Financial Crises: What Have We Learned From Theory and Experience?

Peter Isard, Andrew Rose, and Assaf Razin
Published Date:
January 2000
  • ShareShare
Show Summary Details


Economic Counsellor and Director of Research Department, International Monetary Fund


Professor of Economics, University of California at Santa Cruz


Ford Professor of Economics and International Management, Massachusetts Institute of Technology


Managing Director, Deutsche Morgan Grenfell


Governor, Bank of Israel

Summary of Panel Remarks

Michael Mussa opened the panel by recalling the occasion of the first seminar he had given as a graduate student at the University of Chicago, which was scheduled for Monday August 19, 1969. The topic was devaluation, and as events would have it, President Pompidou devalued the French franc the preceding Sunday. The Brazilian authorities, by floating the real on January 15, the day before the present panel discussion, had provided similar background for this occasion.

The panelists spoke in alphabetical order. Michael Dooley observed that as initial reactions to the Brazilian float, the Brazilian and U.S. stock markets had risen on January 15. Perhaps this reflected sentiment that abandoning the peg would have an expansionary effect on output in Brazil. Regardless of the ultimate outcome for Brazil, however, one thing we had learned from experience was that the bevavior of real output following financial crises had varied widely across different episodes. The United Kingdom had actually prospered following Britain’s withdrawal from the ERM in September 1992, while in contrast, the output losses that followed the 1994 Mexican crisis were very large.

Dooley’s remarks were focused on two related questions. Why have there been such large differences in the output costs that have followed financial crises? And where do the output costs come from? As a trivial point, we knew that cases of large output costs were cases in which exchange rate movements, along with the interest rate policies that governments had followed in response to the crises, had bankrupted a lot of firms and banks, and had even bankrupted the governments themselves. But if financial market participants knew that surges in capital flows into an emerging market economy would create an environment that was ripe for a financial crisis with large output losses, why did we see such large capital inflows in the first place?

One possible explanation was that market participants were irrational. A second line of explanation, consistent with the hypothesis of rational behavior, was that private investors regarded their claims on emerging market economies as “insured.” Despite the fact that in some crisis episodes private investors had suffered large losses, Dooley was a proponent of the second view.

In particular, Dooley argued that private investors create “insurance” by structuring their claims on a country in a manner that will impose substantial costs on the country if it fails to meet its financial obligations. The success of private investors in insuring that financial obligations were met required the threat of a financial crisis that in turn would precipitate a significant downturn in economic activity. An intriguing element of the paradigm, in Dooley’s view, was the idea that even a small miscalculation would trigger a crisis right away. This created a game with official creditors, in which the IMF or creditor-country governments had to decide quickly whether to bail everyone out.

In closing, Dooley conjectured that the large output losses associated with financial crises were accidents. Private lenders to an emerging market economy set up, endogenously, a structure of claims on the country’s government, banks, and firms that they believed would inflict the maximum pain on that economy—and on their own economies—if the claims were not serviced and repaid. Occasionally they miscalculated. And when they did, there was no effective way to renegotiate, and we were stuck in a bad equilibrium in which the large real costs were imposed.

Rudiger Dornbusch addressed the question of how to model financial crises. In his view, the paradigm provided by Salant and Henderson (1978) and Flood and Garber (1984) was still valid, and a good model of a modern-day financial crisis had to include a focus on three basic elements: vulnerability, globalization, and illiquidity. Models that emphasized “special effects” without addressing these basic elements were not very useful.

Vulnerability could emerge in the corporate-sector balance sheet, the banking-sector balance sheet, or the government’s balance sheet. And if vulnerability surfaced in any one of these places, it would likely end up in all three. Moreover, once a country’s balance sheet became vulnerable, there was only one place people would want to hold their assets, and that was outside the country. So any balance sheet problem would inevitably become an exchange rate problem. The degree to which the exchange rate looked over-valued or undervalued according to traditional perspectives was irrelevant when the country’s balance sheet was regarded as vulnerable.

Globalization was also an important part of the modern-day setting. It was silly to focus on small-country models of financial crises. The Koreans were playing in Brazil, the Brazilians were playing in Russia, and the three were holding hands. And because portfolio positions were leveraged, it didn’t take very much to draw attention to the prospect that all three would go bust together.

