite different choices; yet they each likely made the right choice.

A key problem, however, is that most economies approximate the ideals for neither genuinely fixed nor fully flexible exchange rates. On Optimum Currency Area grounds they should adopt some form of intermediate exchange rate regime. The difficulty is that, as was discussed above, it is now clear that the standard intermediate regime adopted at Bretton Woods—the narrow band adjustable peg—is highly unstable in the face of substantial capital mobility. Almost all economists agree with this unstable middle hypothesis. What we do not know is how far away from the middle one needs to move. Advocates of the two corners hypothesis argue that it is necessary to go all the way to one of the extremes—genuinely fixed or highly flexible rates. Others, like John Williamson, are more optimistic that fairly flexible crawling bands are workable. Examples such as Chile, Hungary, and Poland are consistent with this optimism, while those of Indonesia, Korea, and Mexico show that crawling bands can be operated more like pegged rather than flexible rates. The truth is that at present we do not yet fully understand the conditions that have made some crawling bands quite stable while others have been crisis prone. We can be fairly confident, however, that the answers will involve political as well as economic considerations.

There can be substantial political advantages to governments in the short run from ignoring the long-run constraints imposed by the unholy trinity. Political pressures can easily help induce excessive optimism by officials that any current disequilibrium is temporary, so that politically costly adjustments can be delayed and perhaps avoided altogether. The economic logic of running various forms of pegged rates, such as adjustable and crawling pegs and bands, in ways that avoid currency crises is that they need to be adjusted sufficiently promptly that the buildup of substantial disequilibrium is avoided. However, the political logic of governments facing short-run pressures makes this economic logic extremely difficult to implement.

The Asian Crisis Debate

The role of financial sector factors helps explain much of the disagreement in diagnosis of the causes of the Asian crisis and its spread. The widespread habit of not including such considerations in the list of fundamentals relevant for currency analysis helps explain why many economists—not just government officials in the crisis countries—initially argued that a large part of the speculative outflows during the Asian crisis were unjustified and that countries such as Indonesia, Korea, and Malaysia were innocent victims of what Tom Friedman has labeled “the electronic herd.” After all, these countries had low inflation and strong fiscal positions, so with their fundamentals strong speculative attacks on these countries must have been unjustified. The appeal of such an argument is understandable, but the analysis is quite wrong because, as we shall see below, it fails to include several key fundamentals not directly related to domestic economic policy—exchange rate disequilibrium, financial sector problems, and political instability. Indeed, one of the key factors behind the surge of capital inflows into Asia and other emerging market favorites such as Mexico was this oversimplified view of the fundamentals. A major lesson of the recent currency and financial crisis is that such an oversimplified view is highly dangerous.

Views on strategies for reform are of course heavily dependent on analysis of the problem. Thus, the debates over the causes of the Asian crisis and evaluations of the IMF’s strategies during the crisis have had substantial influence on the debates over reform. Miles Kahler has usefully suggested that we can gain insight into these debates by distinguishing two major schools: the fundamentalists who view international financial markets as reasonably to highly efficient, and hence look to bad policies as the cause of the crisis; and the panic-stricken, “who were convinced that many crises were self-fulfilling, that is, expectations about economies that were ex ante justified were validated ex post by the outcome they provoke.”4 The latter analysis suggests emphasis on controls on capital flows and large IMF lending packages. Most economists and academic experts in finance are attracted to efficient market theory and hence their initial inclinations are to favor the fundamentalists’ view of the crisis, with the frequent failure of markets to give strong early warning signals being interpreted as stemming from beliefs that investors and borrowers would be bailed out in the event of crisis.5

The view that financial markets are shortsighted, engage in self-fed frenzies, and commonly overreact is widespread outside of academia. And in the case of the Asian crisis, practioners such as George Soros have been joined by prominent economists such as Jeffrey Sachs and Joseph Stiglitz. Indeed, recent theoretical analysis has shown that under some conditions herding behavior in financial markets can be rational, thus making it easier for economists to accept that such behavior may exist.

