Getting it Right: Sequencing Financial Sector Reforms - Transcript

July 15, 1999


Thursday, July 15, 1999
International Monetary Fund
Washington, D.C.


MODERATOR:
V. SUNDARARAJAN, IMF Deputy Director, Monetary and Exchange Affairs Department.

PANELISTS:
ANDERS ASLUND, Senior Associate, Carnegie Endowment for International Peace.
GERARD CAPRIO, Director, Financial Policy and Strategy Group and Head, Financial Sector Research, World Bank.
NICOLAS EYZAGUIRRE, IMF Executive Director for Argentina, Bolivia, Chile, Paraguay, Peru and Uruguay.
R. BARRY JOHNSTON, Division Chief, Monetary and Exchange Policy Analysis Division, Monetary and Exchange Affairs Department, IMF.

Proceedings

MR. SUNDARARAJAN: Thank you all for coming here today, and particularly pleased to be here to moderate today's economic forum, which will address two questions. And I hope I am phrasing the questions in a proper way, so that the answers are feasible.

First question is: is the increased incidence of financial crisis that we have seen in recent years a result of the liberalization process itself? Which means we will have doubts about the wisdom of liberalization. Or is it the result of poorly managed and sequenced reforms? If the latter, is there an appropriate sequencing of financial sector reforms that minimizes the probability of crisis? These are the types of questions that we are trying to discuss today.

The focus of today's forum is sequencing different components of financial sector reforms not on the broad order of economic liberalization, which is a very different topic.

To participate in this debate, we have with us today four distinguished speakers. Dr. Anders Aslund, here to my left, and he is Senior Associate in the Carnegie Endowment for International Peace. Then I have Gerard Caprio, sitting on the far end. He is from the World Bank. He's Director of Financial Policy and Strategy Group in the World Bank's Financial Sector Operations, vice presidency. And simultaneously he is a manager of Financial Sector Research in the World Bank's Development Research Group. I have Mr. Nicolas Eyzaguirre sitting next to me. He is Executive Director in the IMF for Argentina, Bolivia, Chile, Paraguay, Peru and Uruguay. Then I have Mr. Barry Johnston, sitting at the far left of me. He is a Division Chief in the Monetary and Exchange Policy Analysis Division in the Monetary and Exchange Affairs Department, with responsibilities for research and analysis of financial policy, monetary exchange policy issues.

Let me just make some introductory remarks before I turn to the speakers. The search for orderly liberalization of capital flows, that is, to give an operational content to the concept of orderly liberalization of domestic and international financial intermediation is a topic high on the agenda of all the architects that are now designing the so-called new international financial architecture, those trying to reform the international financial system, drawing lessons from the last few years. Even though the debate is on how to bring about orderly liberalization across broader capital flows, the practical and analytical issues in capital account opening are fundamentally the same as those in domestic financial liberalization. If somebody wants to debate that, we can do that. That is, the principles that govern orderly liberalization of domestic financial markets apply very well to managing external liberalization.

Are there such principles? Do such principles exist? Now, at least in the Monetary and Exchange Affairs Department, Barry, who is sitting to my left, and I, we edited a book we just made available outside, on the issue of how to extract some principles of sequencing of reforms, financial sector reforms, based on country experiences.

Let me just make two remarks on what we found. The notion that financial sector reform should be implemented at the later stage after achieving stabilization is more or less discredited. Now most people at least--at least I believe--that the financial sector reforms should not be delayed until other reforms and stabilization programs are completed, but rather, components of financial sector liberalization should be phased in to support, to complement, to facilitate the stabilization policies and other structural reforms that are going on. This simple proposition has very strange counter-intuitive implications. It is often best to liberalize the short-term flows, money markets first, rather than wait until you liberalize the long-term flows and then come to short-term flows, as some people argue.

But, of course, you have to do this with proper prudential safeguards. But in fact, if you look at country experiences, you would find many countries have successfully done exactly that, deal with the short-term flows first before moving on to long-term.

Orderly liberalization often requires implementation of a critical massive reforms simultaneously. That is, you have to put together a package of reforms involving different components of the financial sector, aspects of banking supervision, aspects of money markets, aspects of monetary operations, central banking. You have to put together a package. It's necessary because of technical reasons, for financial stability reasons, and to be able to be effective in implementing stabilization policies.

And this leads to my basic conclusion of the book of what we are doing today in the international financial architecture. All major liberalization steps, all major adjustments in macro policy should be subject to what I call a systemic stability test. You should ask a question--sometimes you can say financial system stability test. The test is to answer the following question: what are the structural institutional measures that are needed to protect the financial system against various risks and sources of instability that we can foresee in the short to medium term, given the liberalization that you are planning? If you ask that question and try to come up with a package of policies to manage the risks, then you could hopefully reduce the probability of crisis.

With this in mind, in fact, the World Bank and the IMF have started a program called Financial Sector Assessment Program, where, for a set of pilot countries we try to do a stability assessment, and help them to identify the package of reforms.

So this is the type of background against which this forum is being held. Let me first turn to the first speaker, Mr. Caprio. He has a doctorate from University of Michigan. He has published extensively in the financial sector area, including a recent book on reforming finance, historical implications on policy and financial reform. He has held positions before in the US Federal Reserve, the private sector, including in the J.P. Morgan as vice president and head of global economics. And currently he is also working very closely with IMF as co-chair of the Financial Sector Liaison Group, for which has been established Joint Financial Sector Liaison Committee. So, Jerry?

MR. CAPRIO: Thank you. It's an honor to be here, and it sounds like it's going to be a loud honor with the speaker.

It's an honor to be included in the panel, and what my bio seems to have omitted, that it's also a delight to be back where I started my career in what was then called the Young Professionals Program and now the EP.

I should make the standard disclaimer that the comments I'm going to relay to you are mine and not those of the World Bank, its executive directors, et cetera. In this vein, I'm reminded that about a year ago or so I was at a conference, and there were a number of Fund speakers on the panel, and I was sitting in the audience at the time. And they got up, and after one of them made the standard disclaimer from the Fund's point of view, somebody in the audience turned to me and said, "You know, we never believe them when they say that. We always know they're speaking for the Fund."

(Laughter.)

MR. CAPRIO: Which led me to conclude that the difference between the World Bank and the Fund, is when the disclaimer is given from the Fund, no one believes them; and when it's given from the World Bank, everyone believes them because of our well-known decentralization.

I'd like to focus my comments around three points. First, on the dangers of unbalanced financial sector liberalization. Secondly, a few remarks on the difference between financial deregulation and the right regulation of the financial sector; in other words, this is to give me a chance to talk about the appropriate role for government in the financial system, at least briefly. And lastly, to say something about the element of what I would call right regulation sector, which bears directly on the sequencing issue.

