[TRANSCRIPT PREPARED FROM A TAPE RECORDING.]

PROCEEDINGS

MR. PERRY: --lessons that we are learning from the Asian crisis for the future management of Latin American economies, and we have composed a panel in which we will begin with Michael Mussa from the IMF; then, we had expected Charles Dallara to go second, but he seems to be a little late; and then we will go with Peter Garber from Brown University and Arminio Fraga from Soros Foundation, and finally, Joe Stiglitz.

I will first make a short introduction, showing how, in the Latin American part of the Bank, we have been interpreting the lessons, and then, each of the panelists will have an initial round of about 10 minutes. We will then ask some of them if they want to comment on what the others have said, and then we will open it up to questions.

So let me begin by saying that what we have seen in the Asian crisis is that financial vulnerability understood has a high likelihood of successful currency attack, but also, somehow related to the degree of severity of the crisis, we saw four aspects of vulnerability clearly present in most countries--a traditional external sustainability risk emerging because of appreciation of real exchange rates, slowdown in the rate of export growth, export receipts, and even higher rising current account deficits, together with a high liquidity or rollover risk, because several countries had very high short-term debt to reserve ratios; also, widespread currency mismatch risks, unhedged currency exposures, in some cases of the financial sector, in others, of the corporate sector; and finally, risks toward any interest rate hike or slowdown because the corporate sector was highly indebted and because the quality of loan performance was weak after a prolonged credit and asset price boom.

The way we see it, these interacted, one with another, in the sense that vulnerability to an attack rose when this emerging external sustainability risk indicated the possibility of a currency adjustment sometime in the future, but then the liquidity risk invited herd-like behavior, and the speculators knew that the authorities had their hands tied, so to say, because they could not or they did not want to implement a timely devaluation due to the fact that there were great currency mismatches in both the real and the financial sectors, and because they were not in a position to defend the currency with an interest rate hike given the high indebtedness of corporations and the poor long portfolios.

In the paper that we have distributed, we look at all these indicators in comparison for the five countries of Asia that went into crisis, non-crisis Asian countries and Latin American countries, and I am showing just a few of these numbers here.

First, the differences with Latin America on the side of the more traditional external sustainability risks were not so high. Clearly, in most of the Asian countries, there had been a noticeable slowdown in exports, in the growth of exports receipts in dollars, something that you did not see as clearly in Latin America, although in some countries in Latin America, the growth of export receipts was very slow all the time; but there was not a marked slowdown.

In any case, in appreciation, you don't see much of a difference. There were countries in Latin America with stronger appreciation than last year, some with less appreciation. And current account deficits--the stories are not so different there. They are more different when you look at the liquidity risks. Really, the short-term debt-to-reserves ratio for Indonesia, South Korea and Thailand, which we have here, are well above one in all the countries at the beginning of the crisis, and the ratios are much below for the Latin American countries. The larger countries have larger ratios, and the smaller countries have smaller ratios.

As far as currency mismatches, we see some countries in Asia in which the financial sector itself clearly has a strong currency mismatch growing; in others, it is not as noticeable. On the corporate sector, we don't have very good data, but in any case, the impressionistic evidence is that currency mismatches in the corporate sector were not as widespread as in Asia, and the coverage of reserves to M-2, for example, was in general much higher in Latin American countries.

And finally, the vulnerability of the financial sector was clearly less in Latin America, partly because in the last years from 1994 onward, different from the Asian countries, we had not had strong credit booms. Maybe Peru was the only country in the region that had it, but beginning with a very low financial deepening. Somehow, after the 1994 crisis, credit growth had been reduced in most of Latin America, and the financial sector had improved substantially.

But we think these four things are basically symptoms of vulnerability and are not the root causes. The root causes, we find in a financial and especially a capital account liberalization that, by the way, was very strangely sequenced in Asia, in our own view in Latin America, in the sense that long-term capital flows are restricted in most countries, while short-term flows are stimulated. But this took place in the period in which capital flows were rising and over a structure of very perverse incentives, incentives that led the private sector to take excessive risks in the presence of these funds, these flows--rigid exchange rate systems; widespread moral hazards in the financial sector, but also in the corporate sector--the government was seen as not letting anybody go bankrupt, not the banks, not the big corporations--lack of transparency in corporate governance and financial transactions which led the corporate sector to take high risks with nobody to notice it or even the financial sector to finance it. All of these perverse incentives were unchecked basically by lax prudential regulation and supervision.

Now, when we look at these root causes in Latin America, we find a much better situation, so we understand why Latin America has not developed this high financial vulnerability. Capital account liberalization, if it had been done, had been more in the past, and the incentive structure was much better. Most countries have more exchange rate flexibility; moral hazard was much less; transparency was not the best, but it was much better; and regulation and supervision had improved a lot, and there had been substantial restructuring of the banking sector.

However, Latin America was affected somehow, through spillovers. The financial spillovers came especially after the attack on the Hong Kong dollar on October 22. There were demonstration effects. Investors were looking at countries with similar problems. There were "flight to safety" effects. There were "cash-in" effects; some of the funds had to get cash to cover losses of depositors. So that led to the beginning of the end of asset price booms for most Latin American countries and some exchange rate pressures in the short term, especially in the case of Brazil, and then to something of a reduced and most costly access to international capital markets for a while, and that has affected different countries in different ways.

But probably more important have been the effects on the real side. Some countries already had an important part of their exports going to Asia, so there is some direct effect of demand contraction. But more important is what has happened to commodity prices. The sharp fall in commodity prices has not been all due to the Asian crisis, but it has had an important effect, and this is probably the most important effect today for countries like Chile, Peru, Venezuela, Mexico and Colombia.

And then, there is the effect of competitiveness, which we are just still beginning to feel. Our calculations show that this, in the worst of circumstances, could be around one percent of GDP for Mexico, which wasn't playing for just a year, and something on the order of 0.3 percent of GDP for Brazil, and for the rest, basically insignificant.

The other thing is that most of the authorities have reacted--these trade effects have created a trend toward current account deterioration, further current account deterioration, in a situation in which these deficits are going to be a little more difficult, a little more costly to finance. So most of the authorities in the region have reacted in a prudential way by raising interest rates more than the markets require and by having fiscal contraction, and we believe that this has been a prudent reaction, but it magnifies somehow the effects of the growth.

Let me go directly to the last--basically, we think the financial effects are well over, and the trade effects are still playing through. Here, what we have is our estimates--present estimates--of growth. What we see is that clearly, after a year of very high growth in Latin America--5.2 percent weighted mean which, without Brazil, was 6.2, the best year in the nineties for most Latin American countries--we expect now something between 3.1 and 3.4. Not all of this deceleration is due to the direct or indirect effects of the Asian crisis, of course. There are other factors. Some of the countries which grew very rapidly last year were going to have a slowdown this year in any case. But we do think that the effects are not insignificant in any case.

Now, the lessons. The first lesson we see here is that institutions matter a lot, not only for efficiency and growth, which we already knew, but for stability. I think that since the Mexican crisis, we had known that the institutions--the rules and enforcement of rules of private sector behavior for the financial sector were very important. Now we are seeing that those for corporate governance and disclosure, not only of the financial sector but of the real sector, are also important. And they are all very important lessons--and when we talk of all lessons, they are things like prudent macro management of the fiscal, monetary and external sectors; prudent debt management, especially in terms of level and composition; prudent capital account liberalization if you don't have good institutions in place; and either a flexible exchange rate system, more flexible, or a [inaudible] system with very strong commitments that make them very credible.

So let me just stop there and give the floor to Michael Mussa.

MR. MUSSA: We have been asked to reflect on the Asian crisis and its lessons for Latin America. This issue of the Asian crisis seems to be a disease which is going around to a whole host of conferences these days, and I expect it will continue, at least for some more months; then it will calm down, and then the next crisis will come, and we'll have "The Lessons of X for Y" once again.

I think the most important lesson for Latin America from the Asian crisis is that it could happen in Latin America. Indeed, it has happened in Latin America before, more than once, and there is certainly nothing that provides any assurance that it will not happen again. Indeed, I think it is very likely at some point that there will once again be a financial crisis affecting, on a broader or a smaller scale, several Latin American countries.

The task needs to be to make those crises less frequent and less damaging than they have been in the past. And from that perspective, I think there are important lessons to be learned from the experience in Asia.

I think Guillermo has provided us with an excellent summary, really, of all of the main factors that have contributed to the financial crisis in Asia, with the exception of one additional factor which I do want to emphasize, and that is the external developments which contributed to the crisis in Asia.

World capital markets were clearly flooded with liquidity from industrial countries and were pumping capital into East Asian and other economies in the beginning of the decade, much as the banks pumped capital into Latin America in the late 1970s, and this contributed to the credit boom. Movements in the dollar-yen exchange rate tended to undermine trade positions and competitiveness for countries that had linked their currencies to the dollar. And there are other external factors as well.