The “speculators are at fault” view gets a strong boost from a superficial look at the Asian crisis countries. For most of the crisis countries the traditional domestic macroeconomic fundamentals were strong. Budgets were in surplus, not deficit, inflation was low, and growth was high. Indeed a number of these countries were widely believed to possess miracle economies. True, there were signs that the miracles were wearing out, but nothing to suggest that disaster was imminent. As we broaden the range of fundamentals included on our radar screen, however, a quite different picture emerges. Most of the crisis countries had huge capital inflows. While this was a sign that they were doing things right, it was doubtful that inflows of such magnitudes could be sustainable. In some cases, the inflows were combined with large current account deficits that made these countries vulnerable not only to a reversal of flows, but even to a drop in the rate of inflow. On top of this, at the epicenter (Thailand as well as the Philippines) old-fashioned currency overvaluation had resulted from the maintenance of a virtually fixed peg in the face of changing economic conditions. All of these signs were clearly visible, yet they were ignored or discounted by many analysts and investment managers. What was not plain in the published statistics was the development of massive problems in the financial sectors. Official data on nonperforming loans are notoriously suspect. Problems usually do not show up there until it is too late.

Due in no small part to financial liberalization unaccompanied by adequate supervision and the continuance of a mentality that the government will stand ready to orchestrate bailouts, there was an explosion of credit to borrowers who were often making ill-conceived investments in real estate and industrial sectors suffering from overcapacity. Furthermore, much of the lending was financed by unhedged short-term borrowing from abroad. Thus, financial systems were a time bomb waiting to explode, but there was relatively little evidence of this in the published statistics usually followed by analysts. Indeed, even IMF staff—while beginning to raise some notes of caution—had no idea of the extent of these problems, nor did many of the government officials in the countries in question. One of the reasons the announcements of IMF programs appeared to have so little effect in calming the markets during the crisis was the continuing flow of revelations that the financial situation was worse than had been thought. Estimates of the magnitudes of short-term foreign debt of financial institutions and corporations were raised numerous times, while effective reserve levels were found to be much lower than the published figures.

The interaction of currency and financial crises explains why the real effects of the Asian crisis were so devastating while the crises in Western Europe in the early 1990s were not. The interaction of financial and currency issues also helps explain some of the controversy surrounding the role of speculation and currency regimes in the crisis. Estimating equilibrium exchange rates is far from an exact science, but there is widespread agreement among experts that before the crisis, the Thai baht and Philippine peso were noticeably overvalued. The same cannot be said for the South Korean won and Indonesian rupiah, however. These were at most only mildly overvalued and some estimates even put them as slightly undervalued. Surely then these countries must have been the innocent victims of currency speculators.

Such arguments have attracted considerable support, but they overlook a crucial factor. These currencies were roughly correctly valued based on an assumption of continuing capital inflows, which in turn assumed fairly sound financial systems. When massive problems were “discovered” in the financial positions of many banks and corporations, capital flows would naturally be affected. Foreign investors quite rationally reduced investment inflows and often attempted to reduce positions while borrowers scrambled to cover their unhedged foreign debts. Without either panic or overt speculation, the change in the perception of financial fundamentals would lead to a substantial shift in capital flows. This would leave initially correctly valued exchange rates substantially overvalued. In Indonesia this deterioration in perceptions of financial fundamentals was compounded by increasing political instability—another important fundamental.

On this broader analysis, the Asian contagion countries were unfortunate, but not entirely innocent, victims. This does not mean, however, that all of the severe economic dislocation that was suffered in these countries was inevitable. In his description of the panic-stricken view discussed above, Kahler equates self-fulfilling speculation with unjustified speculation. This is a common practice, but can lead to confusion. While self-fulfilling speculation could be unjustified, it need not be.

Economists’ second generation of crisis models makes the important point that to view fundamentals as coming only in two flavors—good and bad—is misleading.6 There is also an important intermediate category where the fundamentals are neither so bad that a crisis is inevitable nor so strong that a crisis could come only from unjustified speculation. In this intermediate zone, countries are in a situation of vulnerability, but with good luck and an absence of bad shocks, crisis is not inevitable. This is the situation where self-fulfilling speculation is most likely to occur. And while such runs may result from blind panic, they may also reflect rational behavior under limited information.