On the dangers of unbalanced liberalization, it was always in the cards that the liberalization of financial markets, and specifically interest rate liberalization, had the potential to present us with some surprises, and the speed and scale of the consequences that we've seen, both favorable and unfavorable, have often been breathtaking. The broad outlines of the major risks were quite clear, I think. In advance, it was evident that moving from rationing credit by quantity to rationing it by price, could impose substantial losses on those who had privileged access to funds, including governments and their agencies. It also seemed likely that interest rates and asset prices, having been very tightly controlled in repressive regimes, would become more volatile, thereby increasing risk. And there were also some early warnings that with monopoly profits eroded, banks could become prone to excessive risk taking, indeed, even looting by their managers, behavior that one would expect would be tolerated by depositors who would be confident that they'd be protected by their governments.

Now, these dangers that were inherent in liberalization were magnified by the manner in which governments went about this. In many cases, the governments favored reforms that were cheap, easy and quick to implement, such as completely deregulating interest rates, formal reductions in directed credit programs, even capital account liberalizations. As tough as those may be, they still, when you think about them, come across as being cheap, easy and quick, compared to reforms that have the opposite characteristics, especially building the institutions in two senses of the word, that are needed for a healthy and sound financial system. By two senses, I mean institutions in the sense that people talk about financial institutions meaning organizations, the intermediaries that are in the sector, but also institutions such as the rules, laws, norms, even the entire infrastructure that influences the behavior of the participants in the sector. Lopsided financial sector reform, doing the cheap, easy and quick first, and not building those sound foundations, the institutional steps, is like constructing a building with some pillars missing, or at the very least, some pillars set in quicksand.

So in short, I would argue that reform of the financial system, as I and others said in the book that we put out in the early '90s on this subject, displays sensitive dependence to initial conditions. So one can give general guidance as to how to proceed with reforms such as start focusing on institution building first because it takes so much time, and liberalize only enough to give some impetus to the liberalization process, such as by bringing interest rates up to zero or near or positive levels in real terms, but not deregulating them completely, retaining some controls.

Also, when you start making recommendations for different initial conditions, you do have to take account of the vastly different starting positions in a large number of respects, including even the size of the system. The recommendations that you would make for a reasonably large financial system, one might suspect are going to be quite different when you're talking about trying to reform a tiny financial system, and the world abounds in tiny financial sectors. By one count that we put together, out of 160 countries that we could get data for, in 52 countries with 200 million people living in those countries, the entire financial system, meaning the assets of all the financial intermediaries, was less than the assets in the bank fund credit union. There are also 80 countries with 500 million people where the assets were less than those that were in the smallest or one of the smallest savings and loan institutions in Frederick, Maryland.

So you really have to think then that the approach to dealing with a sector that tiny is going to be fundamentally different from dealing with a much larger one. Some of the specifics on handling these different initial conditions, perhaps we can come back to in the discussion period.

Okay. Now, what about deregulation versus right regulation? Well, the government's role in the financial sector, I would argue--and I thought it was taken for granted, but I keep hearing from people inside and outside the World Bank that this is not clear--is different and should be different from that in other sectors, notwithstanding a lot of ideology on the subject. The pervasiveness of information asymmetries in finance, the prevalence of inter-temporal trade, meaning you're exchanging a dollar now for a dollar later, and at least in the banking sector, the presence of demandable debt make banking unique even before we bring in the importance of banks as part of the payment system. To quote Dornbush, banks are not like gasoline stations. If you pull into a gasoline station, you can fill your tank up and sample that gas. Now, it's true there's a bit of inter-temporal trade once you've filled up and pulled away, you've paid your money already, but the costs on doing that, and especially to not going back to that gas station, are relatively minor. With finance, it's the wait for the inter-temporal trade tends to be much longer and it's much more difficult to reverse.

Another way of thinking of the information problem is to recall a saying among bank CEOs, that is, heads of commercial banks. And this saying which was recalled at the time of the Barings fiasco by a 70-year-old banker who was working in Asia, is as follows: the saying is that if you, as the head of a bank, have someone working for you, you should watch them. If they're making money for you, you should watch them closely. If they're making very good money, you should watch them closer still. And if they're making absolutely fantastic money, you should fire them, because they must be excessive risk.

Now, think about it. That's the view of the head of a bank. So that suggests that the information problem, no matter what you're called, whether you're inside or outside, or whether you're supervisor, a market creditor, you've got a fundamental information problem in terms of finding out about the health of that bank.

Rapid reforms have routinely failed, not just because they've neglected the time-consuming element of building institutions, but also because they've been explicitly or implicitly, in effect, arrived at by default at an unregulated financial sector that was ripe for rip-offs. If you doubt that, consider that virtually every transitional economy has had significant ponzi schemes that stole substantial amounts of money from many families. To be sure, we see these schemes cropping up from time to time in many countries, but the scale that we saw these in the transition countries, such as in Albania and Rumania, was quite noticeable, to the very least.

Now, what are some of the elements of proper regulation or the government's role in the economy? From a very simple basis, although one can put together a more elaborate theoretical argument for this, the pervasiveness of the information problem in finance should mean that one would want to have as many very highly motivated monitors of the intermediaries that are active in the sector, especially those intermediaries that may be benefiting from a government guarantee, so if there's implicit or explicit deposit insurance in the banking sector. In other words, you don't want to put all your eggs in one basket, trusting one intermediary--one monitor--excuse me--to guarantee the safety and soundness of your banking system. This multiple-eyes approach that we've called it in the bank, involves both creating incentives and providing the ability for monitors to do their job well. Monitors are very simple. First line of defense against unsafe banking, I would argue, are the owners and bank managers, who should have good information about their institution. They'll have better information about the risks that they're taking to the extent that there's a better information environment, meaning that if there are better accounting and auditing systems, good protection of creditors' rights, that will improve bankers' ability to understand the risks that they're taking.

The incentives are even more important. If bank owners don't have much at stake, they can be depended on to take excessive risk with what Justice Louis Brandeis referred to in the 1920s as "other people's money." Since it's so very hard to verify the quality of bank capital, you may well have seemingly substantial capital holdings, but much of the capital has been borrowed. Indeed, in some of our countries, the bank owners have borrowed the capital from the institutions they supposedly own. That means that you want to think of other ways to increase owners' stake in safe and sound banking, such as by making sure that the banks themselves have a healthy franchise value, meaning that you can earn good profits from respectable banking rather than gambling, and/or by increasing the liability of bank owners and managers for imprudent actions.

Now, an example that bank supervisor friends of mine, for some reason, find interesting, is that we know very easily how to limit bank risk. We can execute bankers whenever they make losses. The problem with that, of course, is that it would kill off financial intermediation in the process.

