Transcript

Transcript of an Economic Forum
Ensuring Financial Stability: The IMF's Role
Tuesday, December 16, 2003
Washington, D.C.

(View a webcast of this Economic Forum using Windows Media Player.)

Participants:

Graham Hacche (Moderator)
Deputy Director
External Relations Department
IMF

Stefan Ingves
Director
Monetary & Financial Systems Department
IMF

Larry Promisel
Director
Financial Sector Global Partnerships
The World Bank

George Kaufman
Professor of Finance
Loyola University (Chicago)

Roberto Zahler
President
Zahler & Company

MR. HACCHE: This forum, Ensuring Financial Stability: The IMF's Role, is being webcast live, and the transcript will be published on the IMF's website.

Financial sector analysis and advice to countries have been part of the work of the IMF for at least 40 years, but up until the last five years or so, it took the form mainly of technical assistance to developing countries. The IMF Central Banking Service was established in 1963, and within a few years it had become the world's main source of technical assistance for developing countries in all matters related to central banking. This work expanded enormously in the following decades.

But also in the 1990s another major change occurred. Partly in response to financial and banking sector crises in a number of emerging market economies and industrial countries, the IMF took a number of actions to strengthen its surveillance, and one of them was to increase the attention it paid in its country surveillance to financial sector health and the identification of financial sector vulnerabilities. The surveillance became part of the work of the Monetary and Exchange Affairs Department, which is what the old Central Banking Service had grown into. A key step in this regard was the establishment in 1999 of the Financial Sector Assessment Program, a joint IMF/World Bank effort aimed at increasing the effectiveness of efforts to promote the soundness of financial systems in member countries.

Earlier this year the department was renamed again as the Monetary and Financial Systems Department, reflecting the increased importance of its work to promote the soundness of financial systems.

That gives you some background to the subject of today's discussion. We have a distinguished panel to discuss the IMF's role in ensuring financial stability in member countries. Stefan Ingves, to my immediate left, of the IMF. To his left, Roberto Zahler of Zahler & Company, Chile, and formerly of the Central Bank of Chile. To his left is Professor George Kaufman of Loyola University, Chicago. And the far end is Larry Promisel of the World Bank and formerly of the Federal Reserve Board.

I have asked each of the speakers to limit their introductory remarks to about 15 minutes so that there will be reasonable time later for them to take questions from the audience.

Our first speaker is Stefan Ingves, who became Director of the IMF's Monetary and Exchange Affairs Department, now the Monetary and Financial Systems Department, in 1999. Before he came to the Fund, Stefan was between 1994 and 1998 the first Deputy Governor of the Central Bank of Sweden, after having previously been Director General of the Swedish Bank Support Authority and a senior official in the Swedish Ministry of Finance. In his career he has acquired considerable experience in the management of financial sector crises, including in the early 1990s in his own country, Sweden. Stefan has a Ph.D. in economics from the Stockholm School of Economics.

Stefan?

[See Mr. Ingves' presentation (134 kb PDF file)]

MR. INGVES: Thank you. Let me give you an introduction to the Financial Sector Assessment Program (FSAP), a few comments on our experiences with the Program, and what lies ahead of us.

First a little bit of the background to all of this. We have learned the hard way that having a strong and resilient financial system is important for economic activity and growth, and we truly know how costly it can be if things go wrong. Financial stability is also very important in the context of the global economy, since a crisis in one country may easily spread across borders with large economic and social costs. Therefore, promoting financial stability, and shaping the international financial architecture are important parts of the Fund's mandate.

The launching of the Financial Sector Assessment Program a few years back is part of the Fund's response to these needs. So far, over half of the IMF/World Bank membership has participated in this program. It is important to stress that this is a voluntary exercise, performed at the request of country members, and the disclosure of the results is also voluntary. Over time, more and more countries have chosen to disclose the final reports which normally are available on our website.