Dealing with Capital Account Crises

The perspective from second-generation crisis models suggests that we must move beyond the fundamentalist versus panic-stricken dichotomy and provides a rationale for the IMF to help prevent or mitigate bank runs and other forms of liquidity crises or, in other words, to operate as a type of international lender of last resort. Indeed this is the view adopted by the majority report of the Meltzer Commission.7

Pure liquidity crises are rare, however. Generally, genuine fears of insolvency lie behind financial runs and it is no easy matter, even theoretically, much less operationally, to always clearly distinguish between insolvency and pure liquidity.

Fears of unjustified contagion from one country to another have been a major motivator for IMF loan programs with several countries, as well as the Contingent Credit Line (CCL), which was designed to prevent contagion to innocent victims. There is considerable disagreement among experts about the extent to which there have been innocent victims of currency crises. There is much we still do not understand about the dynamics of financial market behavior and the episodes of international contagion that we have seen over the past decade. My suspicion, however, is that official views, including those at the IMF—at least under its previous management—have tended to give too much weight to the panic-stricken view of contagion due to a combination of a misreading of the Asian contagion and a concern with being blamed if the panic-stricken view is correct. Under uncertainty we must compare the risks of action and inaction incorporating assumptions about which view is correct. From this perspective the costs of creating the CCL, even if the rationale underlying it is wrong, are relatively low. On the other hand, the costs of offering programs to Argentina and Brazil, where there was a substantial probability of failure, were much higher—not so much because of risks that the IMF loans would not be repaid, but rather because of damage to the credibility of the IMF’s seal of approval and hence to its ability to calm markets.

Much of the confusion and controversy about contagion comes from the conflation of different concepts of contagion. While the nature of each of these concepts is beyond the scope of this paper, it should be noted that in response to major crises there are often broad and fairly indiscriminate responses in financial markets. Such indiscriminate responses, however, are usually both fairly short-lived and mild. They typically have little relevance for policy. Strong contagion reflecting serious speculative attacks is much more focused. This characterization holds for the Mexican, Asian, Brazilian, and Argentine crises. The Russian crisis generated a third type of response, where strong speculative attacks were quite limited, but there was a widespread substantial increase in borrowing costs or lack of access to international financial markets for a matter of months rather than days. While many have talked of the rapid spread of the contagion from Thailand to the rest of Asia, it was not panicked responses to Thailand that caused the Korean crisis. The initial fallout from Thailand on Korea was relatively mild. It was months later that Korea was hit hard. Thus, this could not possibly be the result just of panic-stricken contagion.

This said, there is considerable evidence that markets do sometimes overreact.8 One can argue that Indonesia and Korea were not innocent victims; their punishments were worse than their crimes. Self-fulfilling market behavior did not precipitate these crises, but it did make them much worse. A part of the excessive costs of these crises was due to political instability and/or inept policy responses by national governments, but more could have been done by the IMF to reduce the substantial overdepreciation of currencies that occurred.

An important part of the problem here is that the IMF’s institutions and policies were designed in a period of low capital mobility to deal with slowly evolving current account crises. These procedures are not well-suited to deal with modern capital account crises. These move much faster and even moderate disequilibria can generate large-sized capital flows.

The IMF is aware of this problem and has been moving in the right direction by establishing the Supplemental Reserve Facility.9 This can provide larger and more front-loaded loan packages bearing higher rates of interest. Such incremental changes have not gone far enough, however. A fundamental rethinking of the role of the IMF programs in capital account crises is needed. The traditional IMF strategy of enforcing policy conditionality through doling out money in tranches over time made great sense in a world of low capital mobility. And it continues to have a role to play, but it needs to be supplemented by the possibility of lender of last resort (LOLR)-type lending that can provide large quantities of funds quickly based on ex ante judgments and preconditions. If IMF actions are to arrest international bank runs, there is little time to negotiate strong conditionality programs with national governments, much less to provide assurances that such programs enjoy broader national ownership.