So we know that if you raise liability limits too high, you'll get insufficient intermediation. We also know from many countries where we have banking crises, if there's no liability, you're going to get excessive gambling. So we know that the extreme--that the answer is probably somewhere in the middle, and countries such as New Zealand, that have lifted liability limits on bankers in certain very specific cases, such as when they disclose faulty or misleading information, could be interesting. We also know historically that when there wasn't the vast supervisory apparatus that many countries had, but rather there were higher liability limits on bankers, that bank losses were smaller than we've seen in the last two decades.

The second line of defense is the markets or the creditors of banks. Strong disclosure laws will certainly provide information once you've developed some accounting and auditing that the information is there and reliable in the first place. Forcing banks to issue uninsured subordinated debt is another attractive way to create a class of large creditors of banks, who will therefore have an incentive to monitor the risk that those banks are taking. And subordinated debt has the additional attractive feature that if they start getting worried about the health of a bank, they can't run all at once; they can only run as their debt is maturing. So it can also be a way to control the speed of bank runs.

It's interesting that in the wake of the current discussion paper, the proposed revisions of the Basle Capital Accord, that one of the few things that the shadow regulatory committees of Japan, the US and the Europe could agree on, was endorsing a subordinated debt proposal, rather than the elaborate proposals that the Basle Committee has come up with.

Lastly, I would note supervisors. They certainly have a role to play in insuring safety and soundness in banking. I put them last because, in my view, they are the last line of defense. They're an extremely important one, but if you don't have bankers overseeing their own institutions and you don't have markets monitoring, it's hard for me to believe that you'll have effective supervision. They might find out about the problems a bit too late.

There's been a lot of attention among all the international financial institutions on trying to improve the quality of bank supervision. What's gotten less attention, in my view, has been the issue of their compensation. This is often mentioned to member country governments, but when the difficulties of changing the compensation system are noted, it's often dropped and we go on to make recommendations for training supervisors. The problem is it may be the most important element in the equation.

A favorite example of mine, because it illustrates this so nicely, is what was called the Suffolk Banking System in the US between the 1820s and the 1850s. These banks--there was no central bank at all. These banks had their own clearing house association for clearing notes among each other, and understanding that they were running a risk, they had an exposure to one another, they hired their own supervisor. This was a private individual who supervised them. And he got paid a decent salary as his base compensation, but he also got a deferred bonus that was equal--last time I really did this was in 1990--in 1990 dollars it was about half a million dollars a year. So a good deferred bonus, and losses were deducted out of that bonus.

Now, the really nice feature of this system, there may have been a number of incentive properties that it had that were very attractive, but there were only a couple thousand dollars of losses over the 25 to 30-year period it was in existence. Moreover, some bankers tried to skip town with some money they had stolen from a bank, and the bank supervisor put a posse together and brought them back to justice.

Now, I don't expect to live to see that in this world. The reason why I point it out is that in many countries we're at the exact polar extreme in the way that supervisors are compensated, and in some countries they earn 1 percent or less of what comparably skilled people are earning in the private banking industry. Hard to believe you're going to have effective supervision with those salary differentials. So at the very least, changing the level of compensation, much less its structure, should be important.

Lastly, I would note that all this institution building is going to take time, and therefore, there's a role, in my view, for retaining some blunt instruments in order to insure safety and soundness, and again, we may be able to get to this in the question period.

Time doesn't permit me to mention the importance of the role of incentives in the non-financial sector, to make sure, for example, that excessive debt levels or excessive levels of short-term debt are not being artificially favored. I would also notice that one needs to be extremely aware or careful about applying best practice that we deduce from some countries to conditions in other countries. Again, given the importance I put on the initial conditions, you really want to think about tuning your measures to the specific country circumstances.

Since time's running out, I'll just mention one item. A lesson that was learned from the US was that if deposit insurance schemes were not sufficiently well funded, that the supervisors may not be able to intervene in a timely manner to resolve institutions, and therefore, the best practiced lesson is: if you've got deposit insurance, make sure you fund it well. The problem is that econometrically, in a cross-country sample that doesn't work. If you fund them well, you end up finding a greater probability of banking crises and larger losses. There may be a variety of things that are going on, but I suspect at least one thing is that many of the institutional safeguards that we take for granted in higher-income countries that limit looting of public funds, are not present in lower-income context.

Let me conclude then just with a final thought, that as the title of this forum suggests, getting financial liberalization right is indeed crucial, and financial reform done poorly is surely going to promote a backlash against maybe not even just financial reform, but all kinds of reform.

And the final lesson on sequencing in terms of generalities, when you start these elements, goes back to a--in this town, thanks to John F. Kennedy and now Larry Summers--is a famous quote by Jon Monet, the founder of the European community, who when questioning his gardener about a tree they were going to plant that day, and asking him how long it would grow to full height. The gardener's response was it was going to take about 100 years. And so Monet's reply was, "Well, then we'd better get about planting it this morning."

I don't mean to suggest that it's going to take that or anywhere near that long to get effectively working institutional structures in the financial sector, but it is a longer term affair to build those institutions, and one needs to start it now, even if real sector reforms are not well advanced. Thank you.

MR. SUNDARARAJAN: Thank you, Gerry.

So my impression is that you favor a gradual approach and no big bangs in the financial sector, right?

Next speaker we have Mr. Barry Johnston. I already introduced him. He holds degrees in economics and econometrics from Queen's University in Belfast and Manchester University. He has published extensively in the area of the monetary and exchange system and financial sector issues. Before he took up his current position, he was the Chief of Exchange Regime and Market Operations Division in the IMF's Monetary and Exchange Affairs, and in that capacity was actively involved in funds work on capital account liberalization issues. He has published quite a lot on that topic, so he will focus his remarks particularly on the capital account issues. Barry?

MR. JOHNSTON: Well, Gerry has relieved me of the need to give a disclaimer at the start of this talk. I suppose if I were to give a disclaimer it would be that there's no uniformity of view on these issues, and I think that's probably about the strongest disclaimer I can give.

Gerry also mentioned that financial sector reforms tend to be, cheap, easy and quick. I suppose the alternative view is that they're catalytic and beneficial in that sense, but I'm sure we'll get back to those issues during the discussion.

And as Sundararajan mentions, I'm going to focus my remarks on sequencing capital account liberalization and approaches to achieving orderly liberalizations, issues that have become quite critical by the recent currency crisis in Asia, Russia and Latin America.

It's important to recall that countries have liberalized their capital accounts for a broad variety of reasons. They've included to accelerate their domestic real sector reforms, to accelerate their domestic financial sector reforms, to give credibility to their programs of macroeconomic stabilization and structural reform, and more generally, to achieve higher and more sustainable development and economic growth.