I like to think about crises in terms of the disaster movie, "Titanic," a great movie. I think it is relevant to ask two questions. First, when would we have forecast that the Titanic would sink, with the loss of 1,500 souls; and second, what caused the wreck?

Well, the answer to the first question is: Only after it hit the iceberg, and probably not even then. And the answer to the second question, of course, is that the iceberg caused the sinking of the Titanic.

Now, actually, the story is a good deal more complicated than that--the design of the ship, the way in which the ship was operated in a dangerous situation, the number of lifeboats and the way in which the lifeboats were handled, whether radio operators or other observers on other craft in the neighborhood properly identified and responded to the distress signals--all of those things contributed to the ultimate magnitude of the disaster. Similarly, in financial crises, it is always a complex of factors that contribute, and as Guillermo suggested, while we can have a sense of when it is that a country is becoming potentially vulnerable or highly vulnerable to a crisis, there is no way to predict with a high degree of confidence that a crisis will actually come. In fact, there are a couple of countries around the world which continue to surprise me about why they have not had a crisis yet.

Now, I should say there is one theory of what causes crises which I have to reject, and that is The Wall Street Journal's theory, which is that the IMF causes crises.

[Laughter.]

The IMF does not causes crises. The IMF's role is much like that of the Chicago Police. The late, great Mayor Richard J. Daley described the role of the Chicago Police in the great disturbances of 1968 when he said: "Everyone knows the police aren't there to create disorder; they are there to preserve disorder."

[Laughter.]

So I think there are questions that can be raised about how the Fund has responded and the nature of the policies it has suggested for countries that have gotten themselves into difficultly and whether these have been in all instances as helpful as they might have been in an expeditious resolution of the crisis. But the notion that the Fund created the crisis or the conditions for the crisis in the first place is simply outrageous nonsense.

What I think is important to distinguish when we are talking about the causes of crises is to distinguish between those causes where there is something we can control and those causes which are fundamentally beyond our control.

I said an iceberg sank the Titanic, and there is a certain, obvious physical sense in which that is true. However, it is not really very helpful to say that we are going to outlaw icebergs, or at least outlaw icebergs beyond certain Arctic or Antarctic latitudes. That is simply not a solution to the problem. Nor is it reasonable to suggest, as some central bankers do, that icebergs need to contribute constructively to the resolution of the problem. This is just not helpful.

As I noted on several occasions, including the WEO press conference, the most valuable lesson I ever learned about this was when I was about four years old, and my parents red me the story of the three little pigs and the big, bad wolf. It was the little pig who built a brick house who saved himself from the depredations of the huffing and puffing of the big, bad wolf, right? The wolves are out there. Capital markets do sometimes behave in stranger manners; that behavior can be stimulated by moral hazard and by other types of difficulties, but it is probably there to some extent regardless of what you do, and you need to build a solid enough house to be able to withstand most of the storms when they come.

Now, sometimes, there will be storms that no country can withstand. If the United States should once again experience the Great Depression, I don't care what kind of exchange rate regime Canada or Mexico have--they are going to feel some economic impact from that magnitude of external disturbance.

But I think, as Guillermo illustrated in his remarks, there is a good deal that can be done, and I think particularly in the area of strengthening the financial sector, that will not provide invulnerability to financial crisis but will make the pressures that arise in a potential or actual crisis easier to manage and make that crisis significantly less damaging. And the experience in the Asian economies I think has illustrated that lesson very powerfully from the standpoint of what is not a good thing to do.

The examples of other countries that have more successfully withstood the pressures, including Honk Kong and Argentina, I think demonstrate the importance and the value of having strong, well-regulated, well-capitalized financial systems that pay careful attention to limiting and managing their risks in a domestic and international environment in which, necessarily, there is a good deal of risk that needs to be managed.

MR. PERRY: Thank you, Michael.

I'll give the floor to Peter Garber now.

MR. GARBER: In looking for what lessons there are for Latin America from the East Asian crisis, I was struck by the observation, similar to Mike's, that there are probably a lot more lessons from Latin America for Asia than vice versa, because Latin America has been through it in many different ways before. In particular, the 1992 crisis and the 1994-1995 crisis provide a lesson for possible outcomes. One can go into a decade of lost growth, with numerous hyperinflationary results to let the problem burn itself away until it is finally resolved; alternatively, it might be a very sharp, but short, kind of crisis. And there is a large body of experience for Asia in these Latin American episodes.

So I was having difficulty finding what the lesson is for Latin America from this one that hasn't already been learned about soundness of financial institutions, sound macroeconomic policy and so forth.

I did start to consider, though, some of the channels of the problem as they broke out in Asia, and perhaps that may serve once again as a warning for what might be coming at Latin America in the next several years.

In particular, there has been a lot of talk in the Asian context about the effects of moral hazards on channeling capital flows into Asia. The moral hazards cut in on three different levels. One, domestic financial systems have implicitly protected their depositors, and therefore, the local financial institutions are encouraged to take risks. On the second level--and this is perhaps where the IMF causes the trouble--the international financial institutions are willing to provide liquidity support in a crisis, and that heightens the enthusiasm of foreign lenders for bringing money funds into the country. And then, finally, there is a third one which is not so strongly mentioned, and that is that there is a financial safety net in the industrial countries which protects the industrial country banking institution lenders in particular, and that also provided a source and a motivation for bringing funds into Asia.

In particular, the largest sources of funding in Asia came out of Japan which, as we know by now, has a system of "walking dead banks" that are hungry for yields, and as a result, they lent into the Asian markets perhaps for several years longer than they should have. In Korea, we know that the banks were already in trouble prior to this crisis, but they had implicit protection as well, either directly through their governments or indirectly as a strategic country in the eyes of the United States, and therefore, they would receive external funding. So funds went into Korea through the Korean banks and then by way of, either directly or through the new infamous Peregrine, into Southeast Asia. So there is a very strong connection there as well.

The third set of banking institutions that played a major role in Asia were not U.S. banks, but European banks. The European banks as a group were about the same size in Asia as the Japanese institutions, and they took relatively large losses. But the principal banks in Asia were second-tier kinds of banks, banks that really had no business in Asia, no knowledge in Asia; they were just swept up first of all by the "Asia miracle" ideology, and secondly by their own need in a now increasingly competitive European environment to generate profits.

In particular, the Asia crisis exposed that the European banks had, in many cases, very poor risk control internally to the banks, and that brought about a lot of the lending. Some large European institutions, for instance, had branches in Brazil or in Jakarta or in Seoul. The branches in Brazil found that there were Korean banks shopping around to pick up Brazil risk, Brazil credit derivatives, for instance, in markets that they really had no business being in. They were buying these credit derivatives from the local branch of the European institution, and yet that information didn't percolate up to headquarters and then back down to the branches, say, in Korea or in Jakarta, wondering what were these Korean institutions doing.

So there is evidence now that there is quite poor risk control in the European institutions. This comes back now to what lessons there are for Latin America that came out of Asia, and a primary lesson is that you should beware of who your lender is, that there are bad lenders out there that signal problems just as well as bad borrowers, and that soon, Europe, which will unify monetarily, will unleash a lot of bad lenders into the world because of the increased contribution inside Europe. So in the next year or so, there should be a lot of European capital banking funds looking for investment opportunities in the hot markets, which Latin America, having survived this crisis just as Asia survived 1995, will look like a hot market, and funds will flow in in large amounts from bad lenders.

So I would say that this year was not the year for the Latin American crisis; that seems to work on the once-every-six-year model. So you would look for that in the year 2000, and that will be fueled by a lot of willing lenders coming out of the other side of the Atlantic. So wait until next year.

[Laughter.]

MR. PERRY: Thank you, Peter.

Now, Arminio Fraga.

MR. FRAGA: It's a hard act to follow here, but in thinking about what to say here, I tried to move away a little bit from my day-to-day activities as an investor/speculator, and I went back to my lecture notes from Joe and from some other people, and I started looking back not only at the Asian crisis, which is very recent in our memory and still unfolding, actually, and tried to think a little bit about the thirties and the eighties.

When you look at the eighties, for example, and the Latin American debt crisis, we did learn a lesson. The lesson was that borrowing to finance budget deficits is not such a good idea.

Then, looking at Mexico more recently, we learned that borrowing to finance consumption is not that great of an idea, and again, I'm working with my simple, intertemporal model.

And then, looking at Asia, now, we have learned that borrowing to finance investment may not be such a good idea, and then, I say, wait a minute, maybe there was too much investment, but this doesn't seem to be working. And indeed, it wasn't working.

In going back in all of these--and I was either in the markets, or in some cases, I spent a couple of years in Brazil where, in the Government, we were under some pressure in the early nineties--I think the common factor--and this is in Guillermo's text--that is there every, single time is short-term debt. And I will tell you as an investor, the thing that catches my attention is a lot of short-term debt. There are other things as well, but a lot of short-term debt is a definite red flag. You can make a long list of other things, and they all work at some point, and we have to check them all, because in effect, we are looking at a long chain, and we don't know which ring is going to break, but the one that really jumps out is short-term debt.