Recent IMF programs have a poor record of establishing confidence and halting capital outflows. In Asia the IMF was correct that financial sector issues were at the heart of the crisis and needed to be corrected. Unlike many areas of expansion of IMF conditionality, fiscal financial sector policy was as much a core issue as exchange rate and monetary policy. But what was needed was structural reform, which could not be implemented quickly. The pressures to conclude a traditional conditionality program quickly helped contribute to several miscues, and some have argued that the net result did more to harm than to help market confidence. Far better would have been initial LOLR lending in large amounts for short periods at penalty interest rates that could provide a breathing space while a replacement program with well-thought-out conditionality and national ownership could be developed.

The construction and efficient implementation of such a new type of IMF facility would be no easy task, but I am convinced that such a facility is essential for the IMF to operate at maximum effectiveness.10 This is only one prong of a two-part strategy to deal with capital account crises, however. Bank holidays and payments standstills are an alternative way to deal with bank runs and capital account crises.11 In some cases, the magnitude of official funds needed to calm markets would likely be larger than political constraints would allow the IMF to provide. While large-scale funding of liquidity crises may be efficient, from an equity perspective such funding often serves to finance low-cost exit strategies for well connected private sector actors. There is much merit to the argument that in recent crises what was needed was more “bailing in” of the private sector. Again, however, this is a complicated issue. When one is dealing with a wide spectrum of investors and borrowers, rather than just a few banks, the logistics of such private sector involvement (PSI) can be difficult. Furthermore, fears of standstills can themselves perpetuate outflows. Given such considerations it seems likely that better procedures for both LOLR and PSI activities are needed.

Furthermore, many crises are not pure liquidity ones. Insolvencies are also often at issue. But the suggestion that the IMF should lend only in the case of a pure liquidity crisis is neither efficient nor politically realistic. In many liquidity crises, it is not known initially whether solvency is at issue or not, so risk-averse rational actors run for the exits, thus typically compounding the amount and costs of any eventual insolvency. For example, during the Asian crisis, the initial overdepreciation of national currencies greatly magnified the amount of private sector insolvency.

When we come to sovereign debt the issue is even more complicated. The best way to deal with such issues of debt restructuring is far from clear, but it is obvious that there are important collective action problems that make it unlikely that the private sector will be able to work out efficient solutions unaided. Thus the IMF’s new initiative to deal with such crises is highly welcome. It will not be an easy process, but at least it is now on the official agenda.

The Political Dimension

On this point, this chapter has addressed primarily technical and ideological issues. One more crucial area must be examined before the task is complete. This involves the political dimension. Another major source of disagreement over proposals for reform is differences in political assumptions. Often these are left implicit and hence lead to further confusion.

Many proposals by economists are utopian in the sense that they focus on what would be ideal if there were no political constraints. Others make more “realistic” proposals by attempting to take political constraints into account, but there is seldom a consensus on the relevant constraints and this might lead to a range of proposals from experts with the same view of the economic issues. This could easily generate confusion. The solution, of course, is to be explicit about both one’s political and economic assumptions so that each can be addressed separately.

Quite understandably, the IMF has a tradition of downplaying the role of politics. It is staffed primarily by economists and has built its reputation as a technocratic institution. But it is clear that its political facade is just that. While often exaggerated, there can be little question about the strong influence that the U.S. government frequently has on IMF policies. Less well known, but likely as important, is the political leverage enjoyed by many borrowing governments. The self-interest incentives of IMF management and staff make it difficult for the IMF to say no, if such a response runs a substantial risk of being seen as leading to crisis. This and other considerations have often given borrowing countries a great deal of leverage. Instead of the harsh taskmaster perceived by the left, the IMF often behaves more like the perception of the right—that it keeps on negotiating and funding programs that both it and the borrowing governments know will not be implemented. The result of such political manipulation of the IMF by both the rich and the poor has resulted in serious damage to the credibility of IMF programs.12

Another powerful consideration leading to differing analyses of international financial issues is the human desire to avoid blame. Of course national policy officials in crisis countries will be tempted to point the finger at irrational panic and greedy speculators rather than any lapses in their own policy leadership. Likewise international investors have tended to emphasize a lack of transparency as the cause of their failure to detect the early warning signals of impending crisis. (Most independent experts agree that lack of transparency is a contributor, but usually it is judged to be a minor factor compared with the failures to adequately analyze the information that was available.)13

It is essential to the smooth functioning of the international financial system that the IMF’s credibility be restored. This will undoubtedly require the IMF to say no more frequently and the governments of the industrial countries must allow the IMF to do this. This is simple in theory, but will be difficult to achieve politically.