The trend to more liberal capital accounts also reflects global developments, technological innovation, financial market innovations, the liberalization of current payments and transfers, which have generally meant capital controls in general are much less effective. And essentially countries are largely faced with a position where they confront large capital movements, whether or not they have significantly liberalized their own domestic financial systems. The issues we confront are ones on how countries manage facing these situations?

So I'm going to focus on three issues. The first is: what is the conceptual framework that we apply when we look at capital account liberalization? The second is: what lessons have we learned from countries' experiences with liberalization? And the third is: what are the main elements of an orderly approach to liberalization?

Well, as Rajan mentioned in his introductory remarks, at a conceptual level we really see capital account liberalization asone element of broader programs of financial sector reform, basically, how do we develop systems, institutions, instruments and markets to handle a world where financial resources are allocated by market processes?

However, capital account liberalization also has additional dimensions than a program of domestic liberalization. I would mention three elements. First, it usually adds greater urgency to domestic liberalizations, and that greater urgency arises for two reasons. First is that when capital comes from abroad, it will be intermediated by the domestic institutions and markets, and therefore, you have to develop those institutions and markets so they can efficiently use the capital. The second is that capital flows tend to increase competition on the domestic financial system, and therefore, it increases the pressure to actually reform those systems.

The second additional element, when one has a program of capital account liberalization, is the nature of capital flows themselves. They tend to be highly sensitive to the incentive structures created by the interest rate and exchange rate policy mix that the countries themselves establish. This concept is sometimes known as the impossibility trinity, that a country cannot have a fixed exchange rate, an independent monetary policy, and free capital mobility. Therefore, countries liberalizing their capital accounts have to place a premium on having a consistent monetary and exchange rate policy mix.

The third additional element involved in capital account liberalization is that capital flows typically involve additional elements of risk to those that are specifically seen in domestic financial transactions, and therefore, policies and procedures have to be developed to handle those specific elements of risk.

What do we find when we look at countries' actual experiences? Well, here I fully agree with what Gerry said, that countries' specific circumstances are critically important. Indeed, because countries start in different circumstances, there can be no unique approach to sequencing financial sector reform. There's no unique approach. In fact, what we tend to find is that what matters critically in achieving liberalizations is the package of reform, the supporting policies, the synergies within the reform mix itself.

What about speeded reform? Well, here I have to depart from what Jerry had said--that all rapid liberalizations are a disaster. In fact, our own assessment of countries' experiences with capital account liberalization is that countries with both fast and more gradual approaches to liberalizations, have had successful experiences; they've also had difficult experiences. The speed of the form itself does not seem to be a critical element in whether reform is successful or not. What is critical is whether it is part of a comprehensive policy package.

What about the liberalization of the components of the capital account itself? We hear quite a lot of discussion of what I would call general rules that we should apply when we liberalize the capital account. Liberalize long-term flows before short-term flows. Liberalize foreign direct investment flows before portfolio investment flows. Now, these rules themselves I think sound extremely good, but when it comes to apply them in practice, it is extremely difficult in practice to distinguish between different types of capital flows. Many countries, for example, began by liberalizing foreign direct investment flows because these are clearly beneficial in terms of transferring managerial skills and technology, but if you look across countries generally, one of the most regulated sectors of the capital account is foreign direct investment, because these sectors are regulated, not for economic reasons, but for social reasons, strategic reasons, and therefore, simple rules themselves are more difficult to apply in practice.

Now, in the wake of the Asian crisis, there's been much discussion of the need to manage the risks in liberalizing short-term capital flows. I think Rajan mentioned this. Short-term capital flows do play a critical role in the international financial system. They're critical in trade finance. They're critical in establishing foreign exchange markets. They're critical in establishing money and security markets where risks are managed.

What no countries' experiences tell us about liberalizing short-term flows? We have learned three lessons about where problems can arise. The first is: where regulatory frameworks have tended to bias flows toward short term flows, that has created difficulties for countries, and that bias is not necessarily explicitly in the maturity structure of flows, it is quite often in terms of the biasing the institutions which transacted in markets--banks rather than securities markets. The second lesson we have learned is that countries which provided implicit or explicit exchange rate guarantees, and at the same time had an independent monetary policy--in other words, they did not recognize the impossibility trinity--created incentives for significant and volatile short-term capital flows. And the third lesson we have learned is that there needs to be specific attention to the risks in short-term flows. These are risks of currency mismatches, liquidity mismatches, and one needs to put in place procedures to manage those risks.

How do we design an orderly liberalization of a capital account? The first key element is that it has to be accompanied by domestic financial sector reform, and many of the elements that Gerry mentioned are part of that. Part of that is in terms of the prudential measures that countries follow, and adopting appropriate prudential regulations. Quite often, countries with weaker financial systems and regulatory structures use capital controls to achieve prudential measures. What we observe is that as countries evolve their capacities, for prudential regulations of their markets, well-defined prudential measure take the place of capital controls, and that's because capital controls themselves are not well designed to manage risks in financial transactions. So we see countries evolve their own regulatory framework as they develop their systems.

Other elements of a supporting policy package for orderly capital account liberalizations are restructuring weak and insolvent banks. Indeed, where banks are weak or insolvent, one would want to restrict their access to international capital flows, and so there may be a case for imposing controls on selective capital movements. Similarly, as Jerry mentioned, we would want to strengthen in auditing, accounting and disclosure practices. We would also want to develop deep and more liquid financial markets and institutions, develop indirect monetary instruments, and as I've mentioned, adopt a consistent monetary and exchange rate policy mix.

How does capital account liberalization fit into this process, particularly when you've got a weak financial market? Capital account liberalization is not an all or nothing affair. We have recently established information in the Fund and that has identified more than 40 categories of capital transactions and movements. There are elements within the capital account, which can be liberalized and are liberalized at different points as a country liberalizes its financial system. The key issue in designing that liberalization is to identify the synergies of those liberalizations with the objectives of building and establishing efficient financial markets. So essentially, one looks for reforms that can help build your financial markets, build the instruments and create a situation in your financial system, which can then be more robust to shocks.

Conversely, that sequencing of liberalization would also involve perhaps introduction of certain controls on capital movements, as I mentioned, on insolvent banks, where their borrowing would create risks before the financial infrastructure is established.

So let me sum up. What would we see as a quote, "modern approach" to managing and liberalizing capital flows that can reduce a country's vulnerability to capital movements, and also put it in a better position to handle any shocks that arise from capital flows?

There are four key elements of that modern approach. The first is an internally consistent monetary and exchange rate policy that is sustainable over time, and which will itself help to reduce volatile capital movements. The second key element is the identification of prudential management of the additional dimensions of risk that are involved in capital flows. The third element are the supporting financial sector reforms, the development of healthy and sound banking systems, auditing, accounting and disclosure standards, introducing of indirect instruments, accelerating financial market development. And the fourth element is the sequencing of the capital account liberalization itself. Which component of the liberalization does one liberalize at what point in time to avoid biasing those liberalizations towards more unstable capital flows?

