Detecting and Preventing Financial Crises-Recent IMF Approaches, An Address by IMF First Deputy Managing Director Anne O. Krueger

June 12, 2003

Detecting and Preventing Financial Crises—Recent IMF Approaches
Anne O. Krueger
First Deputy Managing Director
International Monetary Fund
Bretton Woods Committee Annual Meeting
U.S. State Department
June 12, 2003


I'm delighted to be with you today. It is a great honour to address this distinguished gathering—even if it is as a last-minute substitute. Horst Köhler is very disappointed that he cannot, after all, be with you today. He was keen to share with you some of our thinking about the challenges that lie ahead.

These are uncertain times for the global economy and, therefore, for many of the emerging market economies that rely on the Fund for help in developing sustainable economic policies. At the same time, we live in an age of great and rapid change. The problems that confront the international financial system are, in many respects, radically different from those just a decade or so ago.

As you well know, the Fund isn't there to do governments' job for them. But we do have an important role to play in helping them. We, too, have had to respond to rapidly changing circumstances. We have done so by what management consultants might call re-tooling, so that we can be effective in the twenty-first century.


Our aims have not changed, of course: merely, in some respects, the ways in which we can most effectively achieve those aims.

Our primary mission is sometimes overlooked: but it remains, as it always has been, crisis prevention. Only in those occasional cases where we cannot prevent crises, does our role become one of crisis resolution.

I want first to say something a little more about the IMF and what we nowadays see as our role. I will then turn to the changing nature of the crises which we seek to prevent—and about how we carry out that function. Finally, I will make some observations about crisis resolution.

The IMF and its role

The IMF is, and always has been, a membership organization. Its mandate is "to promote international monetary co-operation...; to facilitate the expansion and balanced growth of international trade...; to assist in the establishment of a multilateral system of payments...; by making resources of the Fund temporarily available to [members] under adequate safeguards, thus providing them with opportunity to correct maladjustments in their balance of payments; and to shorten the duration and lessen the degree of disequilibrium in the international balances of payments of its members."

In carrying out its mission (in accordance with the articles that govern its activities), the Fund has adopted a number of principles. These include, first, the equal treatment of members, as reflected in policy papers outlining responses to various situations. Then there is the annual surveillance of all members, consulting with them about and not dictating to them their macroeconomic policies (at least on fiscal, banking, monetary and exchange rate issues) and other variables that determine the sustainability of their policy stance. We have overseen the adoption of standards and codes to provide for the greater transparency of members' policies, thus contributing to good policy and economic stability. There are, of course, many more. But these are at the core of our work.

The changing nature of crises

Until the l970s, most crises were "balance of payments" crises. These usually came about under fixed exchange rate systems in which either macroeconomic policies had been inconsistent with a fixed exchange rate peg (as, for example, when a country was experiencing a significantly higher rate of inflation than the rest of the world); or when a significant deterioration in the terms of trade occurred against the backdrop of a highly restrictive trade regime with few foreign exchange reserves.

Balance of payments crises, or current account crises as they then were, were normally preceded by at least a short period of several months during which import demand was increasing sharply (in anticipation of devaluation) and exports were shrinking (as exporters withheld stocks to benefit from the anticipated devaluation). Since crises had typically been preceded by overly- expansionary domestic policies, the IMF responded with financial support for a program that normally entailed an exchange-rate change and the tightening of fiscal and monetary policy, along with other reforms that were deemed integral to that tightening. Sometimes a removal, or at least a relaxation, of quantitative restrictions on imports was also a part of the package.

Indeed, after the collapse of the Bretton Woods system in the early 1970s, floating exchange rates became the norm for industrial countries; and deregulation, of the capital account and in many areas of the domestic economy increasingly came to be the stated policy goal, if not always implemented in full.

Long after the end of fixed exchange rates among the major trading countries, most developing countries continued to peg their currencies to one of the major currencies, often the dollar. But at the same time, the increasing role of private capital flows, and the increasing stock of outstanding debt led to crises (especially after the second oil price increase in the early l980s) where the instigating factor was an inability to service debt, rather than an inability to finance current account transactions. Even then, however, about half of all outstanding debt was to official creditors, and IMF programs were not significantly dissimilar from those earlier undertaken—except that there was coordination with the rollover of debt (and even new money) from private sector banks, the major lenders.