When you look at the financial sector, stability rests on three pillars. One is the macroeconomic environment, since macroeconomic developments always affect the health of insurance companies and banks. Then we have the regulatory and the supervisory bodies entrusted with the monitoring of individual financial institutions, and then we have the market infrastructure in itself and the conditions of banks, insurance companies, stock exchanges, payment systems and so on. All of these components interact, and their interaction determines the stability and resilience of the financial system.

When we do a Financial Sector Assessment Program, we analyze the structure and the soundness of the financial sector in a country. In other words, we do the numbers and look at whether banks and other financial sector participants are vulnerable to potential economic developments. Nowadays, we also do a fairly large number of standards assessments. There is a whole set of best practices now available in several areas, for example, in bank supervision, the managing of a payment system and so on. We use these to assess individual countries' practices and to compare across countries. We also review the financial sector infrastructure in a country, take a look at individual markets, and evaluate how these markets operate, how they interact with each other. And all of this is then added up to an overall stability assessment.

Where relevant, in some countries we also define development needs, things that you need to improve, that you need to develop, to improve the functioning of financial markets. We formulate a coordinated action plan, helping countries to set a policy agenda to enhance the functioning of existing markets or help develop missing ones.

To do this, we are also cooperating with a large number of institutions outside the Fund. This is an international and a cooperative effort, since we do so in house and thus have all the experts that we need to carry out this work. And so far about 75 Central Banks, supervisory agencies and standard setting bodies like the Basel Committee, the IOSCO insurance supervisors, and so on, are now cooperating with us when it comes to performing the work. That means that this work has nowadays a substantial element of peer review built into it.

So far we have finalized or have ongoing assessments in 81 countries, 46 developing countries, 19 transition economies and 16 advanced economies. And we have about another 25 countries or so in the pipeline. In practice, this means that we are fully booked for the coming years.

Let me talk a little bit on the main findings so far. What have we learned? What have countries learned as a result of this joint work? Starting with the advanced economies, most of them have sound financial systems, both in terms of financial indicators and in terms of the institutional and regulatory frameworks. Generally you have very strong compliance with various standards and codes which perhaps is not so surprising because many of them have been instrumental themselves in producing these standards and codes. The traditional source of risk continues to be credit risk, which is something unexpected, but credit risk truly continues to be relevant, particularly when banking sectors run into deep trouble. In addition, the large dispersion in ownership of publicly traded companies and financial institutions tends to create governance problems and problems running the banks.

What are the main challenges going forward? First, the proliferation of financial instruments and markets over the last few years poses an enormous challenge in terms of regulation and supervision. As always, when new markets are being created, there is a need to keep track of what the new instruments do, and how risk is transferred in the systems. This involves options, futures, forward, swaps; all the financial engineering that is going on, which essentially is a technique to transfer the risk from the underlying instruments to other instruments and to other market participants. Second, there is an increased concentration and conglomeration in the financial sectors, and this very often shows up in the growth of large and complex financial institutions. Banks own insurance companies. Insurance companies own banks. All of them own mutual funds which creates a need to monitor risk across activities and increasingly so across borders.

Now, if you look at the emerging economies, what are the main lessons so far coming out of those countries? One conclusion is that significant short-term vulnerabilities remain in some of the emerging economies. On the other hand, there has been some progress in financial sector reform. A lot of things have actually been done, and important improvements have taken place in bank regulation and bank supervision, which usually stems from the fact that the banking sector is the oldest part of the financial sector.

What are the main risks and challenges in this kind of environment? Well, as in developed economies, a major risk is associated with the rapid growth of new instruments. If you create many new instruments you also need to have the capacity to supervise and fully understand the risks associated with them. Basically you need to understand what you are doing.

Another challenge, in some cases, stems from the lack of consolidated risk-based supervision. That has to do with the fact that historically, supervision has been organized on a bank-by-bank basis, and when you end up with more complicated structures than just individual banks, the supervisory processes have to follow along in order to cover all aspects of these more complex structures.