Other lessons are more general, and I think in this instance have to do with reversals in capital flows, are fairly obvious. In the case of Asia, I think the reversal hit very hard because there was so much domestic leverage, and I think that that is something that needs to be highlighted. I can't find in the history books anything like this except in Japan. In the case of Japan, you have 2 percent interest rates. Elsewhere in Asia, you have 20 percent interest rates, and you have debt-to-GDP ratios of 200-plus percent. It doesn't quite add up. And that is something else that caught our attention.

Interestingly enough, in this case in Asia, as opposed to most previous debt crises that I have been able to find, again, in the history books, there was no fed tightening. There were other things like dollar-yen and so on, but there was no monetary policy tightening at the center, if you want to use the old jargon.

My view in the case of Asia is that the severity of the impact of these capital reversals was due to the excessive domestic leverage, as I said. Sometimes, I even think that the capital reversal itself was due to that leverage. Certainly, what we did was based on that. I mean, we were following very closely the financial sectors of many of the countries in the region, and the action we took was predicated on that as much as anything else.

What it does to policymakers in this country--and that is a lesson to be learned--is that it creates a policy conflict, and markets are always looking for policy conflict situations where governments have their hands tied in one way or another. To use an example that is not one of an emerging economy, you can think of the British pound devaluation of 1992. They certainly seemed at the time to have had their hands tied.

Now, I think it is fair to go and do what Guillermo suggested, which is to look for deeper causes of these short-term flows. The short-term flows were there, but what drove them--and there is a lot of literature on t his; I am not going to spend a lot of time discussing it. I think that these days, anybody who doesn't believe in moral hazard is crazy, although one has got to wonder how important it is. Moral hazard, as the non-economists here know, is a very appealing thing. It sounds like something immoral is happening, and it has nothing to do with that, as you know; it's just a question of incentives, as I learned from Joe, is a very, very bad terminology and gives the wrong impression, but politicians love it, because it looks to be something immoral, and therefore we've got to do something about it. But of course, you have banking practices, markets that were closed, markets where there were no foreign banks, markets that were not supervised and so on.

The things that I think have very little to do with the crisis itself in Asia now and others are transparency in information--I do think there was enough information around; some of it, you kind of had to dig to find, but it was for the most part publicly available. Even, for instance, in the case of Thailand, the intervention in the forward markets, if you were talking to dealers in the market every day, you would know that that was going on and that it was substantial. I don't think it had anything to do with the exchange rate regime. We have seen problems with countries where they were fixed, they were floating, they were crawling, they had bands, they got out of the bands, they moved into bands. I think you have had it all; it is more a matter of consistency. I don't think it was crony capitalism, even though that is obviously not something that we would advocate--and so on.

Moving on to my last five minutes, two things come to mind here. One is what have we learned in terms of crisis management, and the other issue that I think is interesting is what can we say now about what we would like to see. And here, I'm talking about the architecture of the world exchange financial kind of system. I think it's useful to think about these two things. They are different, and sometimes, they contradict one another, or they force you to go in a direction in which you may not want to go.

The crisis management situation reminds me of a story of a former professor and now an important economic authority in my home country who used to tell this little story. Somebody got lost in the countryside in Ireland and ran into some guy who was walking around half drunk, and he asked him, "Could you tell me how to get to Dublin?

And the guy said, "I have no idea, but if I wanted to go to Dublin, I certainly wouldn't start here."

I think the situation we now face in world markets is one which is extremely difficult, and even though we have our dreams of systems where moral hazard is diminished, where everybody who makes a mistake gets punished in the marketplace and so on, the reality is totally different.

It really means we are in a second-best world, and if you are dealing with a crisis, should Latin America run into a crisis--and I agree that it's just a matter of time, maybe five years, maybe 20, but someday, these things happen--in my view, the thing we learn from all these experiences is that the crucial thing is to reestablish credibility, and for that, I think the one thing that seems to be missing in many instances is a set of reasonable goals. I think we need both, and I think they have got to be reasonable, and in many cases, these are lacking. Sometimes both are lacking, sometimes one or the other.

And there is no question that all these countries that are in trouble have got to address their banking problems, their subsidies, and so on, but it isn't obviously true that you need to do the right thing right away; you don't have to shut down every, single bank and so on. The key, ultimately, is macro management, and again, my guess here is that we are going to have to live with the idea that maybe we are going to have a lender of last resort in the IMF and the G-7 countries. And if you are going to have one, you might as well have one that can do things that can be aggressive.

The structural reforms help, and this brings me to my last comment, and this is the second set of issues for Latin America that have to do with long-term architectural, systemic issues. Here, it is clear that better information and more disclosure would help. Nobody can argue with that. It is also clear that better banking supervision, regulation, better incentives, all of that would help. It won't necessarily get us out of this Asian crisis right now, but in Latin America, I think it's something that should be dealt with right now, because there is not a lot of leverage, and it is still early in the credit cycle, so there is time for these things to be done, and in fact, they are being done in many Latin American countries. And you also need improved bankruptcy laws, corporate governance--you know, mom and apple pie and all those things; nobody can be against any of this.

Now, for Latin America, I think the obvious thing, again in my opinion, is that budget deficits, current account deficits, should be handled with care, particularly if they are being financed with short-term money. That is our first lesson.

It is always useful to remember that vulnerability in many ways has to do with rollovers and current account deficits. And banking booms, again, have got to be closely monitored. That is another red flag for us as investors.

I think it would be extremely useful if countries would start disclosing better data on their short-term debt. That would be my first stop in trying to avoid the capital inflow problem that seems to precede all of these crises. I agree with Professor Stiglitz that there is an externality. In fact, I had a little bit of that in my doctoral thesis, back many years ago at Princeton when he was there. And one way to start is to make it explicit, let me people know that the overall debt may be growing too fast and that it may be too much.

In fact, I would suggest that the short-term data should discriminate very carefully between trade debt, bank debt, and so on. I think that that would be a very good thing.

And quite frankly, if all of that doesn't work--let's suppose you have a country where you are doing all of these nice things, and a lot of money is flowing in--one has got to wonder if temporary short-term capital controls are not useful. A lot of times, they become permanent, and there are a lot of problems. Anybody who has worked at the Central Bank of Brazil--I see one of my former colleagues here on the left--knows that this is a dangerous thing. But again, it is hard to argue that that may not be something to be on the menu when times are good.

Finally, for the international financial institutions and the G-7, I think that helping with all these issues would be useful, keeping track of short-term debt, as I said, keeping track of bank flows. Obviously, as you know the Asian crisis was primarily driven by a reversal in bank flows for the most part, and the data is pretty clear on that; maybe keeping an eye on your own banks wouldn't hurt. And I would be strongly against turning the IMF into another rating agency. We don't need another one. And maybe, when things change and we aren't in this second-best world, we should be prepared to let some countries go if they run into trouble and don't do what is recommended or seems to be recommended. But that, I think, is a matter for a future that we may never see, because we seem to be going from one crisis to the other.

Anyway, thank you very much.

MR. PERRY: Thank you, Arminio.

Now, let me ask Joe to close this first round of presentations.

MR. STIGLITZ: I'll speak from here to try to maintain a sense of informality.

I think the discussion so far has highlighted most of the key issues. I think Mike was right in pointing out that we are going to have crises in the future no matter how well we manage the world economy. Even with the best of economic advisors, countries still have business cycles, they still have economic fluctuations, and we are going to have currency and financial crises.

The key issue is how do we make those crises less frequent, and how do we make them less severe. I think that that is the key issue that we want to focus on.

The second point I want to raise is that there has been a lot of discussion of causality, what caused it, what is the root cause. I think the fact is that it is multifaceted, and there are many factors, but most of the explanations that have been put forward, I think are fairly unconvincing, in the sense that if you want to use the word "cause," if "x" is a cause, then countries that have more of "x" are the countries that are going to have the crisis, and if you don't have "x", you don't have the crisis. They are necessary and sufficient in some sense. The world is more complicated than that, but those are the things that you look for as determining factors.

So, if you ask the question, is transparency the cause, you look around and ask where was the last set of financial crises--in Scandinavia, countries that have a good reputation for transparency. So clearly, increased transparency is a good thing, and we ought to do it, but it is obviously not going to be the solution.

Exchange rate imbalances. Clearly, it is bad to have the wrong exchange rate, although how you know it's wrong until after the crisis is often a problem. But if you look at Korea, there is no evidence of substantial changes in the real exchange rate in Korea. So yes, it was part of the problem perhaps in Thailand, but not in Korea, and if we are trying to look for a general theory, we are going to have trouble looking at that.

The same thing in terms of trade deficits. At the time that Korea's problems became really bad in December, they were already running a $2 billion surplus per month. And to think about that as the problem sort of makes you pause. If you need more than that to avoid a problem, then most countries should be very vulnerable.