More difficult theoretically is how to determine when a potential IMF program is a good bet. Recent history suggests one type of situation in which the answer is relatively easy. IMF programs have been notably unsuccessful in saving regimes of pegged exchange rates in the face of strong market pressure. Argentina, Brazil, Russia, and Turkey are all prime examples. Thus only under rare conditions should the IMF say yes to programs attempting to save sticky peg regimes.

Requests for IMF programs have also often come after pegs have been abandoned, however. Argentina, Brazil, and Turkey at present, and Indonesia, Korea, and Thailand during the Asian crisis are examples in this regard. Here the basis for saying yes or no is much less clear. Not only must the IMF judge that the proposed program will do the job, but also that enough of the program will be implemented to make it effective. This judgment in turn requires substantive political economy analysis—something for which the IMF is ill prepared. Its commitment to its technical economics persona has made the IMF understandably loath to admit the crucial role of politics. But as Kahler has observed, few are any longer convinced by the fiction that the IMF is apolitical. What is needed is to make the IMF less political in the sense of strengthening its independence from political manipulation while making it more political in the sense of developing a better capacity to make judgments about the likely political feasibility of policy implementation. These two objectives will each be difficult to implement, but it is important to understand that they do not conflict.

Concluding Remarks: IMF Learning and the Brazilian Program

This chapter has sought to establish that there are a number of reasons why there has been so much recent controversy about the IMF and reform of the international financial architecture. Much of the most vocal criticism comes from those with strong political orientations who believe that ideology is a substitute for learning about what the IMF actually does and from snake-oil-selling experts who offer highly oversimplified views of the world in which their pet policies will work wonders.

When we clear away this thunder and lightning we find a good deal more consensus among the views of nonideological experts and much less argument that the IMF has done a horrible job and should be abolished. However, this is not to say that the IMF has been anywhere close to perfect. It has been far too lax in enforcing the conditionality of its programs and has allowed many countries to come back for new programs time and time again in spite of not living up to their obligations under the old programs. Furthermore, the areas of conditionality were expanded well beyond the range of the IMF’s core competencies. Perhaps most dramatically, the IMF (often with the nudging of the major powers) has been far too willing to help finance the efforts of national governments to maintain various forms of pegged exchange rates in the face of substantial disequilibrium.

The IMF has also made a number of mistakes in its policy advice. But what economist has not? The IMF undoubtedly has some arrogant and doctrinaire officials who fit the picture of the ugly IMFer, but what is remarkable about the IMF is how undoctrinaire most of its officials are. It is hard to think of another international bureaucracy, or a national one for that matter, that has been so willing to learn from its mistakes. The IMF is certainly building up its expertise on financial sector issues and there are some signs of greater willingness to recognize the political economy aspects of its operations. Likewise, during Horst Köhler’s term as Managing Director there has been considerable recognition of the need to restore credibility to IMF programs, to scale back the scope of conditionality, not to finance disequilibrium pegged exchange rates, and to address private sector coordination issues in international finance. It is too soon yet to tell how far these shifts will go, but few experts would argue that they are in the wrong direction.

The major possible exception comes from those who argue that the IMF’s new get-tough strategy reflects an excessive concern about moral hazard by the U.S. Treasury.14 There is little question in my mind that moral hazard is a serious issue, but a number of economists on the right have overemphasized its importance with respect to the operations of the IMF. Arguments for tightening up IMF programs need not rely on excessive concerns with moral hazard. The need to restore credibility to IMF programs so they can play a stronger catalytic role in stimulating stabilizing private capital flows is reason enough. Some argue that Argentina is being made a scapegoat for the new IMF image. It is hard for an outsider to evaluate that claim, but there have certainly been some good reasons for the IMF to refuse to accept the policy packages that have been offered by Argentina so far.