Illiquidity was the third basic element in the story. Dornbusch recalled that the discussion of illiquidity had been very important in the 1930s and cited a paper by McKean (1949) on liquidation scrambles. McKean had emphasized the need for countries to pay careful attention to maturity structure in managing their balance sheets, recognizing that when something goes wrong, everybody wants liquidity.

Together, vulnerability, globalization, and liquidity scrambles were, in Dornbusch’s view, the essence of the balance-of-payments and exchange rate crises we see today, which had become fast-action crises rather than old-style current account crises that lacked an interesting capital account story. In trying to anticipate crises, one had to ask the question: Where is the vulnerability hidden? The Brazilian crisis was no surprise. With large stocks of short-term dollar-denominated external debt, a substantial budget deficit, an overvalued exchange rate, and a banking system that couldn’t stand six months of high interest rates, vulnerablity was extensive.

What should come next for Brazil? Dornbusch thought the intelligent answer was to recognize that when a country had experienced zero growth in per capita GDP over 20 years for macroeconomic reasons, it was time to try to get rid of macroeconomics. To him, a currency board was the obvious answer. There were many arguments against a currency board, the leading one being that it wasn’t a perfect arrangement. But nothing perfect was about to happen in Brazil. As Dornbusch saw things, a currency board had the advantages of taking the exchange rate off the table and taking the central bank out of business. In his view, a currency board arrangement had worked spectacularly in Argentina, not only in the sense of promoting inflation stability narrowly defined, but much more so in lengthening the decision-making horizons of economic participants.

David Folkerts-Landau spoke from the perspective of a former staff member looking at the Fund from the outside. He regarded surveillance as the core of the Fund’s responsibilities and was strongly critical of the Fund’s performance in recent years. The fact that the Fund had failed to anticipate the magnitude and scope of the current financial crisis was not the central issue; nobody else had anticipated it either. The crux of Folkert-Landau’s criticism, rather, reflected his view that the Fund had failed to respond adequately to changes in the global economy and the global financial environment.

Folkerts-Landau regarded the Fund’s culture as still predominantly influenced by a paradigm of macroeconomic behavior that was relevant to the 1970s and 1980s, but not to the 1990s. While the resources that the Fund was devoting to financial sector surveillance had increased, thanks mainly to strong efforts by Jacob Frenkel and Michael Mussa, they remained relatively small. Resource constraints, moreover, were not the only factor that limited the effectiveness of financial sector surveillance. In addition, the effectiveness of the Fund’s general surveillance efforts was hampered by the difficulties that the Fund confronted in criticising member countries and by the incentive structure that that the organization provided for its staff members. In Folkerts-Landau’s view, part of the problem was that staff members who played instrumental roles in shaping the Fund’s policy advice to member countries were not held adequately accountable by the Fund’s senior management—that is to say, they were not appropriately rewarded or penalized for the successes or failures of the policy programs they recommended.

Folkerts-Landau was also critical of the Fund’s performance in the areas of crisis management and crisis resolution. He felt that in some prominent cases, such as Brazil, the Fund had been inclined to support programs based on projections that were widely regarded as unrealistically optimistic. As a result, the Fund was now widely perceived as “talking its own book” and no longer operating on the basis of first-rate analysis. This was a serious matter for an institution as well regarded as the Fund, and for an institution that had been given the special role of promoting the stability of the international monetary system.

Jacob Frenkel spoke next. He subsequently submitted a written version of his remarks, which follows this summary.

Michael Mussa focused first on the question of what we had learned about the causes of crises. He viewed crises as very complex events with a multiplicity of causes. A good analogy was provided by the sinking of the Titanic. The simple explanation described the cause as an iceberg. But we know that the disaster should be attributed to much more than the iceberg—in particular, to how the ship was constructed and operated, to how it was managed after it hit the iceberg, to what the radio man on the California was or was not doing, and so forth.