The financial sector weakness--yes, it is really bad to have weak financial sectors--I don't recommend that to anybody. We have models where we can predict which countries have the weakest financial sectors, and again, we have all seen these kinds of models. Some of the countries that had a crisis were countries which were toward the top of the list, but some of the countries that did not have a crisis were higher on the list. Some of the countries that did not have contagion were very high on the list.

A common characteristic, by the way, of some of the countries for which contagion did not occur and which had very weak financial systems according to our models, which had lack of transparency, had something in common in that they had forms of capital controls that limited the risk. That may be something that is worth noting.

Anyway, one could go through the various purported causes, and it makes one feel very uneasy about most of the explanations on hears. I think that one way of thinking about the issue of how do we make crises less frequent and less severe is to focus on the vocabulary that Guillermo did in his talk, that is, how to reduce vulnerabilities to shock in two senses--how do we make those shocks less frequent, and how do--

[TAPE 1, SIDE B]

MR. STIGLITZ [continuing]: --in some sense, the derivative of the economy to the shock, the real impact of the shock less.

And I think that many of the things that we have been talking about are good policies, because I think that while they are not the cause of the underlying problem, they may reduce the vulnerability in some important ways--so that, for instance, having stronger financial sectors means that when you do have a shock, you are able to withstand it, and it is amplified less throughout the economic system.

High debt-equity ratios in Korea clearly made the economy more vulnerable and meant that if there were a shock, it got transmitted to the economy in a more significant way.

As an aside, let me say that there are lots of policies that are bad for economic growth but have nothing to do with a crisis and have been thrown into the discussion. For instance, trade restrictions and policies that create rents, distortionary policies--really bad things to have. You shouldn't have distortionary agricultural policies in the United States, you shouldn't have distortionary energy policies in the United States.

The question you may want to ask is would it have been appropriate in the middle of the Texas financial crisis to have eliminated energy subsidies. And if you ask that question, the answer is pretty apparent. If you had eliminated the energy subsidies in the middle of that crisis, you would have taken a crisis that was serious and made it a mega-crisis, because in fact, it would have taken away capital values from existing assets and really made them undermine them and weaken the financial system and weaken the remaining corporation.

The same thing in the real estate crisis. Was that a good time to get rid of agricultural subsidies? If you had done that, it would have lowered real estate prices, the lower real estate prices would have exacerbated the real estate crisis, which was the underlying problem in the California crisis.

So thinking about how you sequence reforms, I think is one of the lessons that will be learned from this current crisis, and that one shouldn't think that every policy problem ought to be solved at the same time and that everything is wrong is the cause of the crisis. There are things that are wrong--economies shouldn't have bad economic policies, but many of those economic policies predate the crisis and were not changed and therefore should not be viewed as a source of the crisis.

Let me go on to the main theme of what this discussion is, which is the implications of the crisis for Latin America, and I want to do that in two parts. One is, looking forward, what does the Asia crisis imply for the economies of Latin America. And perhaps the best place to focus on is Brazil, because it was a country that was most strongly affected by contagion.

An interesting thing is that it appears now that the financial consequences of the crisis appear to be transitory. If you look at things like stock markets and capital flows, there were big impacts, but those have been largely reversed. Spreads, for instance, on loans jumped from 186 basis points in October to 527 basis points in November and have now fallen back to 215 basis points. There is a lot of volatility in markets. That is one of the things we recognize. The stock market was down 40 percent between October 1st and November 12th; since then, it has risen 50 percent, almost recovering all of the ground lost.

So there is a lot of volatility, but the financial market effects appear to be transitory. What is of concern is that the real effects could be more persistent, and that was sort of shown in the data that Guillermo illustrated--that it looks as if the real effects on the economy, even after the transitory financial effects are gone, partly because of policy responses, may lead to lower growth and higher unemployment, and those effects are going to be far more significant than the temporary losses of book value on the part of some investors; those effects are likely to be more severe throughout the economy.

In terms of the lessons, it seems to me there are two sets of lessons that I want to call attention to. The first picks up on a theme that has been mentioned several times, which is that there are risks, and the question is what are the ways that we have of trying to limit those risks; what can we do to limit risks in the economy.

And actually, there is some research that gives us some insights into that, and that research relates to studies that talk about what are the factors that affect the probability, for instance, of a financial crisis. We know, for instance, that financial market liberalization and opening up capital accounts in, by definition, "an inappropriate way"--that's almost a tautology, but it turns out mostly that it is done in an inappropriate way--has a high likelihood of increasing risk.

There has also been discussion of how foreign exchange exposure can increase risk, and that trying to take actions through regulatory means or others to limit exposure can reduce the risk.

The key questions are what are the trade-offs, and how do we view these things from a public policy perspective. In terms of the second question--how do we view it from a public policy perspective--I think Arminio's point is exactly right. There are externalities involved. The fact that there are public actions, like bailouts, is indicative that either there are strong externalities, or there is a belief by others that there are externality systems at risk that lead to actions which then, in certain terms, lead to a marked discrepancy between social and private risks and risk-taking.

So, if there were discrepancies between private and social risks, there is a kind of market failure, there is a kind of distortion, and that distortion suggests at least to open the possibility of asking, in terms of the design of our international architecture, on the part of lenders, borrowers and the international community, if there are ways of correcting that distortion.

So in that case, limiting risk--there isn't a trade-off. It is actually a case where, by correcting the externality, you are increasing efficiency and limiting risk.

The second issue is raised by the question of what are the growth effects associated with short-term capital flows. There has been a lot of discussion today about how short-term capital flows are related to the risks. Then the question is what are the benefits--we now know there are risks, but what are the trade-offs, what are the benefits?

Well, one way of thinking about the benefits of that, or the net benefits, is to think of the question, if a country is already saving 35 percent of GDP, and it borrows a little bit more--36 percent, another percentage point--what is the margin of return in terms of economic growth, and how do we weigh that extra return versus the extra risk that it exposes?

Now, there are some things that are important where, engaging in these flows, if done in the right way, you can get more portfolio diversification and actually reduce risk, but that is for certain types of equity flows and isn't generally the case for a lot of the short-term debt flows.

The other related way of seeing this is that some people have recommended in the kinds of statistics that keep coming up are issues like what is the reserves-to-short-term debt ratio. People think that if your short-term debt exposure goes up, then you ought to have more reserves.

Take the limiting case, just as a way of thinking about it. Let's say that you think you ought to have reserves to cover your short-term debt, which is a view that some people have put forward. Well, that would mean that if a poor country borrows $1 billion from some American or German bank and is paying 10, 12, 15 percent, the government then is supposed to take an equivalent amount of money and lend it to the United States Treasury or the German and get back 4 or 5 percent.

Now, how this is supposed to be good for economic growth is a little bit of a mystery. Borrowing at 15 and lending at 5, you lose 10 percent. I can understand why some of the more advanced countries like this as a policy, but why the less-developed countries want to engage in this kind of transaction from the macroeconomic point of view leaves me a little befuddled.

The net effect of this is that there have been some studies that have tried to look at relationships between capital market liberalization and economic growth in a cross-country way. Obviously, every cross-country study is controversial, and particularly those who don't like the results always say they are particularly controversial.

Danny Roderick has recently done an interesting paper--I find it interesting--and what he shows is that basically, there is no relationship between capital market liberalization and economic growth, that it does not result in faster growth. That is quite different from the wide body of results that show that trade liberalization is systematically related to economic growth. There is a large body of literature, basically the same kind of cross-country methodology, which substantiates the view that trade liberalization does promote economic growth. The results on capital market liberalization, particularly for short-term capital, are clearly less convincing.

The final set of lessons, I think, that Latin America will want to draw have to do with how to manage--or not manage-- crises, particularly when they are private sector crises. People have talked about the enormous experience that Latin America has had with crises and managing crises, but one of the things that has been repeatedly emphasized is that this is a private sector crisis, that private sector borrowing in most of the countries is the root cause, or is intimately involved. And how to manage that kind of crisis, there may be some important lessons.

Some of the lessons that Latin America learned earlier may also spill over. I don't want to think of these as two separate sets of ideas, but I think the distinctive nature of it being a private sector crisis raises some interesting issues, particularly as the private sector is becoming more and more important in Latin America.

Let me illustrate a couple of examples of the kinds of questions that I think are going to be raised. One, which is the question that was actually raised by the earlier experiences, is are there ways to minimize the real effects of how the crisis is managed. The "lost decade" that one often talks about after the crisis in the eighties is a high price for an economy to pay, and if that's the price we have to pay for every crisis, then we really ought to think about risk mitigation. So if that's the beset we can do, I think we really need to think very deeply about kinds of risk exposure.

Those same questions about how to minimize the real effects are now being raised in other countries in East Asia. As the unemployment has begun to soar in Korea, for instance--is has already more than doubled--people are beginning to worry about ways to minimize these real effects.

I don't think we have the answers yet, but clearly, the questions are becoming faced more visibly.