A third issue in many cases is the weak interagency cooperation. In very simple language that is to say that sometimes it's hard for people to talk to each other, but that you have to do so in order to carry out supervision efficiently. And then, there is a wider range of deficiencies in the non-bank area, and that's because it's a historic fact that surveillance is mostly focused on the banking sector.

Let me now move to the developing economies. In developing economies you often have a weak regulatory and supervisory framework. There are a few rules of the game and there are sometimes even fewer people keeping track of them. Also, there are weak risk management procedures in financial institutions because there hasn't been an interest in developing them, or has not been enough time to do so. And then there is an often inadequate market infrastructure in the sense that very often the banking sector is very, very dominant and there is not much of a bond market, the insurance sector is small, the payment system is deficient, and so forth.

Now, what are the challenges and risks in that kind of environment? One challenge is clearly low diversification of economic activity and the vulnerabilities you end up with coming out of that. If a country is producing mostly one thing and has a limited number of markets, the banks still lend to the very few producers you have in that country, and that produces vulnerabilities to occasional shocks.

Inadequate macroeconomic policies again prove to be important. If you have large macroeconomic imbalances in an economy, it is quite likely that those imbalances in one way or the other, sooner or later, will feed back into the loan portfolios of the banks.

Weak legal and institutional settings are also quite important. For example, credit markets fail to function properly, and financial vulnerabilities increase if it's impossible to seize collateral. That's just one example because developing countries tend to have many deficiencies in their legal and judicial systems. In this environment, almost any type of rescue operation, when it comes to dealing with banks in trouble, tends to grind to a halt, and it is very, very difficult to change things.

Now, in doing this work, what are the issues for us going forward? We need to sharpen the scope of the FSAP, refine the selection of issues and standards assessed. Now, let me restate that this work that has been going on for about five years. We started with a blank sheet of paper five years ago. And there is more work that needs to be done, more figuring out to do when it comes to improving and doing these things right.

We are moving into a phase where we also need to make updates because countries change things, they improve, and they say: "Well, you assessed us a few years back. Now we have changed a hundred different things. We want you to come and take a second look, because we want to tell others that we look better than last time." And that puts demands on us when it comes to doing the updates.

Also, we need to think hard about how to deploy new tools in financial sector surveillance, and how to perform follow ups of various kinds. This includes follow up of standards assessments, but also the type of statistical information we collect, and the statistical information we use in our assessments.

We also need to put further thoughts into how to develop new methodologies for looking at developmental issues related with the financial sector. What does it take to make the financial sector an integrated part of the economy in any given country, and what good things for growth and welfare in general come out of that?

We can always do more research on financial stability issues. The trick there is, based on our concrete experience, having done FSAPs in more than 80 countries and having another 25 in the pipeline, how can we use that experience in countries that are coming up, in order to support the practical, very hands-on policy advice that we give to authorities.

And it's also a responsibility of ours to bring our country-based knowledge back to the standard setters and tell them what works, what does not seem to work, and whether there are gaps in these various methodologies that are out there. By now we have done, I think, 60 to 70 Basel Core Principles Assessments, just to pick one example, and then it's up to us to tell the Basel Committee what we have learned in that process, so they can, in turn, use that information and that knowledge in their own discussions on how to further develop the Basel Core Principles, and address gaps in these standards. This is an issue that always comes back.

To close, when we do the work, it is one thing to make an assessment at a given moment in time, and another to implement change. If there are things to do, it's also up to us to work on better mechanisms for the provision and follow up of technical assistance. The final objective is to help countries deal with the problems and issues they have, so that we can make it easier for them to truly improve.

So much about the introduction to this topic.

MR. HACCHE: Thank you.