An even more complex story needed to be told about the economic disasters that had befallen a number of countries around the world over the past few years. Certainly the fundamentals mattered. Brazil had a fiscal problem. Russia had not only a fiscal problem, but also a massive problem associated with a culture of nonpayment and a government that was incapable of carrying out some of its most basic functions. A number of the Asian economies had problems in their financial sectors, and in the financial structures of their business sectors; and they also suffered a variety of adverse external disturbances, including a sharp decline in the volume of world trade in 1996 and unfavorable movements in the dollar/yen exchange rate.

Another contributing factor that Mussa regarded as important was the nature of the policy response as the crisis developed. In Indonesia, which at the outset of its crisis had a very powerful government led by a very powerful president, it was the president’s failure to act initially in a sufficiently constructive manner that helped to transform a situation of severe economic difficulties into an economic catastrophe. Disfunctional policy responses had also been important, in Mussa’s view, at the initial stages of the difficulties in Korea and Thailand.

Moral hazard was an additional contributing element. Significant parts of the imbalances or disequilibria that developed in these economies, and that turned into real problems as the economies fell into difficulties, were the consequences of actual or perceived guarantees, by the national governments, of various businesses and financial institutions. Still another contributing factor was contagion—not only financial contagion, but also the real spillover effects that countries feel when trading partners experience severe economic contractions.

The second issue addressed by Mussa was the perception that the economic costs of crises have been excessive—beyond what reasonably needed to occur, given the circumstances of the economies before the crises started. The cumulative output loss for Korea, over a horizon of roughly 6 years, was projected to amount to about 50 percent of Korea’s annual GDP. The projected output loss for Indonesia looked even worse. It was Mussa’s perception that once we get into the crisis mode, the system disfunctions. It disfunctions at the national level because of our incapacity to resolve bankruptcy problems and the like. It also disfunctions at the international level as creditors scramble madly to pull their money out, which then pushes the economy into a situation akin to national default, which in turn creates a massive disfunction in the real economy as well as in the financial sector. Regardless of whether or not one viewed this as a situation of multiple equilibria, Mussa felt that we needed to focus considerable attention on trying to understand why the system disfunctions and what could be done to avoid the various types of disfunctional responses of a system in crisis.

Turning next to some of the issues that Folkerts-Landau had raised, Mussa argued that the most that the Fund could be expected to do in the area of predicting crises was to identify areas of vulnerability. With regard to the Fund’s surveillance activities and internal dialog, considerable improvements in recognizing countries vulnerable to crises had been made in recent years. It was the Fund’s job to identify vulnerabilities, to work on policy-reform efforts aimed at correcting those vulnerabilities, and to help deal with crises when they came. But in Mussa’s view, it was not feasible for the Fund to identify, in any public way, the countries that it regarded as on the hit list for the next financial crisis. That was a function that needed to be performed much more by private financial markets, which did not have the type of membership responsibilities that constrained the Fund.


    FloodRobert and PeterGarber. (1984). “Collapsing Exchange Rate Regimes: Some Linear Examples.”Journal of International Economics17113.

    McKeanRoland N. and (1949). “Liquidity and a National Balance Sheet.”Journal of Political Economy57506522. Reprinted inFriedrich A.Lutz and Lloyd W.Mints (eds.) Readings in Monetary Theory.HomewoodIII.:Richard D.Irwin1951.

    SalantStephen and DaleHenderson. (1978). “Market Anticipation of Government Policy and the Price of Gold.”Journal of Political Economy86627648.



When we ask ourselves what we should learn from financial crises, we may first ask ourselves which crises we try to understand because, unfortunately, there have been several periods of crises over, say, the last three decades. Certainly there are important differences between the debt crisis of the late 1970s and early 1980s, the ERM crisis of the early 1990s, the Mexican crisis and its “Tequila-effect” in 1994/95, and the recent crises in South East Asia and Russia with the ensuing world financial turmoil. However, there seems to be one common factor that manifests itself in each episode in a somewhat different disguise: moral hazard.