Just briefly, three other aspects of it that I think Latin American will have to think about in terms of lessons. The first is the relationship between politics and economics. You cannot restore confidence in an economy when there is political upheaval, and there are systematic relationships between political upheaval and economic weakness.

When unemployment rates soar, there is going to be political upheaval. So that while economists would like to focus more narrowly on economic issues, you cannot really separate the politics from the economics.

The second one is actually related, and we saw it in Mexico, and we saw it in Indonesia. Capital flight is important. When you are looking especially in the private sector, do not just focus on investors in New York; look at investors in their own country and what will affect their behavior. Their information set and their risk stance are markedly different from those in New York or London, and therefore, one has got to focus on that.

Thirdly, think about the safety net. In East Asia, it is very clear that people who were not involved, were not borrowing--innocent bystanders, as we sometimes say--have been very adversely affected. The bad lenders that have been talked about, the guys who pushed the money onto these countries, as somebody put it, have been largely protected. They have lost something, but they have been largely protected. A lot of the workers who were thrown out of jobs have seen a much larger cut in their income.

And the same thing goes for small businesses that were not borrowing abroad; they are facing credit crunches and much higher cost of credit. They were not engaged necessarily in benefitting from the capital flows that were talked about earlier.

So the general issue and one of the lessons that I think Latin America will need to think about more is how to construct a safety net so the distribution cost impacts of dealing with the crisis are more equitably shared within the society.

Thank you.

MR. PERRY: Thank you, Joe.

Before opening it up to questions and discussion, I would like to give a short period to Charles Dallara. Unfortunately, this morning, something came up, and Charles called me to tell me he was going to be late.

So, because of the time, Charles, I would like you to make a short comment, and then we will open it up to the public.

Charles Dallara, from the International Institute of Finance.

MR. DALLARA: Thank you very much, Guillermo.

I do apologize for my tardy arrival. I am sure that most of the important things have already been said by my distinguished colleagues on the panel, and therefore, I will stress only one or two points and then move on, as I am sure many are anxious to get to a period of questions and answers.

Let me pick up on two points which Joe Stiglitz just mentioned. The first one relates to the behavior of market participants. It seems to me that one of the key lessons for Latin policymakers and indeed, emerging market policymakers around the world, in light of the still evolving crises in a few Asian economies, is the need for much better skills at reading the tea leaves of markets.

It is fairly easy for us all to agree that policymakers must act at an early stage, macroeconomically and structurally, to correct the emerging imbalances in their economy. It is not always easy, of course, to do that.

But the most difficult part, I think, is that many times, policymakers have to mobilize support for taking difficult policy actions even when the markets are continuing to send them positive signals of ongoing support through access to capital markets, oftentimes even access to capital markets at declining costs, even in the face of gradual and noticeable deterioration of imbalances.

It is not easy, of course, to mobilize support to renew the fight against a fiscal imbalance or an external imbalance at any time. It is particularly difficult if you have to do so in the absence of a crisis at home and in the absence of strong pressures from the marketplace. If the markets are sending you clear signals that we no longer have confidence in your policies, the environment for mobilizing support for strong actions, as we have seen in Asia in the last few months, becomes much more politically manageable.

But the fact is that during much of 1996, even in the face of growing evidence of emerging imbalances in a number of East Asian economies, the spreads on lending to some of these economies continued to narrow; the cost of borrowing for them was continuing to decline. At some point, of course, when the market confidence breaks, it can be quite strong, quite impressive, quite powerful, and even quite devastating. And the task of rebuilding that market confidence is enormously complicated and challenging, as I think we have seen.

It seems to me that that task for Latin policymakers today is to avoid being lulled into the notion that simply because the markets are continuing to fund them, that all is well at home, that the markets will forever tolerate the imbalances, that just because Argentina's current account deficit is growing dramatically, as long as the markets are continuing to comfortably finance it, it should not be a source of concern. I am not trying to single out Argentina, because we have a number of countries in Latin America where imbalances are growing, but I wanted to stress this point, that one as a policymaker must make a much more considered effort to try to understand the complicated signals that markets are really sending and not to take false comfort from the fact that market access may continue even at declining spreads.

I have often suggested to some of my friends in the policy community in Latin America that when they go to New York, they not meet with the senior investment bankers or even the senior commercial bankers, but that they meet with the most skeptical analysts and asset managers that they can find, because there, they may get more of the seeds of the future doubts that may grow in the markets if problems continue.

The second point I would stress, and then I will conclude, has to do with market transparency. It seems to me that it is increasingly evident that transparency is the ally, indeed, almost the weapon, of the policymaker today. And those countries which are prepared to do anything possible to release the full content of their IMF programs where there are programs in place, to release the content of their Article IV consultations, to communicate macroeconomic and financial data in the most timely fashion, are going to be best-positioned to sustain and enhance market confidence. I think the lessons are very clear and quite powerful here.

There are obviously many other issues--financial sector, macroeconomic, exchange rate--I am sure they have already been discussed, and I will not dwell on them further at this point but will turn it back to the chairman.

MR. PERRY: Thank you, Charles.

Now, we will open it up for questions. Please identify yourself and say to which panelist the question is addressed.

MR. ____________: Thank you. I would like to make a comment and then a question.

I couldn't agree more with Mr. Stiglitz on the very valuable point that one thing is to have, let's say, bad features in an economy, and a very different thing is to single them out as causes of a crisis. I think that that confusion has been very much in the discussion of this crisis so far.

I agree with Mike Mussa that there are always a lot of causes in a crisis, but I think we can hardly argue that--it is bad not to know how to swim, but had that been different, it would have prevented the problem of the Titanic.

So, following that line of argument, one thing that has been stressed very much as one of the causes of the crisis is the exchange rate regime. One can argue that maybe, if you can afford an independent monetary policy and if you have a well-established central bank independence, emerging markets and countries in our region in particular may consider floating their exchange rates. But I think we can hardly argue, as Mr. Fraga highlighted and others, that the core issue of the crisis was to have fixed exchange regimes.

In fact, there are a number of experiences, including my own country back in the eighties, where floating exchange rate regimes have not prevented a big, big macro financial crisis and haven't prevented capital inflows from coming in.

Based on that point, I don't think that more flexible exchange rate regimes are going to prevent--in the next wave of exuberance--flows from going to emerging markets that most likely, as convergence theory predicts, will again to grow more, presumably have higher interest rates in an attempt of the authorities to prevent overheating, and so on and so forth.

We have signed out short-term debt and other stock indices as very important things in this crisis.

Finally, I guess my question is what can we do about that. As you know, I am a Chilean, and you know of our particular experience, and I would like to hear the distinguished panelists on this.

MR. PERRY: I am sorry, I'm going to have to ask all of you to make your questions very short.

MR. _____________: Okay. I'm sorry. I guess my question is on top of hearing about open positions of banks, what can be done in order to prevent the natural tendency that will have agents, like corporations or households, elsewhere having open positions when countries are in abundant liquidity.

Thank you.

MR. PERRY: Thank you.

MR. STIGLITZ: I think there are at least three sets of policies that one can think about. The first thing, which I think everybody would probably agree on, is to correct the distortions that led to excessive short-term borrowing. For instance, in Thailand, a number of these countries actually had policies that were distortive in the sense that they actually exacerbated the problem. So I think the nature of markets could lead on their own to the problem, but they have policies that exacerbate it.

Secondly, in the case of the banking system, the financial system, I think there needs to be better regulation. The real problem is not too regulation, it is not the right kind of regulation in the financial system, and there are many dimensions of that.

For instance, if you had well-designed risk-adjusted capital adequacy standards, you would have had better incentives, and interest rates would have reflected better the risks associated with lending to firms that were highly-indebted, had very high debt-equity ratios, or firms that had large capital exposures.

To be fair, no government has really done a very good job of designing risk-based capital adequacy standards. Even in the United States, they have focused on credit risk and not on capital asset value risk associated with interest rate changes. Even when it was pointed out to the regulators that this was not the right way of managing risk, there were reasons they had of nontransparency, basically, I think, for not wanting to have the right basis of risk adjustment.

I think one needs to have regulations that address the issues of speculative real estate booms, breaks on the system, speed limits on expansion of credit, because all of these are things that investors recognize as signs of problems, and the regulators ought to as well.

The third and I think most controversial set of issues have to do with versions of capital controls or taxes that address the externality problem that I talked about before. I don't think one should view this as creating a distortion. The fact is that there are externalities, and you want to correct those externalities.

There are issues of the feasibility of those, practical issues, and they would differ from country to country. One that a lot of people are looking at is the Chilean system. Another one is provisions that affect tax deductibility of short-term debt of corporations denominated in foreign currency for those countries that have corporate income taxes.

I think there are a number of proposals of this kind that one can start looking at that will help stabilize or mitigate the risks.

MR. PERRY: Michael, would you like to say something on this?

MR. MUSSA: Yes, I agree with most of what Joe had to say on this issue. I think it is important to recognize that developing and emerging market countries and their businesses and banks typically do not enjoy the luxury that businesses and banks in a major industrial country do--be able to do all of their borrowing for domestic business purposes in domestic currency and, if they have business located abroad, to fund that in terms of foreign currency.