Turning to our second speaker, George Kaufman has, since 1981, been the John F. Smith, Jr. Professor of Finance and Economics and Director of the Center for Financial and Policy Studies at Loyola University Chicago. He has published a vast amount of material on money and banking. There's a 25-page list of articles and books on his website. Among the many positions he has held, he has, since 1986, been co-chair of the Shadow Financial Regulatory Committee, and he is currently President of the North American Economic and Finance Association. He has consulted for many official bodies and private firms, and he has been a visiting scholar at a number of the Federal Reserve Banks. He obtained his Ph.D. in economics at the University of Iowa.

George?

MR. KAUFMAN: Thank you. My talk is going to follow rather directly on Stefan's last point about a swifter mechanism for the provision of follow-up technical assistance. I'm going to argue that while financial surveillance such as the FSAP program by the IMF or internal programs, are an important and necessary, but not sufficient condition for minimizing the cost of bank problems on the economy.

It is also necessary to turn identified near-insolvent banks around before they become insolvent, and resolve identified insolvent banks efficiently at lowest societal cost. This requires I think two things. One, knowledge of how to do this, a set of best practices that Stefan has referred to; and secondly, advance planning, a set of plans on the shelf for quick and timely implementation.

Let me give you warning here that best practices are highly country specific, depend on the legal, political, social institutions of that country, but a number of rules apply broadly, and those of well-performing countries should serve as targets for others. Best practices depend on when bank problems are detected for individual banks to some extent. Are they detected before or after bank insolvency? At times it is difficult to tell precisely as accounting is not always accurate, and both bankers and governments in trouble often do fast and fancy accounting footwork to disguise the problem as long as possible, for example, late and under reserving for loan losses or booking imaginary revenues.

Let me first turn briefly to the troubled banks before insolvency. One, you try to identify and take action as soon as possible to rehabilitate the banks before it is too late. For example, prompt corrective action, or PCA, was adopted in the United States as part of FDICIA reform after the S&L and banking crises of the 1980s, are not a bad best practice. PCA imposes regulatory sanctions on troubled financial institutions modeled after the sanctions that the market imposes on nonregulated firms. The sanctions become progressively harsher and more mandatory if a bank does not turn around and continues to deteriorate financially. It uses a carrot-stick approach. The carrots are greater powers, greater freedom if you're financially sound. The sticks of course are the sanctions that we've talked about.

PCA requires careful and continuous surveillance and monitoring of banks and effective intervention. If this fails, the banks become insolvent. So now I'm going to move to resolving insolvent banks.

You need to close the bank legally but generally not physically. So there's a difference. When we talk about closing banks, there's a big difference as I'll emphasize throughout the paper, between legal closure which means terminating the interest of the shareholders and changing senior management, and physical closure, which of course is what people think about that you just close up the teller cages.

So what we want to do is close the bank legally but generally not physically at least cost. That is, you want to wipe out the interest of the shareholders, changing senior management. In addition, if there's negative net worth at the bank when it is resolved, it's best practice to have the government share the loss with the de jure uninsured depositors and creditors. But the bank is generally kept operating through recapitalization, merger or temporary government ownership through a bridge bank. Best practice requires minimizing two types of losses, one, credit losses, which are the negative net worth of the institution, and two, liquidity losses to depositors which occur if you freeze or block access of the depositors to the market value of their accounts. And in some countries the freezing part, which we in the United States do not really experience, is the greater fear of bank failures than the credit loss part.

Let me start off with credit loss. You want to minimize credit loss by closing the bank legally on a timely basis as soon as its capital or net worth approaches zero, and it's best to do so at some pre-announced positive value. If you are successful in closing the institution before its capital turns negative, then the losses are limited to the shareholders only. Of course, this is not only true for banks, it's true for any firm, that if you resolve it before its net worth reaches zero, the losses are confined to the shareholders and not to the creditors.