But before we proceed with the issue of moral hazard, we may question the timing of the learning process: when should one start learning? When the horses have left the stable? This reminds me of the criminal who was heard mumbling on his way to the electric chair, saying to himself, “Well, this will teach me a lesson!” The fact is that the present situation is not so much the end of an era as it is a mid-point—each point is a mid-point. One may of course ask the Fund, “How could this organization have made so many mistakes?” But for that matter, a similar question could be put to the crisis country’s authorities (today this would be the Brazilian authorities): “How could you have made so many mistakes?” This reminds me yet of another story: There was this little schoolboy, who had written an exam, and when returning the exam to him, the teacher exclaimed: “My goodness, how could one person make so many mistakes?” And the pupil answered: “I did not do it alone, my father helped me.”

So, what is the solution? David Folkerts-Landau proposed a deep change within the staff, a change of the incentive structure; one could go on and on—a change of top government officials, etc. Reshuffling international organizations, governments, and other important national institutions requires a great deal of motivation and time.

What Have We Learned?

I remember that, when I came to the University of Chicago together with Rudi Dornbusch and Michael Mussa as graduate students, I was impressed by Milton Friedman’s capacity to go from one country to another, not knowing much or anything about the details of the country, but coming home with very strong recommendations. When asked about how he was so sure about the necessary steps, he used to say: “I know what data to look at”—needless to say, the relevant data were the monetary data.

When I joined the Fund, the many young and talented economists, going from one country to another (and membership has doubled since then) knew very clearly what economic prescriptions to recommend after very short missions in the country. When I asked them, how they know what to do, they told me that the key figure to look at is the country’s government-budget deficit.

It turns out that, in many cases, classified as crises today, neither the money supply nor the budget deficit would have been useful indicators for predicting the crisis. So now there is a new fashion—probably the right one—emphasizing the functioning of the financial system, the soundness of the banking system, the existence of moral hazard. These concepts were not mentioned in the textbooks of the past. This suggests, that with this time perspective, when asking “What have we learned?” the question is not about what we have learned since last Friday, although we may state that quite a remarkable number of things have happened within such a short period of time. A crucial point in this respect is that we are in a new era—an era of globalization. The meaning of globalization has at least two facets:

  • Geography matters much less than before. It has become less important to know where a crisis started. In this sense, the name “Asian crisis” is a misnomer, because in many respects, countries in Latin America exhibit similar characteristics to those associated with the Asian or the Russian crisis. The first element of globalization is that geography has been substituted by a more functionally-related phenomenon.

  • Time is not linear any more. Not in a backward-looking sense, but in a forward-looking sense: the time available to policy makers, to surmount an economic crisis by choosing the right actions, gets shorter and rapidly so.

The Butterfly Effect

You probably recall the book by James Gleick (1997). It starts with a butterfly that moved its wings somewhere in the Pacific, and through the laws of physics in a frictionless world, it created a typhoon and a hurricane in some faraway place. Of course, this does not really happen, because in reality there is friction. And it is precisely globalization that brings us closer to the (frictionless) chaos—we may call it the “butterfly factor.” Rudi Dornbusch was perfectly right by noting that vulnerability, or the balance sheet problem, is not of a single entity. Indeed, looking at Korea, and recognizing the extraordinary cross-guaranties among firms and other institutions within the same conglomerates, helps us understand that the notion of a balance sheet has stopped being merely a notion of an enterprise, and has become a notion of the transmission mechanism of difficulties. And Rudi Dornbusch is also right in pointing at the exchange market, where turmoil manifests itself most rapidly. This is vividly illustrated in Figure 1.

Figure 1.Nominal Exchange Rates Against the U.S. Dollar.

Developed Economies: United Kingdom, France and Germany Emerging Economies: India, China, Australia, New-Zealand, Taiwan, Mexico and Brazil Crisis Economies: Thailand, Malaysia, Philippines, South Korea, Indonesia, Singapore and Russia

Another financial market importantly sensitive to crisis is of course the stock market. Also in this market the difference between crisis countries and any other category of countries is vividly visible; see Figure 2.

Figure 2.Equity Indices.

Developed Economies: United States, United Kingdom, France and Germany Emerging Economies: India, China, Australia, New-Zealand, Taiwan, Mexico, Brazil and Argentina Crisis Economies: Thailand, Malaysia, Philippines, South Korea, Hong Kong, Indonesia, Singapore and Russia

The Currency Board

We should always keep in mind one thing and try to address it up front. In a currency board system, a liquidity crisis gets easily translated into a financial crisis. And Argentina’s experience, i.e. the post-tequila effects, must teach us a lesson, not as a case against currency boards, but rather about what should be done in order to avoid the situation of the liquidity crisis being transformed into a full-fledged financial crisis.