By the l990s, however, private capital flows to emerging markets were growing very rapidly, and sovereign debt to the private sector greatly exceeded that to the official sector. Many of the headline crises of the l990s were thus "capital account" crises, rather than the older current account crises.

Much of the IMF's work in the past decade has been learning more about capital account crises. What are the warning signals that a crisis is imminent? And how best to respond both to the warning signals, and, when necessary, the crises themselves?.

What makes capital account crises different?

Capital account crises differ in several ways from current account crises.

    1. In the first place, they can occur much more rapidly and require a much more immediate response than current account crises. This is because it is not only the current flows of goods and services that can require financing (and be subject to speculation) but also the stock of outstanding financial instruments.

    2. Capital account crises take place when holders of the country's debt lose confidence in the country's ability in the future to service its debt. Thus, a crisis can come about — at least in principle — even if the country's underlying macroeconomic policies are reasonably sound. At the same time, the shift from bank loans to bonds as the principal source of debt finance increases a country's vulnerability to a turnaround in confidence.

    3. Equally, the greater reliance on bond finance means that when there are genuine doubts about the sustainability of macroeconomic policy, those doubts can translate into a crisis very rapidly indeed. Policy inconsistencies are treated with great suspicion by the financial markets. It is very difficult for governments to buy time in such circumstances.

    4. Fixed exchange rates tend to compound the problems. One of the key lessons of the l990s is that, with open capital accounts, countries can be extremely vulnerable to capital account crises if there is any doubt about either the sustainability of the exchange rate peg or the sustainability of debt servicing. Thus, with a fixed exchange rate, there are two major sources of vulnerability: anticipation of exchange rate sustainability, or lack of it; and expectations regarding the future course of debt-servicing.

    5. The appropriate policy response must therefore aim to restore investors' belief that their holdings of the country's obligations will be serviced.

Responding to change

The Fund—and, of course, its membership—has responded to the challenges posed by the changing nature of crises in several ways.

First and foremost, we have placed much greater emphasis than ever before on the sustainability of macroeconomic policy. By this, I mean that we have gone far beyond what might be regarded as the traditional prescription of developing policy that delivers macroeconomic stability over the medium term.

Given that, as I noted earlier, crises can occur even when governments are pursuing what would be regarded as conventionally sound policies, they—and we—have to pay much close greater attention to factors such as debt sustainability and debt management. Governments need to be able to demonstrate to the financial markets that their overall debt burden is manageable: and is likely to remain so under most circumstances.

Whereas analyses of countries in earlier years focused on likely trends in the balance of payments, attention now turns as well to the future course of total debt, recognizing the country's planned fiscal trajectory and financing needs. Anticipating the primary surplus for future periods and the likely real interest rate the country is likely to be paying on its outstanding indebtedness enables a calculation of the evolution of the stock of debt. Contrasting that evolution with likely growth rates of real output enables a judgment (and sensitivity testing) of the country's ability to sustain its current policies. In Fund surveillance, debt sustainability is now routinely examined. In countries with Fund programs, debt sustainability going forward is a necessary condition for Fund support.

The balance between domestic and foreign debt is not the central issue here. Creditors will take an overall view of debt sustainability. The denomination of the debt, however, can be important, depending on the exchange rate regime is in place.

Partly for that reason, we have grown much more sceptical about the benefits of fixed exchange rates, and today far fewer countries, especially among emerging markets, are attempting to maintain fixed exchange rate regimes.

As a result of the experience with the capital account crises of the past decade or so, we now place much greater reliance on structural reforms. I know that when we in the IMF talk about structural reform people tend to assume we are talking about things like labour and product market reform. Of course such changes remain essential when developing sustainable policies.

However, we are increasingly aware of the need to place greater emphasis on reform in other areas of the economy. Countries need properly functioning and well-regulated financial sectors. As capital has rapidly grown more mobile, so have weak financial sectors become more vulnerable. Banks and other financial institutions need to be well-structured and properly supervised. The laws governing financial relationships—and here I include bankruptcy laws—need to be clear and effective. The Fund has worked hard to encourage governments to press ahead with such reforms.