In the United States we can resolve banks quickly because a bank is failed by the regulatory agency. It does not go through the normal corporate bankruptcy code, although bank holding companies do. The sooner a bank is resolved after it becomes insolvent, the smallest the credit loss, and the smaller will be political opposition to resolving or failing the institution. If there is delay and the credit losses become large, then there's a probability or an increased probability of political intervention and there's a call for guarantees. If there's more than one bank failing there may be a call for blanket guarantees, something I'll talk about in a few seconds, and for large banks, too big to fail.

Let me turn now to minimizing liquidity loss, and this of course is achieved by not freezing the accounts at the insolvent banks. One has the government or the deposit insurance agency, as in this country, pay the de jure uninsured or de facto unprotected depositors an amount equal to a conservative estimate of the present value of the recovery value of the bank. If you have small credit losses, say, less than 5 percent of assets, the unprotected, which are generally the large depositors, would get 95 cents on the dollar the next business day at a new or assuming bank. This they, the depositors and society, can live with. De jure insured and de facto protected depositors get par value the next business day. If the liquidity loss is large for either the insured or the uninsured depositors, then there will be political intervention and call for relief.

It's important to note that deposit insurance differs from other kinds of insurance, life and property insurance, for example, where payment is delayed. When you have a car accident the insurance agent is not standing there giving you a check. With deposit insurance, that is important that that happens, and because we do this in the United States, our liquidity losses have not been a major problem here. One should try to minimize both credit and liquidity losses as much as possible, given the institutional structure of each country. Blanket guarantees are admission of failure in crisis prevention and reflect the lack of pre-crisis planning and surveillance.

Let me now turn to the history of too big to fail in the United States, which I think is important and interesting for a number of reasons. One, we have a large number of banks in the United States. Two, we have good historical records; if nothing else, banks always kept records that are available. Three, we effectively have no state-owned banks, which I'll fondly call SOBs. Too big to fail is a misnomer. It doesn't really mean what it says. And lastly, we have had major reform attempted in the FDIC Improvement Act in 1991.

In the United States, with very rare exception, insolvent banks, including large banks, are failed and closed legally in terms of what I said, the shareholders wiped out and senior management changed, but they're not closed physically. They're kept operating through recapitalization, merger, or a temporary bridge bank organized by the FDIC. But the failure has not always been on a timely basis, so the credit losses often have been larger than necessary. Too big to fail in the United States is a question not of failure, which we do, but of which depositors and creditors we protect. In the United States we've had no very big banks become insolvent before the 1970s. In 1974 the bank regulators kept open the insolvent Franklin National Bank in New York, which was the 20th largest bank, and in 1980 they kept open the First Pennsylvania National Bank in Philadelphia, which was the 23rd largest bank in the country and the nation's oldest bank. And they kept them open through providing financial assistance through the Fed's discount window and loans from the FDIC.

The Franklin National Bank eventually failed although the depositors were protected. The First Pennsy recovered. It had taken an interest rate risk in the late '70s. It was betting interest rates were going to fall as they were rising, and they of course lost. When interest rates fell then in the early '80s, it recovered. This recovery, amazingly enough, meant not only no loss to the FDIC. But it actually had a gain because as part of the financial program it accepted warrants which it was able to sell back to the bank once interest rates fell.

But the term TBTF or too big to fail really started with the insolvency of the Continental Illinois National Bank in my hometown of Chicago, right across the street from the Federal Reserve Bank of Chicago to which I'm a consultant, so I watched it day by day in 1984. The Continental was the 7th largest bank in the country, and more importantly was the largest correspondent bank with more than 2,000 inter-bank connections. There was a great fear then of contagion. The FDIC injected capital into the parent bank holding company, which in turn was down streamed to the bank and up streamed again in the form of dividends to the holding company. This kept both the bank and the holding company solvent, operating, and protected all depositors and creditors of both the bank and the bank holding company. But the FDIC became the major shareholder in control and changed senior management. Continental Illinois was effectively nationalized. The old shareholders got claims against the residual value of nonperforming loans, which turned out to be zero five years later, and their claims then were wiped out.