And I think that all of these elements are likely to be necessary in a program of orderly capital account liberalization. Thank you.

MR. SUNDARARAJAN: Thank you, Barry.

So we have a large number of reforms to be implemented. The question is how quickly we can do that, and if you try to do it too quickly, what happens?

So let us reflect on some of these with our next speaker, Mr. Eyzaguirre. He is Executive Director in the IMF. He is a graduate of University of Chile and Harvard University. He has published numerous books and academic papers on macroeconomics and finance. He is a central banker. He was the director of Research and Chief Economist of the Central Bank of Chile, when he also helped design the now famous capital account policies. And see, so he has some special insights to offer us. He has also been very active in addressing countries on structural reforms in the monetary and financial policy area.

Mr. Eyzaguirre, please?

MR. EYZAGUIRRE: Thank you, Mr. Sundararajan. I guess we share the honor of being high in the ranking of most unpronounceable names, last names. Just two disclaimers. The first one is that in spite of my effortless tries and those of my English teacher, my English still is very bad. The dangerous thing is that I seem to be losing my Spanish.

(Laughter.)

MR. EYZAGUIRRE: The third disclaimer is that I don't have a watch, so if I take too long, please tell me.

Okay. First idea. I believe this dilemma between whether you should aim towards big bang or gradualism is to a large extent a false dilemma. The transition between closed economy with a repressed financial system towards a market-based open financially economy is a very, very difficult one, and as Mr. Caprio was highlighting--I couldn't agree more with him--the amount of institutional building that takes place is enormous, sometimes overwhelming. So I guess most big bang attempts will unavoidably end in a crisis with a subsequent reversal, and at the end of the day, it will take quite long for a country to accomplish the full deregulation, even more than if you go gradual. You cannot really separate the liberalization process to building blocks, like saying, well, first I'm going to go to the monetary area, then to the fiscal area, then to the domestic financial area, then to the capital account area, and so forth. The amount of inter-linkages and sort of vicious circle that can emerge between those four areas are important enough so that you will need to seriously look for another approach. I believe what was getting interestingly apparent is that you should aim at something like critical mass of reform entailing measures in all areas that should be coordinated in a concurrent set of reforms. So a good sort of analytical view should be to differentiate what are the core areas for reform that should be taken at an early stage, and which ones are more subsidiary and can be postponed to a later stage.

Having said that, I will try to identify some of which I believe are those core conditions in the different areas based very much on the experience of failure and to some extent, success that I have seen in the countries that I represent, especially in my own, Chile.

First, in the macro area, I believe that some degree of fiscal soundness is really a prerequisite. If you really have a big pending fiscal issue and you want to begin a process of deregulation of your financial system, you will be continuously shocked from the macro area to a degree that that the newborn financial system may not be able to handle. When you move from credit allocation ceilings to interest rate deregulation, and you have an unsolved fiscal position, the crowding out would imply quite high interest rates that are going to be a difficult issue to handle as well. Since after starting your stages of financial liberalization, lots of agents facing that high interest rates will try to get financing any way, doing some what is called gambling for resurrection.

But maybe even more importantly, to have a sound fiscal position is key to make corrections in order to handle the financial crisis that for sure--either to a small extent or to a big extent--will arrive down the road. If you just check the amount of time that it has taken to recover to an economy hit by a financial crisis, and you correlate that with the initial fiscal position, you will see that the difference is enormous in favor of the ones, of course, that did have sound fiscal positions in the first place.

Same lines of reasoning can be argued for achieving price stability, first, because it provides macro stability. Price stability provides some credibility to the central bank. There's no economy really completely isolated from the rest of the world, so a credible central bank is quite important to avoid attacks on your currency, even if your capital account is quite regulated still. In this area, but more in a sort of a structural arena, I believe our experience has been that an earliest possible separation of the duties of the central bank in terms of price stability, the fiscal policy in terms of fiscal policy, and of the supervisory authority, to avoid the supervisory authority to be mixed with multi-considerations or with political consideration through the fiscal side, is very, very important.

In relation to the financial sector reform, given what I said at the beginning, you cannot wait until you have perfect monetary and fiscal stability to begin this process for two reasons. First, because financial reform is very important to enhance growth, and second, if you look at the experiences of countries that decided to begin to liberalize, one of the key considerations there is that it is very hard to maintain macro stability forever in a very close and repressed economy, not the least because you cannot deliver growth, and if you cannot deliver growth, social problems will mount, and that will impact your fiscal position pretty soon. So the more important thing here--I may be repeating to some extent what Mr. Caprio said--interest rate liberalization should be cautious and maybe some band or suggestion on interest rate at the early stages is well taken. Rates will have a tendency to overshoot at the beginning, and some guidance may be appropriate. Also you have the very well known excessive expansion of credit at the early stages of financial liberalization; this should fully factored in when thinking about counteracting measures. By the same token, if you're going to realize interest rates, you want to dispose of the direct monetary control instruments just gradually, as you can build institutions to have open market operations and more sophisticated market based instruments of monetary control.

But I believe probably the more crucial area, the issue to focus at an early stage are the areas of supervision and regulation, and of course, enforcement capabilities. In the experience of a number of countries in my part of the world, the inter-related lending that arises at the early stages is a very, very difficult and pervasive phenomenon. So you may want to put controls there to enforce them, but more importantly, to work in part of, in the area of corporate governance in order to make it possible for the new banks and for the public in general to have the transparent information to assess to the extent possible the credit decisions.

A typical development that gets somewhat delayed and is a source of instability is an inappropriate deposit insurance scheme, where either you do not have any but there is the general expectation that if institution fails, the government is going to bail it out, or you have it, but poorly financed or without clear boundaries to it.

Last, but not least, among the key areas for supervision, I would put the loan classification, provisioning and interest capitalization. It is a common practice of the newly deregulated banks to begin to lend again to borrowers that cannot pay their interest due. So it begins a full process of distressed lending or ponzi games and goes well beyond fundamentals.

Now, I believe where it is possible to say something more definitive in terms of sequence is in the capital account deregulation. If I would choose, I would clearly put macroeconomic balancing and the financial--domestic financial sector deregulation first on the list. And as I can advance a little bit on that front to proceed with capital account deregulation. There is a number of pervasive and perverse developments that can arise if your financial system--domestic financial system--is not well supervised and you opened your capital account. Foreign lenders are going to say soft budget constraints of bank or will be willing to lend distressed banks, on the basis of an expectation of a bailout. That happened in Latin America in the '80s and even in the '90s. But more importantly, because the capital account entails new risks, very different ones, transfer risk, solvent risk, exchange rate risk, and the more important one, you do not have an lender of last resort facility. So if you are an emerging market and you have say a short history of stability and you belong to a neighborhood that is difficult, let's say, you may face, regardless of what you want to do or how well you do it, considerable instability of foreign lending, and since you cannot, by definition, hedge that risk, because most emerging markets are not able to raise funds abroad in their own currency, so you cannot hedge away that risk. You really want to think it twice.