A different argument is that with its announcement of a $30 billion loan package for Brazil in August 2002, the IMF had ended its brief flirtation with its new get-tough policy. To some commentators this has been a source of alarm—to others a desirable rejection of what is seen as right-wing ideology coming from the U.S. Treasury.

I would argue, however, that the Brazilian loan package need not reflect a reversion of policy back to the big bailout days. While the press and critics of the IMF have begun to label all IMF programs as bailouts, this is highly misleading. The worst types of bailouts allow some to get their money back safely while others suffer. This occurs frequently when governments are fighting to maintain pegged exchange rates, where international loans can help some people pull their money out at favorable rates before depreciation occurs. This certainly occurred with IMF loans to Russia and this is far from the only case. The Brazilian loan is quite different, however.

Some commentators quickly concluded that the IMF program in Brazil had failed. This was because the favorable effects on the currency and financial markets from the announcement lasted only a few days. What these commentators missed was that the circumstances in Brazil were quite different from those in Argentina or in earlier high-profile failed IMF programs in Brazil, Russia, and Turkey. In all of these cases, the IMF programs were motivated in substantial part by the desire of national governments to maintain various forms of pegged exchange rates. However, Brazil has learned the dangers of pegged rates and floated the real. There is no longer a fixed target for speculators to shoot at, nor can insiders pull out money at overly favorable rates.

Furthermore, it was difficult to find serious fault with Brazil’s current economic policies. Usually with IMF programs major policy changes are required. The typical cause of failure of programs is that these changes do not come close to being fully implemented. With this particular program, it was only the maintenance of current policies that was required.

So what was the problem? The previous description makes it sound as if Brazil’s current economic team was largely the innocent victim of capricious financial markets. The first part of this statement is correct, but the second is not. Irrational markets are not the only reason that victims can be innocent. Neighboring Uruguay followed good economic policies but it was hurt badly by the crisis in Argentina. Uruguay was largely innocent but the contagion that hit it was due to strong economic and financial interdependence, not irrational panic by investors.

In Brazil the problem was twofold. Past fiscal excesses have left Brazil with a large public debt. At normal interest rates, the resulting burden of debt service was substantial but manageable. At sky-high interest rates, it was not. Consequently Brazil was vulnerable to crises of confidence. The current crisis of confidence is based on sensible evaluation, not hysteria. The cause lay not in current economic policies, but in the polls. Elections were to be held in October 2002 and the two leading candidates were both from parties of the left. Thus, the markets had quite plausible fears that current sound economic policies would be abandoned after the election. A return to hyperinflation was not likely, but at current debt levels a much less dramatic loosening of policy would be sufficient to force default.

Note that what was at work was not just some capital market conspiracy against parties of the left. There are many cases of left-of-center governments following quite sensible macroeconomic policies. Chile provides a current example. The problem in Brazil was that the leading candidates had been unwilling to make clear-cut commitments to maintaining current prudent macroeconomic policies.

One of the major reasons for the short-lived effects of the announcement of the new IMF program was the continued ambiguous responses of candidates Mr. de Silva and Mr. Gomes. Initially, they were widely reported as backing the program, but as time passed their failure to make strong commitments understandably undermined confidence again.

What the IMF did was offer Brazil a lifeline. However, this could help only if the presidential candidates grabbed it. Quite wisely, the IMF back-loaded its commitments so that most of the money would not be provided until after the election, and then only if policies remained sensible in terms of long-run viability, not short-term political popularity. The preelection part of the program offered the candidates some time to make their decisions. The downward pressure on the Brazilian real was not coming primarily from large speculative outflows. Rather it came from a current account deficit and the maturing of debt obligations that in current circumstances, risk-averse lenders understandably did not want to roll over. In the absence of strong speculative outflows from Brazilian citizens, the limited financing from the IMF would be enough to substantially reduce the odds of a major balance of payments crisis before the election.