Through 1991 the concept of too big to fail narrowed. Bank holding companies and banks were both legally failed and fewer uninsured depositors were protected. By 1991 too big to fail really meant fail, but too big not to protect all third-party uninsured depositors, although not other uninsured depositors or creditors of the bank or bank holding company.

But too big to fail proved very costly and all banks were not treated equally and so it became unpopular in Congress and the FDICIA legislation effectively prohibited the FDIC from protecting any uninsured claims in costly resolutions or resolutions where there were loss to the FDIC.

But as always, there are exceptions, and the exception is that there could be protection of uninsured claimants if not protecting them, quote, "Would have serious adverse effects on economic conditions or financial stability and [protecting them] would avoid or mitigate such adverse effects." In other words, too big to fail was transformed into systemic risk exemption. But the systemic risk exemption is not easy to evoke. There are a number of ex ante and ex post barriers. Three important ex ante barriers are: one, it has to be a recommendation in writing by two-thirds of the board of the FDIC and two, it has to obtain two-thirds of the Board of Governors of the Federal Reserve. This recommendation has to be approved by the Secretary of the Treasury after consultation with the President of the United States. And friends of mine who were involved in writing this legislation tell me that consulting of the Secretary of the Treasury with the President of the United States does not mean catching the President on his way to another meeting and saying, "By the way, we have this bank in Chicago that's failing and we need to protect all the depositors." It means sitting down and doing a thorough analysis.

And thirdly, there has to be written notice, if the Secretary of the Treasury accepts the recommendation, to both the House and Senate Banking Committees.

Ex post there are two barriers. One, a review by the General Accounting Office of the rationale and the actions taken, and I think most importantly, if there is a loss to the FDIC, which is likely, it has to be repaid quickly through a special assessment on all banks, so that the banks would be paying to keep one of their competitors in business.

So far there have been no systemic risk exemptions. Uninsured depositors and creditors have shared in any loss with the FDIC in all bank failures. But no big banks have been in trouble. Therefore, we haven't had a real test yet.

On the other hand, Fed Chairman Alan Greenspan has said publicly on a number of occasions that no bank is too big to protect its uninsured depositors or creditors either fully or partially.

I think it's important to point out that the Systemic Risk Exemption does not require full protection. It could mean partial protection, so you still would be able to maintain some market discipline. In other words, if you catch the bank too late and credit losses are very large, you could still protect the depositors up to some extent without not giving them a haircut or sharing the loss.

Let me now turn to the conclusions. One must follow up and build on the surveillance programs that the IMF has or the country itself has with action to prevent costly bank problems. This requires, I've argued, three things. One, target best practice. Two, advance planning. You need time to do both the rehabilitation and the resolution, particularly for big banks. Unwinding a big bank is not easy and you cannot do it overnight. You need the availability of plans which are important for effective intervention and implementation. Unfortunately, a recent OECD survey showed that no country has such plans in place at this time. And lastly, one has to tailor the best practices to the institutions of the specific country.

Thank you.

MR. HACCHE: Thank you very much, George.

Our third speaker, Larry Promisel, joined the World Bank in 1998 and is Director of Global Partnerships in the Bank's Financial Sector Network. In this position he is responsible for the Bank's relations with a wide range of partners working in the financial area including the IMF and the International Regulatory and Supervisory bodies. He is co-chair of the Bank/Fund Financial Sector Liaison Committee, and takes the lead from the Bank's side in implementing FSAP. Before joining the Bank, Larry worked for 30 years at the Federal Reserve Board including as Deputy Director of the International Finance Division. He was responsible for the coordination of all technical assistance provided by the Federal Reserve System, two foreign Central Banks, and he also chaired a BIS Working Party that produced a report in 1992, the so-called Promisel Report, which was one of the first official analyses of systemic risk.

Larry?