Even though I'm very cognizant of the crucial importance of not delaying this part of the story, let me provide some illustrative calculations. If you think that an emerging market can afford a current account, deficit financed by a foreign influence of, let's say, 3 to 4 percent of GDP, within sensible capital operations, we may be talking of a rate of growth--I mean an increase in the potential rate of growth of some 1 percent to 2 percent per year, so that's big numbers. So you cannot really afford to postpone this indefinitely.

What my views are on this follow closely what Chile has done. I'm not of the view that you cannot distinguish transactions. To distinguish a foreign direct investment from a portfolio flow or from a short-term loan, bank short-term loan is not that difficult. But what I found weak in the arguments of some that are quite opposed to the idea that you can differentiate, is that--there's the whole argument that controls may be evaded and that you need some institutional capability to differentiate and enforce those controls. But what's the alternative? The institutional capability you need in a system without controls is also very high because you will need to have implementation capabilities at the level of the regulatory agency, and, to assess those risks, and that's a difficult thing. So I would clearly favor deregulating first FDI, for their well-known arguments, and maybe some long-term governed borrowing.

I'm also quite open to the idea that portfolio outflows can be deregulated at a quite early stage. Basically what you need there is to make sure that you won't face a negative externality arising from the risk that if one corporation of your country raises funds abroad and at the end that corporation fails, that may produce a negative externality on the rest of the economy. So maybe you want to start with your better companies and to demand them a high credit rating by international standards.

Last but not least--I'm running out of time I imagine--short-term flows, they fulfill a service no doubt, trade credit and all that, but are particularly volatile, and if you are an emerging market, your currency is not hard currency, you will be facing attacks. So at the least you would like to isolate to the extent possible the solvency and liquidity of your banks to the wide movements of these flows. There are a number of ways you can do it. Maybe we can touch on that in the Q&A period. Thank you very much.

MR. SUNDARARAJAN: Thank you.

Next speaker, Mr. Aslund. I introduced him already. He was a professor at Stockholm School of Economics and director of Stockholm Institute of Open Economies. He has published extensively. His well-known books include How Russia Became a Market Economy, Post Communist Economic Revolutions, how bigger bang has--you know, many of us have read it. He has been active in advising Russian and Ukrainian governments, he is a member of the Russian Academy of Natural Sciences.

MR. ASLUND: Thank you very much, Mr. Chairman, and it's a pleasure and honor to be here today.

My presentation today will be rather different from the preceding ones. The other speakers have spoken about what should be done. I will rather concentrate on what could be done, and I will focus only on two countries which I know the best, Russia and Ukraine, and I will look up on what really came out of the financial crash in Russia and the crisis in Ukraine last year.

And I very much come down to in the end the question: what do we prefer, stability or change? In very broad terms. There is very little that we can determine in a newest way, because the issue here is really that the states are very weak. There are choices, but they are much more limited and much less attractive than we often want to think.

When last year both Russia and Ukraine faced very severe financial crisis, the reason for these crises are elementary. There was nothing strange about them as the Asian crisis. Both countries had too large budget deficits, and they couldn't refinance them. Both Russia and Ukraine had overvalued real exchange rates because of corridors that had allowed too little devaluation for a few years, not keeping up with inflation, and both countries suffered from problems servicing the short-term debts because of limited international reserves and small federal revenues, rather than because of a large overall government debt.

And the big difference between the countries was really that Russia had attracted very substantial foreign portfolio investments, $46 billion in '97, that was 10 percent of GDP; while Ukraine, because of a less attractive investment environment had attracted rather little. And this is really the reason why Russia went into a serious financial crash and Ukraine saved itself. Ukraine was less vulnerable, and therefore, Russia also had the larger budget deficit, et cetera.

The cures for this were of course obvious. The budget deficit had to be less. The exchange rate should have been devalued earlier and faster, and the government should have restricted its own borrowing, at least of short-term international capital by not issuing so large a volume of treasury bills. But the question is rather why this was not done.

But let me first move on to the effects of the crisis, and I'll look first on the immediate effects, and then the effect that they look today, one year later, and the perspectives are very different.

To begin with, of course, Russia and the crash looked horrendous. Industrial output fell by 15 percent September last year. The IMF, last October, forecast a decline of GDP this year by 9 percent of GDP. And eventually GDP fell by 4.6 percent in Russia; in Ukraine by only barely 2 percent, and this was essentially an effect of Russian crisis. You should not blame it on the Ukrainian policies at the time. Both Russia and Ukraine devalued, Russia by four times and Ukraine by two times nominally.

Inflation-wise, Russia got 200 percent inflation June to June; Ukraine just doubled its inflation to 20 percent. So both countries saw real devaluation of 40, 45 percent in both their exchange rate, you could say, approximately right, perhaps a bit more devaluation than they really needed, but that's not normally a problem.

Russia defaulted on its treasury bills while Ukraine has not really done so. Half Russia's banks collapsed, but only a few in Ukraine. And the standard of living fell sharply in Russia by something like 30 percent. You can see that the financial crisis in Russia was socially more costly than the whole prior transition period in Russia, while in Ukraine it was much smaller decline.

So if we look up on it from this perspective, Ukraine seems to be doing much better in terms of output, inflation and social costs, simply by failing to attract foreign capital. However, if we look up on the effects now, one year later, it looks quite different.

Both Russia and Ukraine have undertaken substantial changes. The budget deficits are very small in both countries. Both countries have been forced to undertake radical cuts in public expenditures. The treasury bill markets are dead in both countries, and both countries have lingered on the verge of external default, but essentially avoided it just doing some restructuring. And both countries are getting IMF funding now with some delay.

And the essence here is that both countries are managing to do a lot of the things they couldn't do before, because the financial crash in Russia and the repercussions in Ukraine was really that the hard budget constraint became credible for the first time. What the reformers never managed to do the financial crash did by being an exogenous force coming from outside.

But there are several changes that are much stronger in Russia, which give much more hope. And one is that Russia has really been forced to undertake a substantial bank restructuring. While there's a lot of complaint that the bank practice are not correct, the bad banks have been forced to close, and it's less important what really happens to them. The important thing is that they're gone.

As a result of the hard budget constraints, all kinds of arrears are falling in Russia now, month by month, and monetization is rising sharply in Russia. It's also in Ukraine, but these tendencies are weak in Ukraine.