We also recognize that many countries, some of them with very limited resources, can be overwhelmed by the speed of change. The IMF has in recent years worked hard to provide technical assistance to enable governments to develop and implement the sort of changes I've been describing. Such assistance might take the form of expert help in developing statistical reporting systems, or legal advice in drawing up new bankruptcy laws, or systems for trading public sector expenditures. The more the Fund can do to help countries develop effective policy frameworks as well as effective policies, the more successful will be our efforts to prevent crises before they start.

At the same time, because of the role of confidence in determining investors' behavior, the importance of transparency of policy has increased greatly. Encouraging—and helping—governments to develop sustainable policies has gone hand in hand with efforts to make such policies more transparent. The IMF itself now publishes many more of the documents it produces (including, with the agreement of the governments concerned, Article IV — surveillance — papers, as well as program papers). We have developed standards and codes meant to lead to greater transparency that we encourage governments to adopt. The easier it is to understand what is happening, whether it be at the policymaking level or, for example, in measuring financial sector stability, the easier it will be to reassure creditors and investors. If they are continuously informed of changes in economic policies and circumstances in individual countries, the likelihood of a sudden swing in sentiment is greatly reduced relative to those times when release of information "surprised" the market.


The cumulative impact of these changes has been significantly to strengthen policies in a number of countries. As such, we believe that the defenses against crises are much stronger than they were in the l990s. In light of the shocks to the international economy in the past three years — the bursting of the stock market bubble in the U.S., the disaster of September ll, and uncertainty over the geopolitical situation in the Middle East, among others — it is perhaps remarkable how few crises there have been.

Brazil is perhaps an excellent example of the Fund's role in crisis prevention when it is imminent and when underlying policies are sound. I'm sure Arminio Fraga will have more to say about his country's experience later; but let me make some brief comments in the context of the wider points I've been addressing.

By 2002, Brazil had adopted a floating exchange rate, had put in place an inflation targeting regime, and had fiscal policies promising a sufficient primary surplus to enable debt sustainability (and a projected reduction in the debt/GNP ratio), although the debt/GDP ratio was high.

However, Presidential elections were scheduled for the autumn, and market observers became increasingly doubtful whether the macro policies already in place would be sustained. The spread over libor rose to over 2400 basis points, the exchange rate depreciated to 3.9 reis per dollar, and rollover of debt became increasingly difficult and short term during the early summer. In response, the authorities negotiated a supporting program endorsed by all three major presidential candidates, which promised a continuation of the macroeconomic framework then in place.

That program has been carried out — indeed, strengthened with a higher primary surplus — since the new administration came into office. The exchange rate has since appreciated to less than 3 per U.S. dollar, while the spread over libor is now less than it was prior to the crisis — between 700 and 800 basis points. Most observers now believe that the macroeconomic framework already in place will be sustained. One can either conclude that there was no crisis, or that a crisis was well and rapidly resolved. Either way, the transparent policies of the Bank of Brazil, the macroeconomic framework (including inflation targeting and a floating exchange rate), and the Fund support were all vital elements of the turnaround.

Crisis management

As I said at the outset, crisis prevention remains our ultimate goal. But no matter how much we learn, and even when analysis is entirely on target, there will be occasional times when crisis occurs. In part, that is because there are always new unexpected events. But in part, crises will anyway sometimes happen because some governments cannot or will not adjust their economic policies in order to make them sustainable. As the lessons of crisis prevention and management become more widely accepted and understood, it is to be hoped that the political pressures on governments to undertake sustainable policies will reduce the instances where crises do occur.

Meanwhile, as the experience from the l990s and capital account crises is analyzed and better understood, the international economy is becoming increasingly integrated. That integration is enabling higher living standards and poverty reduction, but as the structure of the international economy changes, it is likely that there will be new, and as of now, unanticipated, sources of crises. The Fund, as well as the entire international policy community, will need to continue learning and adapting to new challenges as these emerge.

This inevitably means that alongside the efforts we have made—and are continuing to make—to strengthen crisis prevention, the Fund is also working to improve the effectiveness of crisis resolution. The aim, when crises do occur, is to be able to respond swiftly and decisively—and to minimise both short-term and longer-term harm to the affected country.