MR. PROMISEL: Thank you very much.

The Financial Sector Assessment Program (FSAP) began in 1999, largely in response to the financial crises in Asia. Reflecting the origins of the program, considerable emphasis has been put on stability issues -- that is, short-term vulnerability issues. But this is only a relative statement. Development issues -- that is, longer-term structural issues -- have also received significant attention.

Also reflecting the program's origins, priority has been given to assessments of systemically important countries. But this, too, is a relative statement. Countries that are not systemically important, from a global perspective, have not been ignored. For example, fifteen countries in Africa have participated in the program already. Among them, probably only South Africa would be deemed systemically important, although others are important in a regional context.

Indeed, there is only a handful of countries remaining in Africa where an FSAP assessment along the lines we have been doing would seem to be warranted. For the other countries, in Africa and elsewhere, with less-developed financial systems, the emphasis in the FSAP must change if it is to be most useful to national authorities. While still analyzing why markets or institutions might fail, the analysis more fruitfully should emphasize which markets and institutions are missing or underdeveloped and how residents can get better access to a wider range of financial services.

The information base for forming the development assessment under the FSAP overlaps with that of the stability assessment. Observing many of the standards set for bank supervision, securities, insurance regulation and financial market integrity will also help ensure that all elements of the financial sector are functioning efficiently with a good flow of information, professional management and the kind of credit culture that helps build a deep and effective financial system. At the same time, the formal assessment methodologies that help FSAP teams in assessing stability issues do not exist for how to address development issues in the FSAP.

Over the next ten minutes or so, I will put forward some of the development-related issues that I think might usefully be addressed more thoroughly in the FSAP. These issues, which were presented in a paper for the Fund and Bank Boards earlier this year, fall into five categories: infrastructure for access; monopoly power and related distortions; nonbank financial intermediaries and organized markets; specialized institutions; and corporations and households. I will not address other areas, including taxation of financial intermediation, that are also important.

Infrastructure for Access. Since access to credit is one of the keys to balanced and sustainable growth, unleashing enterprise and innovation in any economy, the conditions under which providers of funds can do so with confidence are needed. This requires a well-designed and well-functioning set of collateral and bankruptcy laws; competent and impartial courts, adequately funded and free of political interference; and an information infrastructure. The needed information infrastructure itself has several components. It will include (for the larger companies) coherent accounting rules and a competent and credible auditing profession; and (for smaller concerns and individual entrepreneurs) a credit information industry. Some of what is needed to support the contractual environment is legislative; some involves the development of a private industry (auditing and perhaps credit information). Human capital and an enabling regulatory regime will be required.

So far as the legislative and regulatory components are concerned, there is a considerable overlap between the information that is gathered in the course of the various standards assessments and the information that is needed to make a diagnosis of the condition of these elements of financial infrastructure. Probably more can be done to incorporate this information in the development assessment.

Credit information and similar types of financial ancillaries, such as credit rating agencies or equity analysis, represent an area that appears only tangentially in the standards and codes, or in the stability assessment. But these deserve attention from the development perspective. A stand-alone private credit information industry is probably in its infancy in many countries; the status, performance and potential of that industry need to be looked at, at least superficially.

Enhanced communications and information technology are increasingly being exploited in the financial sector. They hold out great potential in terms of access. To realize that potential -- with due regard to security concerns, for example -- will require special legislative and regulatory initiatives. In this area, and in others, FSAP teams need to assess the capacity of the government to perform a proactive catalytic role to make sure that all of the elements do in fact come together.