Banking Supervision, Globalization and Financial Crisis

This brings me to the concept that has recently gained a lot of attention—the soundness of the financial system. There is now greater understanding and consensus that in order to carry out a successful macroeconomic policy you must have a sound financial system, especially a strong banking system. If your banking system is not sound and there is a crisis, it may transform itself into a macroeconomic crisis. By the same token, if you have a macroeconomic system that is not stable and if your banks are not sound enough, then an unstable macroeconomic system can transform itself into a banking panic, or a banking collapse. So, we see the recent interaction between macroeconomic stability on the one hand, and soundness of the banking system on the other.

Furthermore, in order to be able to pursue an effective monetary policy, you must make sure that you are not constrained by a fragile banking system that may prevent you from imposing or adopting the appropriate policy measures because you are afraid to shatter the banking system. You’d better have a strong and sound banking system. Furthermore, since one of the important channels of transmission—the effects of monetary policy—works through the banking system, in order to ascertain the full efficacy of monetary policy, you’d better have a banking system that functions well and that is sound. All this suggests (and the lessons from Asia reveal) that in the modern era the soundness of the financial system has become an extremely important element of successful policymaking. Therefore, much effort is now being put by Central Banks the IMF and the BIS (Bank for International Settelments) into strengthening bank supervision.

How Should We Define Successful Supervision?

There are two issues that must be emphasized here. First, financial markets are very innovative. And second, because of the fact that they are so innovative, these markets are relentlessly developing financial instruments (like derivatives) and new technologies. Therefore supervisors with the institutional and technical knowledge of yesterday are not adequately equipped, to deal with financial developments of tomorrow. This situation implies that banking supervisors must improve their knowledge continually because the subject is very dynamic. It follows, therefore, that the concept of banking soundness is also a dynamic one and should not be defined in terms of a rigid set of technical conditions but rather adapt itself to the changing realities.

As we analyze developments in Asia, we realize that most of the problems there stem from moral hazard: Financial intermediaries extended loans that were very risky. Those who deposited financial resources with the financial intermediaries did so under the (implicit or explicit) assumption that the government provides a guarantee to the intermediaries. Therefore they had no incentive to monitor the quality of the loans extended by the intermediaries. They just assumed that all was safe. Managers of the intermediaries did not have the incentive to gauge carefully the riskiness of their loans because they always trusted the implicit government guarantee. They also just assumed that all was safe. In many cases these loans were extended in order to purchase assets such as real estate. The growing demand for such “assets,” financed by the readily available credit, brought about a continuous rise in their price since the mechanism that normally puts a check on such a trend, i.e. the rise in the perceived risk associated with the growing stock of debt, was absent. The implicit assumption that all is safe since the government provides insurance without charging the appropriate premium, created the environment of a “One-way bet” and planted the seeds for a bubble—for the analysis and understanding of which the economics profession owes a large debt to Bob Flood. Depositors put their money into the financial intermediaries with the idea that it was safe and secure. The financial intermediaries, with these resources, gave the loans to risky enterprises, like real estate, thus blowing up asset prices. And basically there was a vicious circle that kept everyone happy, as long as it worked. The depositors saw their assets grow. Real estate developers got financial resources from their banks and financial intermediaries. Financial intermediaries saw the collateral on their books go up in price, and everything was all right up to the point at which the first burst in the bubble took place. At that time, when it turned out, that the government was not fully securing particular loans, suddenly the flow of resources into the intermediaries dwindled. And therefore the engine that had generated the rise in asset prices stopped working. Asset prices tumbled. The collateral on the books of the financial institutions dwindled. Loans had to be called back and soon certain institutions became insolvent and the crisis of the financial sector turned into a major economic crisis.