So of necessity, many emerging market countries are going to access world capital markets, and they are going to access them in terms of foreign currency, and there is going to be a risk exposure there that most industrial countries don't need to worry about.

Laying that aside for a moment, I think it is telling, despite all the deficiencies of regulation and risk management of financial institutions in the United States and other industrial countries, that they are not exposed to the type of foreign exchange risk that has typically characterized many developing countries.

I like to point out that in Thailand, the financial system had $35 billion of net dollar exposure. That was more than one-sixth of GDP. For the U.S. economy, that would be more than $1.5 trillion, or 3 trillion DM. Now, if the U.S. banking system had net deutschemark, short exposure, of 3 trillion deutschemark, neither Alan Greenspan nor Hans Teitmeyer would be able to get any sleep at all.

The fact of the matter is, though, that the pressure on banks from their own risk management is such that they do not expose themselves to that magnitude of risk where a 10 percent movement in the exchange rate would wipe out their equity capital. And it is important in terms of the incentives that are provided to the recognition and management of risk in emerging market countries where, of necessity, there is going to be a fair bit more of it that is going to need to be carried by both businesses and banks, that they recognize and manage that risk.

There, I disagree a little bit with the premise of the question about the exchange rate regime. I think we need to be really careful about the exchange rate regime, and I think it has been an important because of some of these crises--and here, I disagree a little bit with what Joe said earlier.

It is not the case that everyone who smokes gets lung cancer, and it is not the case that everyone who gets lung cancer smokes; but that does not mean that smoking is an unimportant factor, causative factor in lung cancer. And the issue of the exchange rate regime and how it interacts with other macro and microeconomic policies is, I think, an important issue about which one needs to worry.

If you are going to have a floating exchange rate regime, I don't believe most emerging markets are going to float the way the dollar/deutschemark exchange rate floats, where we have gone from 135 to 185 over the course of two years and with a monthly standard deviation of 4 percent a month. That is simply not realistic for most emerging market countries. But if you have a floating exchange rate regime, it is important that the exchange rate float up and down on a fairly regular basis so that people can see that the market influences the exchange rate, enterprises and banks can recognize that there really is foreign exchange risk to which they are exposed every day that they are in business, and to have an accounting and other systems that force the recognition of that risk in terms of the bottom line of those enterprises, because in the end, the controlling of those risks needs to occur in the enterprises and in the banks, that if the controlling of the risk depends upon the strength of the supervisory system as the first line of defense, you are a dead duck. So you have got to find some way to make the enterprises and the banks recognize it and manage it properly.

The other alternative, of course, is to have a really hard peg, which means having policies that are consistent with maintaining that peg.

There is, however, a problem with really hard pegs. Even really hard pegs can get blown away. Nothing--nothing--would have sustained the peg of the Chilean peso in the early 1980s. It was just inconceivable, given the magnitude of the disturbance to the world economy, that that exchange rate peg could have been sustained. I think Argentina has been very well-served by the convertibility plan. If, in 1995, we had seen the dollar strengthen by 50 percent against the yen rather than weaken, and if we had seen U.S. interest rates go up by 4 or 5 percentage points as the Federal Reserve tightened to combat a rise in U.S. inflation, this would have been a very different story.

So, while I don't want to say, and it certainly is not the case, that the exchange rate regime was the dominant factor in all of these instances, I think that it is an issue that requires continuing attention in the management of economic policy and in the communication that needs to occur between policy in the business and financial sector about how it is they manage their risk.

MR. PERRY: Let me say something here, first, that I concur completely wit this view. And let me suggest that it is precisely those countries that are not prepared either through a very strong commitment to a peg, like in the case of Argentina, or to a pure float, which I don't know too many countries yet that are prepared for that, are precisely the ones that care or may care about the possibility of using some sort of capital controls, because they are giving some degree of guarantee of exchange rate stability that would have the effect of having the private sector take excessive currency risks. And then you would probably like to compensate that somehow.

Peter?

MR. GARBER: Along that line, we have had a lot of suggestions of possibilities of controls of one form or another, more or less stringent, plus we have had an argument that transparency is a good thing, and I'd like to argue that these are mutually exclusive reforms. If you impose controls, unless they are stringent and pervasive, they will be evaded through subterfuges and methods that preclude understanding exactly what an off-balance sheet of the financial sector is, so that now transparency will exist in that kind of situation. So you want one or the other, but you can't have both.

MR. ________________: Thank you. I think that probably the best persons to address this question are Dr. Stiglitz and, I think, Dr. Mussa.

First of all, I was hoping this evening to take away with me some normative principles that would apply to the management of small countries like my own. I've gotten hints of them, but I haven't really gotten them. So I want to put out some hypotheses very quickly.

The first hypothesis is that what protects countries from movements, from disturbances, in the international financial system are, one, deep and extensive internal markets, highly-developed markets, and two, extensive and highly-developed foreign exchange markets. This is the mechanism which helps the country to absorb rather than be overtaken by these disturbances. That is the way the United States, for example, and even countries in Europe, like France, the UK, Sweden and so on, are able to do so.

So my hypothesis is that a country like Barbados, with 270,000 people, will not have these markets by definition, specially since the Barbadean currency is not held abroad; so we will not have those markets, and therefore, we must use nonmarket means of protecting ourselves from these possibilities. So one means is capital control, which I notice is creeping in, but which was very, very--I mean the free marketeers a year ago would have been absolutely upset by your even hinting at capital controls. So, capital controls.

The second means of defense, I would suggest, is that you have to hold foreign exchange reserves even though there is a cost of so doing--but it is a cost of stability that you have got to pay. It is observable, although none of you observed it, that those Asian countries which survived were holding high levels of foreign exchange--Singapore, with 3 million people, holding $75 billion of foreign exchange; Taiwan, $85 billion; Hong Kong, $80-odd billion. China is using capital controls and also has relatively high reserves, $125 billion, so they hardly know that this thing has happened, more or less. So I want to put that out.

Now, as far as exchange risks is concerned, I just want to observe for the benefit of Dr. Mussa that fluctuations in a country--I'm going to use the example of Jamaica, Barbados and Guyana from my own region--does not mean moving 50 percent in six months. It means moving, in the case of Guyana, from 5 to 1 to 140 to 1, which means, as somebody mentioned, destruction of capital assets and all kinds of difficulties from which it is very hard to recover.

I just wanted to suggest those hypotheses so you. Thank you very much.

MR. STIGLITZ: I think the point you focused on is the point that actually, I think we have raised a couple of times already, which is that small economies are going to face more risk. It's like a small company, with very thin markets, and very thin markets are going to be more volatile. Small countries are going to be less risk-diversified, and therefore, it is more likely that there will be one commodity which they are going to be dependent on. A shock to the price of that commodity is going to have macroeconomic effects. There is a whole variety of reasons that smaller economies are going to face more risk.

The implication of that is that given that they have a different degree of risk, they have to have a different set of policies. When I say "set of policies," what I mean is that it may be the same framework, but the levels at which they dial the various instruments will be different. And there are two sets of policies that I think we want to think about. One is what you might call the precautionary policies--and you emphasized one--like reserves. The optimal level of reserves is going to be different when you face a higher level of risk; I think you are perfectly right.

The second one is the appropriate set of regulatory policies, the appropriate way of designing the banking systems, are going to have to be different.

And the third one, which I think a lot of people have been talking about, is that the types of prudential capital controls may have to be different, precisely because of the nature of the risks which they face.

MR. PERRY: Michael?

MR. MUSSA: When we are talking about a small, open economy like Barbados, my view is that this is an outstanding candidate to have a currency board-type regime, an absolutely rigid peg of the exchange rate. I think the exchange rate has been pegged at, what, 2 to the U.S. dollar, for 25 years. I think that that is the right answer for Barbados.

There is a question of how one handles the banking and financial system, and there, I think the New Zealand example is a relative one to consider, where they have invited a lot of foreign banks, largely Australian, but also others, to run much of the domestic banking system. They have got, then, the capacity to diversify whatever are the credit risks beyond Barbados to other parts of their operation. And there is always American imperialism, but Canadian banks can do the job, denominated in U.S. dollars, if that's what the issue is.

I think it's important to recognize that this type of regime obviously does not insulate Barbados from a variety of real disturbances. If a massive hurricane hit Barbados, there is nothing that its monetary policy can do about that event; or, if suddenly, Castro disappears, and all the tourists go to Cuba, their monetary policy isn't going to help very much in dealing with that disturbance, either.

But it seems to me that that is the best option for an economy like Barbados in terms of its monetary exchange rate and banking policy.

Jamaica is a somewhat larger and more diversified economy, but looking at the economic history of Jamaica in the post-War period, it is hard to come to any other conclusion than the conclusion that Argentina came to, that they would be better off without any independent monetary policy, because their policy record has really been a disaster over a long period of time, and I think they, too, would be better-served, even though sometimes they are a less obvious candidate for it in terms of the size of the economy.