Yet there were other, greater problems. Restoring confidence was up to Mr. de Silva and Mr. Gomes. They could not have it both ways. Of course they wanted to sound favorable toward the IMF programs to reassure the markets while stopping short of full endorsement in order to keep their options open and their anti-IMF supporters on board. But there was never much chance that such a strategy would work. Markets are sometimes gullible, but not that gullible. What the IMF did was put the ball in the candidates’ court with a strong set of inducements to make the right decisions. In doing this the IMF had unquestionably taken a risk, but a much more responsible one than with Argentina and Russia in the past. Thus, despite some superficial similarities this new program did not reflect a retreat from the IMF’s much needed new policy of toughening standards for its programs. Brazil was following sound policies—this time.

Therefore, recent developments give us reason for optimism that the IMF is continuing to revise its policies in sensible directions. That current IMF policies and the structure of the international financial system are not as flawed as many critics claim does not provide a sound reason for thinking everything is fine, however. There is still a long way to go to implement the reform that most middle-of-the-road experts believe is needed. The political and bureaucratic obstacles to sensible economic reform are often great, so slow progress is probably the best we can hope for. But it certainly beats movement in the wrong direction.


There were also major crises in Europe in the early 1990s, but the debates they stimulated were narrower.

For earlier discussions of sources of disagreement on international monetary issues, see Benjamin J. Cohen, Organizing the World’s Money: The Political Economy of International Monetary Relations (1977); Richard N. Cooper, “Prolegomena to the Choice of an International Monetary System,” in World Politics and International Economics (C. Fred Bergsten and Lawrence B. Krause, eds., 1975); and Thomas D. Willett, “Functioning of the Current International Financial System: Strengths, Weaknesses, and Criteria for Evaluation,” in International Money and Credit: The Policy Roles (George M. von Furstenberg, ed., 1983).

For an excellent, more detailed overview of the debates over reform of the international financial architecture, see Barry Eichengreen, Toward a New International Financial Architecture (1999) and Peter B. Kenen, The International Financial Architecture: What’s New: What’s Missing? (2001). More specifically on the IMF debates see Thomas Willett “Understanding the IMF Debate,” V-4 The Independent Review 593, and John Williamson, “The Role of the IMF: A Guide to the Reports,” Institute for International Economics Policy Brief 00–5, 2000.

Miles Kahler, “The New International Financial Architecture and Its Limits,” in The Asian Financial Crisis and the Structure of Global Finance (Gregory W. Noble and John Ravenhill, eds., 2000), at 235–60.

The degree to which moral hazard contributed to the recent crisis has become the subject of considerable dispute. There is considerable agreement that the prospects for international bailouts were a major factor in lending to Russia, but nationally generated moral hazard appears to have been much more important for Asia. For more discussion on the range of views on capital flows and the behavior of international financial markets, see Thomas Willett, “International Financial Markets as Sources of Crisis or Discipline: The Too Much, Too Late Hypothesis,” Princeton Essays in International Finance (May 2000).

For a useful nontechnical overview of this literature, see Eichengreen, supra note 3.

For analysis of this and other reports on international financial reform see Willett, supra note 3.

See Willett, supra note 5.

For a useful discussion by IMF staff of capital account crises, see Atish Ghosh and others, IMF-Supported Programs in Capital Account Crises, IMF Occasional Paper 210 (2002).

See Thomas Willett, “Restructuring IMF Facilities to Separate Lender of Last Resort and Conditionality Programs: The Meltzer Commission Recommendations as Complements rather than Substitutes,” Claremont Colleges Working Paper (August 2001).

See Ken Rogoff, “International Institutions for Reducing Global Financial Instability,” 13 Journal of Economic Perspectives, at 21, and Williamson, supra note 3.

For discussion and references to the literature on political influences on the IMF, see Thomas Willett, “Towards a Broader Public Choice Analysis of the International Monetary Fund,” in Organizing the World’s Money, David Andrews, Randall Henning, and Louis Pauly, eds. (2002).

See, e.g., Miles Kahler, “The New International Financial Architecture and Its Limits,” in Gregory W. Noble and John Ravenhill, eds., The Asian Financial Crisis and the Structure of Global Finance (Cambridge: Cambridge University Press, 2000), at 235–60.

See, e.g., Ricardo Hausmann, “Baffled By Brazil,” Financial Times, August 14, 2002, at 11.

    Other Resources Citing This Publication