Figure 3.High Yield Market: Issuance and Spread


Two issues arise: How should supervisors deal with fragile banks? How should they deal with prospective crises? Should they deal with five of, say, seven potential crises, or should they “solve” seven of, say, five potential crises? Before the crisis sets off, nobody knows when and where the crisis will emanate. And the supervisor stands before a strategic choice. Should he act with excessive prudence and “prevent” seven out of five potential crises, or should there be less than perfect prudence—prevention of only five out of seven potential crises? Which is better? If you are going to prevent “seven of the five” crises, it means that you are over-regulating the banks; you are excessively cautious. You are therefore paralyzing the system, by deterring the development of the market for risk taking, which is the essence of entrepreneurship. You are—metaphorically speaking—stopping too much traffic. But if you are preventing only five of the seven crises, then you leave some possibility of crises in the system. That is the strategic question.

If you adopt the strategy of preventing five of seven crises, then you should be aware that crises might occur, and that the authorities should be ready to deal with them. An important lesson that has been learned is that it is essential to deal with an insolvent institution early on, and to close it rather than to inject funds into it without sense and make a big problem of an initially small one. It should be understood that a supervisor should not think of any bank failure as the end of his career, because this would imply that no bank will ever fail. And we know that a system, which does not allow any bank ever to fail, is an excessively prudent system. We must remember that most problems that were allowed to grow until they became part of the “too big to fail” syndrome, were initially small problems that were ignored for too long because they were just “too small to be bothered with.” Furthermore one should make sure to avoid cross guarantees between related businesses, conflicts of interest between the public and the private sector. Corruption has to be eradicated. A very efficient way of dealing with such problems is to increase transparency and accountability. All these terms have become slogans, but that doesn’t change the fact that they are relevant.

A former colleague of mine at the University of Chicago, the Nobel Laureate, Bob Lucas, once asked me how many times I had missed a plane. And when I told him that I never missed a plane, he told me “You must be wasting a lot of time at airports.” He asked me how many times I had gotten a ticket for speeding. I said, “Never.” “Well,” he replied, “you must be driving systematically under the speed limit.” Well, analogies may have their limits, but there is a point in this. The point is that if you are not going to waste too much time in airports, you must be ready, maybe once or twice, to miss a plane, but also be prepared to know what to do about it. This suggests that in spite of the desire to avoid crises, we must be aware of the fact that if we avoid all crises at all cost, we are overdoing it. It is better to design mechanisms and instruments that reduce the cost of the few crises, once they occur, than to upset the efficient mechanism of risk taking by over-regulation designed to eliminate the likelihood of crises at all cost. Experience in the past half-year or so shows that financial markets tend to be overshooting. Over the previous two years international financial investors have clearly been chasing excessively after yields, without paying due regard to the risks associated. This is common in situations of moral hazard, because it implies that somebody will bail you out if trouble appears. There is an implicit insurance, while the financial investor did not have to pay any insurance premium.

Moral Hazard and the Exchange Rate System

A similar problem is encountered in relation to the exchange rate risk. If a country’s authorities promise a fixed exchange rate system (this is a close relative to the currency board) it is also planting the seeds for moral hazard. The business sector will take this promise into account and assume that there is some kind of a commitment. After the collapse, especially in the aftermath of the Russian crisis—the most recent World Economic Outlook presents a fascinating analysis of the world financial crisis—there clearly occurred an overshooting of risk premia and an excessive drying up of institutional savings. While before the Russian crisis one could observe an excessive chase after yields and a neglect of risk, after the Russian debt default the wheel turned around and we could observe an excessive chase after liquidity, with confidence sitting on the fence, willing to forego tremendous profit opportunities. Interest spreads went up and new issuances dropped, as you may observe in Figure 3.