Because capital account crises can be so large and with such sudden onset, it is essential that, in these circumstances, sufficiently strong actions be taken to enable holders of sovereign debt to regain confidence. That necessarily entails taking tough measures. It is essential that any programme which the Fund negotiates with the affected country is sufficiently robust to persuade creditors, and the financial markets generally, that it will be effective. It has, in other words, to restore the prospect of sustainability.

Easier said than done, of course. Tough economic measures are usually politically unpalatable—especially so in cases where the country is the victim of circumstances unforeseen and largely beyond its control. It is a harsh fact of life in today's international economy that sustainable policies are those which bring the prospect of economic stability and prosperity in the future: not those which appeared to promise it in even the recent past. Governments, along with the international institutions, must be prepared constantly to adapt to changing circumstances.

I said earlier that policy programmes aimed at preventing crises now go beyond the core of macroeconomic policy; they embrace a wide range of structural reforms at the micro level; and, critically, they incorporate careful assessments of debt sustainability. The sorts of measures required for effective crisis prevention are, in broad terms, those that we aim for in crisis resolution. But once a crisis has erupted, governments are obliged by circumstances to move swiftly to implement such measures; often the speed of adjustment can be as painful as the adjustment itself. Governments that, for one reason or another, have been unwilling or unable to implement appropriate reforms can suddenly find themselves doing so in the worst of circumstances.

Governments can also find it harder if the international response to the crisis is unco-ordinated. This is an area where the IMF has tried to take the lead by strengthening links among the international institutions, for example, and among different groups of creditors. The more closely such groups can work together when trouble erupts, the smoother the process of restoring stability will be. Encouraging emerging market borrowers to include of collective action clauses in bond issues is a step in this direction: such clauses will, eventually, provide for much more orderly debt restructuring on those rare occasions when it might be necessary. It has been gratifying to see Mexico, Uruguay, South Africa and others adopt such clauses.

In the end, though, tough remedial measures will still be needed when crises strike. And it is hardly surprising that—just occasionally—the IMF gets the blame! We are, from time to time, criticised for making unreasonable demands on governments in crisis in return for the emergency funding they need. Of course, such scapegoating goes with the territory. There's little to be gained from being overly defensive about it.

Nevertheless, I think it is worth making a couple of observations. The first is to remind people that the IMF does not make demands. Even at times of extreme crisis, IMF programmes are drawn up in close consultation both with the country seeking help and with the Executive Board—as I said at the outset we are a member-based organisation. In a very real sense, countries are being helped by their peers. The nature of crises might have changed; and they tend to arise more swiftly. But the dilemma that affected governments face as a consequence are not necessarily all that different from many such circumstances in the past. A beleaguered government can often feel isolated and powerless. But it is in nobody's interest—and never has been—to make unreasonable demands, and the Fund does not seek to do so.

My second point concerns the confusion between cause and effect. By definition, when a country comes to the IMF for help in a crisis, it does so because it is in trouble. The policies it has in place are, for whatever reason, deemed unsustainable by the financial markets in general, and by its creditors in particular. At that point, of course, tougher measures are likely to be needed; and that the IMF can sometimes get the blame for their introduction.

But I would suggest that there is another way of looking at this issue; and that is to consider how international financial crises would play out if the Fund weren't there. Would they be easier to resolve without the resources and expertise of the IMF? Would governments be able to restore confidence in their economic policies more easily without us? I can certaintly think of no recent occasion when any government has tried to. Even when it has proved difficult for the Fund and a government to conclude negotiations on a new programmes—and sometimes such talks can go on for many months—the government in question has continued to try to reach a satisfactory outcome.

We need not limit this question to one about crisis resolution. Have our critics stopped to think what the world would be like without the IMF? Do they seriously believe that without us financial crises would be easier to prevent? Some might, of course. But I believe that without the efforts we and our member governments have made together in recent years, the challenge of developing and delivering sustainable economic policies would be much greater for many emerging market economies. The measures I've outlined today provide a benchmark against which governments can assess their progress and the effectiveness of their policies.

It is not, as I have tried to emphasise, a static process. Preventing, as well as resolving crises, is a dynamic process. That, of course, is more challenging. But it is a challenge that needs to be met. It is one that, for our part, we are trying hard to meet.

Thank you.


Public Affairs    Media Relations
E-mail: E-mail:
Fax: 202-623-6278 Phone: 202-623-7100