MR. DALLARA: I will only add briefly two words, because I don't disagree with anything that Joe or Michael have said, but it does seem to me that in the environment you face in Barbados, and in similarly small, open economies, that there still is a strong case to do what you can to develop and deepen your local capital market, if not for the scale that will make a difference globally, for a scale that will make a difference in terms of the domestic borrowing and capital-raising requirements.

Furthermore, I would underscore an issue that has lurked around this discussion but has not been stressed, and I think it is relevant for the small open economy as well as the larger, perhaps relatively closed one, and it is that the capacity of the authorities to influence the mix of short-term debt, medium-term debt and equity, both portfolio and foreign direct investment, is not insignificant.

I mean, I look at 1996, and there are some very interesting few numbers that I see that contrast Latin America with Asia. Both the five Asian economies which have gotten into serious trouble, or at least for periods of time were in crisis or near crisis over the past nine months, and all of Latin America attracted almost an identical amount of private capital in 1996--roughly $90 billion--and over half of that in Latin America was foreign direct investment, 50 out of 90. Seven out of 90 was foreign direct investment in these five countries in crisis.

That underscores to me the important ballast that can be given to an economy, small or large, particularly small, by that flow of direct investment and the flow of medium to longer-term debt, which is obviously in some cases and perhaps in your own infrastructure-related. It strikes me that one area that we don't debate in the context of short-term capital flows at all is the need for a more attractive regulatory environment for infrastructure finance, but I think it is a key issue of lengthening the maturity structure of the debt flows, and I think it can make a real difference--in just a few big infrastructure deals, you can make a real difference in the overall average term of the lending that comes into a small or large economy, and tackling those regulatory issues that are necessary to attract the infrastructure environment is--

[TAPE 2, SIDE A]

MR. PERRY: We are going to collect several questions and put them to the panel. So I will collect three or four questions from the audience.

Antonio Morales?

MR. MORALES: Antonio Morales, President of the Central Bank of Bolivia.

My question is addressed to Mr. Dallara and probably to Mr. Garber. Since this is related to the financial system, what role, if any, do the credit rating agencies have in adding to transparency, or have they been counterproductive, adding more noise than transparency to the situation in Asia?

Secondly--and this is related to a comment made by Mr. Garber--if we look at the credit ratings by the credit agencies, the European banks fare a lot better than the American banks. How do you reconcile that with your comments that we heard before?

MR. PERRY: Peter, do you want to respond to that very specific question before we collect other questions?

MR. GARBER: Well, I think the record of the credit rating agencies in Asia has been a well-documented, dismal one, that they seem to be late each time, at each stage of the process.

In terms of credit rating, it depends on which banks you are talking about. Many of the banks in Europe are government-owned, so they would get a good rating, but they would be dismally run. There is also going to be a major change in the competitive environment in Europe which may not be built into the ratings, and I am looking forward to that.

MR. DALLARA: Just a word on the ratings and the role of the rating agencies, because I think it is a real challenge for a country such as Bolivia, as well as larger countries which are tapping the capital markets these days.

It seems to me that the important issue to stress, both on the issuing side as well as for the buyers of the paper and the underwriters and the lenders, is to stress the need to treat the ratings as only one relatively modest component of the assessment of creditworthiness. And I think that that has been a large part of the problem, that there has been much heavier reliance on ratings than has been warranted, and it has become for too many investors--and this includes many institutional investors who are relatively new still to the emerging markets game, large, sophisticated institutional investors in the United States who know U.S. markets in and out--they really don't have a lot of "bench strength," as one might say, internally in their organizations with regard to assessing emerging markets' risk, so they rely heavily on what underwriting firms tell them and on what rating agencies tell them, and I think therein lies part of the problem. It becomes too easy to become dependent upon the ratings. I certainly think the rating agencies are due for criticism, but I actually think that there is a more fundamental systemic problem of undue reliance on ratings by issuers, by lenders and by those buying the paper and the underwritings, and I think it speaks to the need, really back on the private side of this, for better independent risk management, which is not unduly reliant upon ratings.

MR. PERRY: Thank you.

Yes, please.

MS. BERTHELOT: Danielle Berthelot from the [inaudible] PSI Network in the Bank in LAC.

I have a question for each panelist. I would like to know the view of each panelist on the reforms taken by the Brazil Government in November 1997 to mitigate the spillover effect of Asia.

I also would like to know from each panelist if they had been in the position of the Government in 1997, what they would have done.

MR. PERRY: I will collect a couple of other questions, to give the panelists a moment to think about what they would have done.

Yes, please.

MR. _____________: I have a very brief one. I am Robert Chureta [phonetic], with The International Investor.

I would like to ask particularly Mr. Fraga, in light of what you said about a dangerous error of financing investment, what is your assessment of China presently--and anyone else who cares to answer that.

MR. PERRY: I will ask the panelists to take the opportunity of these questions that were directed to all the panelists to answer them and also give any final comment that they would like to give. So we will end this turn in the reverse order that we began.

Joe Stiglitz can go first.

MR. STIGLITZ: Well, China has weathered the storm in many ways better than many other countries in the region. People are anticipating a slight slowdown in their economic growth, but from very high levels and a very slight slowdown.

My sense is that the Government is very carefully recognizing the adverse effects that regional growth will have on China's growth and taking that into account in terms of their necessary adjustments to macroeconomic policies to sustain their economy.

There has been a lot of discussion of problems within their financial sector. I think it is important--and this is not the place to go through all those issues--but it is important to keep in mind that there is a very fundamental difference between a private financial sector and a public financial sector--and this is actually something that came out before--that there may be problems in public financial sectors misallocating resources because they don't have the right incentives, and that is reflected in the low growth that China has experienced over the last 20 years--it has only been compounded at 10 percent or 12 percent a year--and if they had had a private banking system, they clearly would have grown at 20 percent a year.

[Laughter.]

But the fundamental problem that one worries in a state banking system is misallocation of resources. Recapitalizing a state banking system does not get involved in many of the same kinds of issues that are involved in recapitalizing a private banking system. In thinking about the problems that China faces, I think one has got to keep in mind this very big difference in the nature of their banking system. It is important for any country to make sure that funds are invested well. And if it turns out that as a result of the banking system, funds are not being invested well, then it becomes something they need to deal with very quickly.

In the case of Brazil, again, I don't want to go into the details, but I just want to reiterate the fact or the concern, obviously, that what has been a transitory financial disturbance is now being translated into real effects that are more persistent, and that obviously is going to be an issue of concern.

MR. PERRY: Charles?

MR. DALLARA: Just very briefly on China, I do think that we are witnessing one of the most ambitious attempts at structural reform in the parastatal and financial sector that I have seen in emerging markets in many years, if ever, and even if the Chinese authorities succeed in only half of what they are set out to accomplish in the next 36 months, I think it will have been a great deal.

However, I do think that the clock is ticking on the credibility of the financial system in China, and I think the authorities know this. I think the markets have been really willing to bend over backwards and finance a very opaque system in China for years, because they are willing to bet on the future. But at some point, I think the market requirements, the market standards for transparency and for credibility in the banking system, are going to jump up, and if they jump up too high, too quickly on the Chinese authorities, it could create some serious problems. I don't know when that will happen, but I think they are very wise to move quite aggressively.

I do think it is one thing to try to clean up the balance sheet, which is going to be a huge problem; it is another thing to move toward privatization. Both are necessary, I think, in the next five years in China.

On Brazil, I think it is very difficult to second-guess. It is tempting to put oneself in Pedro Milan's [phonetic] shoes for five minutes and then step back out of them and let him face all the real challenges. But the fact is that I think Brazil did what it needed to do quite effectively in November. It seized the moment of the prospect of a crisis rolling across the Pacific and both substantially tightened monetary policy and made a significant move on fiscal and, since then, has taken a number of useful steps.

I do think, though, looking out over the next number of months, that we are embarked on another quite sensitive time for Brazil, especially in light of the significant--understandable--but somewhat risky reductions in interest rates that have taken place in the remaining months, and whether there is going to be a need for some further fiscal actin in the pre-election period and the ability politically to pull it off, I think is a big question.

MR. PERRY: Arminio?

MR. FRAGA: I am glad nobody is asking about the dollar or Japan. Starting with China and maybe continuing where the other panelists left off--and maybe here, I speak with a little trepidation, because I am not really an expert there--the numbers on the credit side--I think the reason they haven't been affected by all this turbulence is because they have no short-term debt, no current account deficit, a lot of reserves. They are just somewhat isolated in that respect.

Trying to look ahead a little bit, I recently went through the very good and detailed report by the IMF published in September 1997, "Recent Economic Developments," and checked the data, and much to my surprise, I found that loans to GDP in China have been stable at around 90 percent for the last three years, and they really do not have any public sector debt.

I have changed the way I think about China a little bit because of that. They do have the growth in savings. And if the strategy is to grow their way out of this problem, I think it seems that this is feasible.