World Financial Crisis and Economic Reform

One of the lessons obtained from the financial crisis in Asia, and also one of the issues the IMF and the World Bank have been working on very hard—is the need to have early identification and prompt closure of insolvent institutions. Some economies have tried to confront the financial crisis by considering a retreat from the strategy of openness regarding the capital account of the balance of payments. The approach was, in some places, “Let’s close the capital account,” which is similar to the reaction “let’s close the window when too much wind comes in through the window,” while forgetting that with the open window, you not only get wind but also oxygen. Another idea was to put “sand in the wheels”—or in our context—taxes on capital flows (as if this would make things simpler). I think that anyone who has tried to implement the idea of imposing controls on capital flows in practice rather than just preaching it, has recognized how futile and expensive such a strategy is. Any mechanic knows that if you put sand in the wheels, it’s very difficult to get it out once you wish to get rid of it. The problem is that it is illusory to believe that you can prevent only the volatile capital flows, while hoping that the “good” capital flows will continue to flow into the country. Like in a diet, it is very difficult to separate the “good” from the “bad” cholesterol. Closing or hampering the functioning of the capital account is like saying that in order to prevent a car accident, you should close the road. Well, we know that’s not the right way to deal with the problem of car accidents. The right way to deal with the occurrence of car accidents is—among other things—to widen the road, rather than narrow it, in order to lower the likelihood of an accident, and also to install seat belts in order to lower the cost of accidents, once they occur. In discussions on the strategy of capital market liberalization the question whether we should put “sand in the wheels” has become a major issue in the international financial system. Should we slow capital flows, or should we—metaphorically speaking—“widen the road” and install seat belts?

As a matter of fact, we have seen heated debates between the IMF and the World Bank on that issue, where the IMF is more reluctant to impose capital controls and some in the World Bank are less reluctant. The IMF is reminding everyone of what Churchill used to say that “markets are like parachutes. They work best when they are open.” And there is a lot to that; in this debate my own view is closer to that of the IMF.

Another lesson that we have learned from the recent Asian financial crisis—and the IMF came out with this conclusion—is the following:

Those countries that have not yet fully opened their capital account of the balance of payments should proceed slowly, whereas those countries that have already opened it, should not retreat. Not because you should not retreat in order to “save face” but because once the market participants have learnt to function within open markets, this is irreversible, just as there is no way you can divert water in a permanent way. You can distort it’s course, but the water will find its way.

Another issue that came up in the debate on financial systems, was whether reform should be slow or fast? Should it be gradual, or should it be drastic? In my own judgment, it should be implemented according to a specific order. First, we should create the preconditions. Once the preconditions are in place, the system may be changed rapidly. You know Mendes France, the Prime Minister of France long ago, wanted to convince the French to drink less alcohol. So he had signs put in the metro stations of Paris saying, “Stop drinking alcohol, because it will kill you slowly.” And somebody wrote underneath this sign, “That’s okay. I’m not in a hurry.” The fact is that while one may not be in a hurry to create the preconditions and the capacity of supervisors, and to create the legal and the prudential system, once you are ready, you’d better do it quickly and not in stages. In the United Kingdom where cars ride on the left-hand side of the road, it was said that when they considered to adopt the American system where you drive on the right hand side of the road, there was a proposal to do it gradually. On Mondays the trucks will move, on Tuesday the bicycles, and so on .... Of course, such changes can not be implemented gradually. Such changes must be adopted at once; this is a case for a drastic move.

Capital markets are unique. They have memory, and yet one can be baffled. I remember in 1995 when there was the so-called “Tequila Crisis” in Mexico. We saw a lot of money flowing out of Mexico and Latin America into Asia, to the “tigers.” If I had told you then, in 1995, that within 2 years we would see money flow out of the “tiger countries,” instead of coming in, and if you had asked: “Where will the money go, that will be flowing out?” and if I had answered: “Back to Mexico, to Latin America,” you would have said that I was irrational. Yet this is exactly what has happened. And the question is, why? Are the markets irrational? Don’t they have memory? Somebody said once that there are two kinds of investors, those with short memory—the bankers—and those who have no memory—the institutional investors and international organizations. I think that the markets do have memory, but the fact of the matter is, that markets respond quickly to policy changes, and they are good in doing so. We have seen dramatic changes in economic policy in Latin America in general, and in Mexico in particular and, therefore, it is not by accident that we see the money that left these countries making a U-turn and coming back.

I was told that in Chinese the word “crisis” consists of two characters: one meaning danger and the other meaning opportunity. The successful handling of a crisis combines the recognition that a crisis contains both the elements of danger and of opportunity. This is another way of saying that the problem of moral hazard needs to be addressed.

Thank you.


    Other Resources Citing This Publication