Income differentials have fallen sharply in Russia as an effect of the new upper middle class being severely affected by the financial crash last year. Not at all that noticeable in Ukraine. And in particular Russia, half of so-called oligarchs have lost out, and this means that economic powers have been much more broadly distributed. You can say that in Russia the powers moved in the economy from the financial capital to the manufacturers, while in Ukraine, it's still with the commodity traders, the first stage of transition to capitalism.

And unlike Ukraine, Russia is now already seeing an inflow of portfolio investments again. The Russian stock index is up almost four times from last October; the Euro-bonds are up three times. And the most important indicator, Russia's growth is not a decline of 9 percent this year as had been forecast, but already in the first half, has an industrial growth of 3 percent, and it's clearly heading toward double that amount for the year simply because it fell so drastically during the second half of last year, while Ukraine remains stagnant.

So what we are seeing is that Russia, thanks to the serious crash, has possibly gone through a significant revival, this is only to say, but the indicators we have so far suggest that, while Ukraine is stuck in stagnation. And what's behind this? If you look up on the fundamental problem of the transition it is an omnipotent oligarchy, and it had to be beaten.

When discussing Russia/Ukraine, we must consider the economic power structure, and the fundamental problem was that these countries were dominated by rent seekers coming out of the old Communist elite, who wanted to make money on the rents to government subsidies and government regulations. And at the heart of this rent seeking were the Russian banks that were not banks at all, but you can say all powerful general companies doing whatever is needed, and therefore, it was of course impossible to regulate. To suggest that they should be regulated is just anachronistic. It could not be done, and it was not done.

So the only way they could be beaten is by long periods of attrition or an exogenous shock. Similar, the budget deficit could not be brought under control as long as the elite thought they could get more subsidies from the government. Again, an exogenous shock was needed. The same with the high treasury bills, which were a main source of rents. And structurally, since the Russian banks were not banks, they blocked the actual development of banking, and you do need banking in economy. So the crash took out the banks that were not really banks, and left the real banks behind. So therefore, the payment system is financially seemingly much better after the crash. And therefore, we can now see some revival.

So then what conclusions can we draw from this? The first conclusion I draw, it is an initial task--the main task of the initial transformation is to impose hard budget constraints, and in order to do so, you need to disarm the old oligarchy. The financial crash is actually quite an effective means to achieve this, and we can notice that Poland went through two serious financial crashes before it became a most successful transition economy. It's possibly the only way of break these power structures up, and we shouldn't shy away from that observation.

Secondly, an early and radical liberalization can weaken the oligarchy, either immediately, if you have a successful liberalization, or later on because it facilitates a later crash.

Thirdly, in Russia, the banks were so strong politically at the outset of the transition to a market economy because they had got all the powers already before the transition. So any suggestion that the banks should be regulated is simply out of context.

And, finally, the Russian and Ukrainian Governments cannot do very much to resist vested interest. So the question is rather: what are the changes? And therefore, you have very little choice about sequencing of reform. The simple response is do as much as you can whenever you can do something. Orderly liberalization was out of question from the outset. The choice is really one of bold and bust or petrifaction and stagnation, and hard as it may sound, I do think that the first is preferable. Thank you.

MR. SUNDARARAJAN: Thank you. So we have some time for questions, or optimistic remarks on the future of Russia. Yes, please?

QUESTION: Just a comment on the difference between smaller and larger economies. For very small economies it's clearly going to be a very, very long time before they develop their institutions, based on their domestic markets and thus to be able to open up their markets. Is there a need need for different approaches for small and large economies.

MR. SUNDARARAJAN: Thank you. Any comments on that? Some sort of sequencing is driven by the market. In the initial stages, governments can play a catalytic role but it is always a interactive process within the market and the government. Any comments on that?

MR. CAPRIO: Just one point. I fully accept the interaction. It's not necessarily the case that in very small economies it takes such a long time. It may take a long time if one is determined to have a sort of separate financial system in that small economy, but the other way to do it may be to go to a regional banking model or importing your financial services. If you want to have a safe and sound banking system in a very small economy, meaning you want to have domestic banks that concentrate all their risk internally, you're going to need to take a number of rather extraordinary measures to induce safety and soundness. So another approach is to allow good foreign banks to come in or to become a partner in a regional banking system.

MR. SUNDARARAJAN: Any other comments?

QUESTION: I have a question. It's for Mr. Aslund. Would you think that the Asian crisis was also unavoidable?

MR. ASLUND: The western economies have had serious financial crisis in the last 20 or 30 years. Basically I think that this is pretty necessary for most countries from time to time. I don't think that we are sufficiently good to keep up with common sense without being shaken up from time to time, and that most countries get into a such kind of vested establishment. So therefore, I think that this is something that is rather broadly true, but of course you have countries that have sufficient checks and balances. MR. SUNDARARAJAN: Yes, please?

QUESTION: Are the other speakers on the panel in broad agreement with Mr. Aslund?

(Laughter.)

MR. SUNDARARAJAN: Well, I will add my view--I'm not sure that the financial crisis is strictly a necessity, particularly when you look at the collateral damage. So you need to achieve the same type of cleansing, whatever, you know, of the new equilibrium. How you achieve new equilibrium without going through your crash? That is really the question that we are addressing today, and I think this whole idea of institution building, idea of creating institutional checks and balances along the way, as you keep liberalizing, as you keep managing, the reform process could produce an exit policy which could be more efficient than a wholesale crash at a particular time.

In a sense we are talking about how do you organize an orderly exit, and to recognize problems promptly, rather than postpone them. It becomes in a sense a political economy type question, but, you know, there are a lot of reforms which makes great sense in theory and cannot be implemented in practice because there is no incentive in the system for implementing them. You know, that is reality.

So the question is then how do you create incentives for implementing good practices, good policies, and that's the question for today's new architecture in a sense. Let's say you are focusing so much time on standards, how do you create--you know, transparencies, and other instrument--what he's talking about, hoping that transparency will create in more smoother crashes, not big crashes.

(Laughter.)

MR. EYZAGUIRRE: Whether it was necessary or not, it seems to me a little bit of a philosophical question. It's like, you know, C.S. Lewis liked saying, "Only after pain will come wisdom."

(Laughter.)

MR. EYZAGUIRRE: I don't know. What I know is that whether or not it was necessary, it was unavoidable, given the policies that were undertaken in a number of years before the crisis, I really don't think that IMF package or all that contributed to the crisis, rather the opposite. It was just an accident waiting to happen.

Now, what is good in this is that you do not have to feel the pain yourself in order to learn, that is, you can learn from others' experiences, and I guess the world as a whole is learning from that experience, as, for instance, did a lot of Latin American countries from our own Chilean experience in 1982. So this is very much a process of gathering experience, and at the end I don't think--back to the question of the person from Barbados--that everybody will need to undergo a process of 200 years to get it right.

MR. SUNDARARAJAN: Jerry, any comments?