I now think of China--because I think a lot of these loans are never going to be repaid--I think of China as a country that has 40 percent government debt-to-GDP ratio and 50 percent loan-to-GDP. And that is okay. I mean, it's not something that scares me too much, particularly when you save as much as they do. And if it is true that they are going to start gradually reducing the amount of directed credit, and the policies are going to be moving toward more of a free market banking model--with all the dangers that that brings, of course--then I think there is a landing strip for China, and we'll have to see. But I am actually pretty optimistic. I am probably more optimistic than the average report that comes through my desk these days, and we probably get about two a day.

On Brazil, I think the events of the third quarter of last year caught Brazil at a somewhat vulnerable moment. Brazil has done a tremendous job in a lot of the structural issues that we have been discussing here--privatization, banking reform, et cetera--but has been taking a very gradual approach to its twin deficits. It was slowly reducing its public sector deficit and slowly addressing the current account needs with this gradual crawl of the exchange rate, and it was caught.

I think the response was correct. It was very aggressive, thanks to two things. One is thanks to the fact that the banking sector was no longer a problem, with a very, very low loan-to-GDP ratio--in Brazil, that's in the low 20s, so they were able to tighten without worrying about bankrupting the economy, which was not the case in Asia--and thanks to the fact that they are privatizing, and they have a lot to privatize, and most of it is ahead, so there was a bridge in terms of financing available for the next four or five years.

As a result of the crisis, I think it also clearly dawned on them that they couldn't be too gradual on the fiscal front, and since then, some crucial constitutional change have been either voted or are in the process of being voted, including civil service reform and the first step in a two-stage social security reform.

So that also in the case of Brazil, I see a path that seems feasible and that is being managed prudently, particularly after the shock of the third quarter.

MR. PERRY: Thank you.

Peter, and then Michael will close.

MR. GARBER: I think it's clear from what the previous panels have said that China can't be run off of its current exchange rate, either because of controls or because of the relatively low short-term debt. So if it does devalue, it would be because of competitiveness reasons with the rest of Asia, but even in that dimension, export surges haven't really hit yet, so that is something that will be coming in the next quarter or next two quarters. So it is in the air what will happen, even if they feel pressure whether they will feel the regional responsibility of not devaluing themselves.

So, rather than explore the probabilities of their devaluing, I'd like to explore the connection of China devaluation to the topic at hand, which is Latin America.

So far, the Asia devaluations haven't really affected the export markets of Latin America, and Guillermo pointed out the evidence for that. There is very little overlap in third countries; there is not that much direct trade, either. And even if there were overlap, the export surges, again, have not occurred. I know that Mexico, because of the automotive industry, is worried about a Korean export surge when that finally happens, if the Koreans decide to cut their prices in a serious way.

So there seems to be little worry about the trade effects so far of the Asia crisis on Latin America, but China is a different story. In Mexico, my friends have calculated that there is a 20 percent overlap between China and Mexican exports in third markets, so that would have a very serious effect on Mexico if there were a devaluation in China, and also, it might have an even greater impact on Brazil in terms of--I'll skip the Brazilian reforms.

MR. PERRY: Michael, I'll ask you to close now.

MR. MUSSA: I think the two countries are very interesting and very interesting to compare and contrast with other countries. Brazil did what it needed to do to sustain the Real Plan [phonetic] in the face of a sudden change in the international financing environment. I think that if they had just sat there and done nothing, the market would have blown them away. That is not an absolute certainty, but I believe that without affirming a first monetary policy and fiscal policy, the market was after them, and the exchange rate would have been very seriously attacked.

The consequence, of course, is a slowing of growth of the economy and, even more so, of domestic demand, but one needs to ask what the alternative would have been if the response to the slowing was to ease monetary policy and fiscal policy, and I think we would then see something between Korea and Indonesia.

So I think one does what one needs to do, and the credibility of undertaking that on their own I think was very important in terms of the favorable market reaction to it.

They have now been able to ease back on monetary policy, and I think that that is a useful development. It cannot be pushed too far too fast, and there needs to be a readiness to tighten again if the pressures return, but I think it is prudent when the pressures have been relieved, at least to a point after some period in which policy has been maintained tight, to then be able to ease back and give a little bit more breathing space to the economy. But the Brazilian economy is going to have a slow year in 1998, and that's probably necessary to allow the current account balance to adjust in a downward direction, which I believe is essential to maintain confidence in financial markets.

In China, I see the situation as different. For China, I think also, sustaining the exchange rate is important there, I think more for international reasons than domestically, although I think China also recognizes that if they were to devalue the yuan, that would precipitate another round of Asian devaluations, and it is by no means clear that they would end up as important net-gainers from such an episode. So they recognize not only their systemic responsibilities, but what the responses of the system would be in the event that they acted, and I think they are reasonably well-positioned to hold the exchange rate because capital is not freely mobile into and out of China; it is very difficult to mount a speculative attack on the yuan, and the problems in the financial sector, the way I describe it is that the Chinese financial sector is much like the U.S. social security system, that is to say, the value of the claims is largely dependent upon the willingness of taxpayers to pay in the future to make them good. But it is a state banking system and has lent money to state enterprises, so fundamentally, they are viewed as ultimately the liabilities of the state.

Now, China has also been in the favorable position that as its growth rate is now clearly slowing from the very rapid pace of recent years, they have been able to ease monetary policy in the face of an inflation rate that has gone almost to zero and also to adopt a more expansionary stance for fiscal policy in the face of domestic weakening.

China has the latitude to do that; Brazil does not. And it is important to understand that what is the right policy response for one country in the face of a slowing of growth is not necessarily the right response for another country, and I want to develop that theme just a minute or two more.

In Japan, we have seen a weak economy, and the International Monetary Fund has recommended easy monetary policy and expansionary fiscal policy, a virtually unique event in the history of the International Monetary Fund to recommend expansionary fiscal policy, and the coincidence with my arrival at the Fund did not make this outcome entirely unexplainable.

[Laugher.]

But Japan is in a different situation from Korea, because Japan does not have large net external debt; indeed, it is a very large net external creditor. So that when the yen depreciated significantly against the dollar, that did not create large capital losses for Japanese banks and enterprises that had borrowed in foreign currency. That was not the situation of Japan.

And other circumstances in Japan really distinguish it from the situation in Korea, so that what has been appropriate for Japan in terms of easing monetary and fiscal policy in the presence of a weak economy where we have a country without net foreign indebtedness with a large current account surplus is very different from Korea.

Similarly, if we look at other industrial countries, for the United Kingdom in the crisis of the summer of 1992, they let the exchange rate go, and fairly shortly thereafter, they substantially eased monetary policy, which contributed to a significant deprecation of sterling, now reversed--indeed, more than reversed. But at the time, it was the right policy given the economic situation in the UK. The right monetary policy was a relatively easy monetary policy which would enable the economy to recover better from the recession into which it had fallen, partly as a consequence of a tightening of monetary policy to bring down inflation, but later on as a consequence of needing to hold monetary policy tight to defend an exchange rate which was no longer realistic or appropriate to the macroeconomic situation.

Contrast the situation of Italy, which also was forced out of the ERM at the same time as sterling. In Italy, it was not feasible to have an early, substantial easing of monetary policy, because a huge fiscal deficit had created concerns about the credibility of fiscal policy in the medium and longer term to service that debt. The flexibility was simply not there to ease monetary conditions in Italy; it was there in the United Kingdom.

The message is that one policy does not fit all; that the policy that is appropriate to one country in what appears to be a certain set of circumstances may not be relevant to another country that faces similar problems but where the circumstances are, in other important respects, different. And in assessing what is the right policy response, the recognition of those differences continues to be an important challenge, and that is one of the reasons why I reject much of the criticism of the Fund--not criticism which says our fiscal policy recommendations may have been a little bit too tight, or our monetary policy may have been a little bit too tight; that is always an arguable point--but those who believe that it was a feasible policy for Korea or Thailand or Indonesia to ease monetary policy and easy fiscal policy as a way to reflate themselves out of their difficulties I think are talking nonsense.

The example of Brazil, I think demonstrates what needs to be done in the circumstances where a country is threatened with a potential cut-off in access to international credit, and it is very important for us to be able to distinguish between the alternative cases and to formulate the right type of policy response that is appropriate to the occasion.

MR. PERRY: Thank you, Michael.

In closing, let me use an image. Michael used the image of the Titanic very effectively, and I am going to borrow one from [inaudible], which is the superhighway. He says we should continue building superhighways--and he is talking about financial integration--because they get you more rapidly to the places where you want to go, and that is true. But at the same time that we build superhighways, we must be sure that we have good cars, with good brakes, that we impose speed limits, that we ask the passengers to wear seatbelts and not to drink before using the highways, and have now and then a police car appearing to check that those things do happen.

So that basically, some good rules and some good institutions are critical for financial integration.

Thank you.

[Whereupon, the proceedings were concluded.]

---