MR. CAPRIO: Just a couple quick points. I would also argue that although we've had banking crises for as long as we've had financial intermediaries that were like banks, going back, in other words, millennia, the crises were never the scope that we've had in the last two decades. So it's hard for me to believe that it's literally impossible to keep them from being so large. Indeed, if you went to one case like Argentina in the 1980s, people might have quickly concluded it was just impossible to prevent crises in that country. Although it's impossible to make any banking system immune from financial crises, the turbulence, that at least up till now the Argentines have withstood, given much higher and risk sensitive capital requirements, a greater proportion of foreign banks in the sector, a subordinated debt requirement, greatly beefed up supervision, much stiffer disclosure laws. And they weathered tequila successfully. They weathered, thus far at least, East Asia quite well, suggests that making those changes can have an impact, even in environments that seem to be quite difficult.

MR. SUNDARARAJAN: Thank you. Barry?

MR. JOHNSTON: I think that he has put his finger on a key issue, something I've struggled with for a long time, are financial crises inevitable, indeed desirable as part of a reform process?

I suppose that as officials who are advising countries on how they reform, our basic role is to identify what leads to crises, what are the elements of a financial sector reform program which creates shocks to the system? And we have now brought up a body of knowledge about what sort of shocks financial reforms give to the system and what sort of policies you can put in place to mitigate those shocks. Some of them are the ones that are being mentioned here, some phasing of interest rate liberalization, certainly institution building.

So I think the whole issue of sequencing and the whole issue of advice to countries is basically to help them mitigate the costs of those banking crises, because they are costly, in terms of losses, at least temporary losses, can be quite considerable. And the ultimate objective is to maximize the long run discounted value of your income created by having the shock or by avoiding the shock through good policies. And I think the latter is what certainly all of our advice is trying to do.

MR. SUNDARARAJAN: Thank you. Any other comments, please? QUESTION: What do you think about Malaysia's capital controls?

MR. SUNDARARAJAN: Yes. We have an expert on Malaysia right here, Mr. Barry Johnston. He can answer that question.

MR. JOHNSTON: Which aspects of Malaysia's capital controls, since there are many different aspects of them? Essentially, the jury is still out in terms of what has been the impact, the final impact of those controls. We're still really trying to assess what has been the overall effect of those controls on the Malaysian economy. The initial response of international markets to the controls was highly negative, and Malaysia was taken out of the MSCI index. On the other hand, the domestic industrialists were quite favorable to the controls. They liked the reduction of interest rates. They liked the stability of the exchange rate. More recently, Malaysia modified its controls, went from a moratorium on repayments of debt to an exit tax in February of this year. That seems to have improved the international environment towards capital flows.

But as I say, we're still assessing that experience. And what has been evident is that the experience of the Malaysian economy, the trends in the economy have not been significantly different from the other Asian countries which didn't impose controls, so that in that sense the positive and negative effects of these controls would probably be less than was expected ex ante by commentators, because they're cyclical effect does not seem particularly different from other Asian countries that recovered without the use of controls

MR. SUNDARARAJAN: Thank you. There was a question here?

MR. : Two questions. First, do you think it actually makes sense to have some kind of ceilings on interest rates when banks are in a weak financial position and actually as a way to recapitalize the banks? Second, can you attribute the different experiences of South Asia and East Asia to their approach to capital account liberalization?

MR. CAPRIO: I thought I had an easy answer to your first question, yes, interest rate ceilings, but definitely not to use it to recapitalize the banks. Let me step back.

Even if you're in good fiscal shape, if you've got large losses in your banking system, you're in bad fiscal shape because that's a contingent liability and it's going to come back on to the budget. And just because you don't want to clean it up doesn't mean that it's not still there, and indeed, it's probably growing. All the experience we have indicates that if you leave open banks that are insolvent, the insolvency grows and usually grows dramatically. So you're looking at a truly unsustainable fiscal situation if you leave that open.

So one way or another you're going to pay those costs, and the risk of paying it through hyper-inflation rises dramatically the longer you let that situation exist. So I think--you know, if you think of these banks as machines that make bad loans, you want to first correct that machinery, and then you want to inject the capital. You don't want to do it in the reverse order. That's very important. Until you're certain that, or reasonably sure that you've got the insolvency cleaned up and a regulatory apparatus installed or an environment that's going to be conducive to safe and sound banking, you may certainly want some ceilings on deposit rates to make sure exactly what you're saying, that you don't have these bad banks in effect existing and bidding up the losses. But you definitely don't want to use that deposit rate control to try to recapitalize your banking sector, because you risk still larger losses.

And there have been a number of cases where banks in several countries have been allowed to do this, and they get a nice clean audit from an auditor, and less than a year later they've gone bust again. Why? Because they are machines for making large losses, and the problem was not addressed.

MR. SUNDARARAJAN: Thank you. On the comparison between East Asia and South Asia, what is the difference, do you have any comments on that?

MR. JOHNSTON: There are a number of differences between East Asia and South Asia. The Indian economy is much larger, more closed generally for trade as well as capital. You're talking about an economy, which had a stronger current account balance of payments position than the East Asian economy. It's a bit like comparing chalk and cheese to say that the only element that was different in the openness of the capital account. Certain elements of the controls in India, and probably China as well, limiting access of insolvent borrowers, probably give them some protection and help them avoid a crisis, or at least would have reduced the risks in that context. Whereas, we know that in East Asia, one of the problems there was borrowing by insolvent corporate and banking institutions. So there are elements of the capital account, which may have made a difference, but I think it's a broad--a far too broad generalization to say that it was openness versus closure of the capital account that resulted in the difference between the performances of these economies.

MR. SUNDARARAJAN: Thank you. Mr. Eyzaguirre?

MR. EYZAGUIRRE: Yes. I think that that comparison is not a very fair one because there are a number of other differences as well. But above all, no doubt, the capital account related framework puts you under new risks, but also gives you a higher mean in terms of mean and variance. So if you--I mean if a plane crashes, it's more worse than a bicycle crash, but the plane is far more useful to travel far away. You can have the better of both worlds. My own country, Chile, is an example that has not suffered distress from the capital account, from the current account, yes, but not from the capital account. And the high rates of growth that both East Asia and Chile were enjoying are in a good extent, thanks to the liberalization of the capital account.

MR. SUNDARARAJAN: Thank you very much.

In all cases where you have highly debted companies and a large portfolio of bad loans, there is merit for going more carefully.

Thank you very much for all of you for coming. It has been an honor for me to moderate.

(Applause.)

- - -



IMF EXTERNAL RELATIONS DEPARTMENT

Public Affairs    Media Relations
E-mail: publicaffairs@imf.org E-mail: media@imf.org
Fax: 202-623-6278 Phone: 202-623-7100