Monopoly Power and Related Distortions. In a protected or distorted environment, participants in the finance business can often flourish without any risk of failure, even though the sector may not be providing the range and quality of services the economy requires. One characteristic of such systems is a tendency toward financial sector monopolies. A small financial system dominated by a handful of institutions with substantial market power can in some cases be a stable model, yet it may be sorely deficient in terms of allowing local non-financial firms to grow and compete in the local and international market. A complacent oligopoly becomes sated on the low-hanging fruit and makes little effort to enhance service quality or pricing, or -- even more important -- to find ways of doing profitable business with more difficult clients such as small and medium-size enterprises. Monopoly and the wider issue of the contestability of markets, including protection of dominant domestic financial institutions, are thus important development issues. Similar issues relate to taxation of financial intermediation and to privileges for state-owned banks. Indeed, the question of privatization policy is one on which stability and development concerns may seem to go in opposite directions and may need to be reconciled.

Concentrations of power are not always easy to detect. Concentration ratios for banking, for nonbank financial sectors, and for the non-financial corporate sector are only one indication. Apart from being alert to indications of collusive behavior, assessors need to be aware of cross-segment ownership structures: do the main banks own the main stockbrokers, the main insurers, the main fund management companies? Is the design of legislation and regulation dominated by a bank-centric central bank? These are subtle issues, but important for development.

Nonbank Financial Intermediaries and Organized Markets. Although insurance companies and other nonbank financial institutions are prone to failure, and stock market volatility can transmit itself to the rest of the economy, the smaller role these businesses play in developing countries and the fact that the systemic implications of bank failure appear to be greater, have meant that nonbank financial institutions and markets are not typically the focus of stability assessments.

Yet nonbank financial institutions have been characterized as a spare wheel that can prevent excessive reliance on banking from making a financial system vulnerable. A healthy nonbank financial sector increases the diversity of financing instruments, allows ventures that for one reason or another would never be bankable to obtain finance and get off the ground, and can provide some competition for the banking system. Thus, development motives call for a strong nonbank financial sector.

This has sometimes been thought to warrant costly initiatives to create stand-alone national markets, such as a stock market, in circumstances where those markets might not be economically viable. It has also been held to justify a monopoly for state-owned insurers. And the role of the social security system is central in influencing the development of the nonbank collective savings and fund management industries. All of these need attention with national development in mind, with the focus on functions rather than on institutions.

Once more a catalytic as well as regulatory role for government will be needed, with some limited potential for operational responsibilities by a state-owned or, more likely, regulated monopoly, as with securities clearance, settlement and depository functions.

Specialized Institutions. A long tradition in development finance has been the establishment of specialized and subsidized institutions for filling supposed gaps in provision of financial services. Current development thinking tends to minimize the role of subsidized institutions of this type as players in the financial system. Subsidies can inhibit the emergence of market-driven financial service firms that would deliver services to the small-scale and to the poor, or inhibit the existing banks from entering that market. Directed credit and selective credit controls also cause distortions.

Nevertheless, specialized financial institutions (whether subsidized or not) do exist with the mission of serving particular segments of the economy traditionally underserved by the mainstream banks. These could include agricultural development banks, low-income mortgage banks, credit cooperatives and similar entities. They are often supervised separately and have a quite different management approach and may function under different fiscal conditions. Specific issues affecting specialized micro-finance institutions may need to be included in an FSAP, including the question of whether or how to regulate such entities cost-effectively and with a sufficiently light touch so as not to crush them.

Corporations and Households. Knowing the financial condition and ownership structure of the nonbank corporate sector helps the vulnerability assessment and throws light on whether corporations have access to finance that they need in a development context. From the stability point of view, looking at just the major companies may be sufficient, but answering development related questions seems to call for a more comprehensive information base going beyond those companies that are listed on the stock exchange. The availability of information on unlisted corporations differs widely across countries. Collecting and processing the information can be time consuming, and to do it properly may represent a major investment of resources in the collection of statistics, which could be wasted if not properly integrated into an institutional development effort with the national statistical authorities. One task for FSAP teams is to judge how much effort should be put into this, given competing priorities within the overall assessment. A further question to be addressed is how reliable and useful is the accounting information that can realistically be expected from even the middle-size firms.

Likewise, to understand how well the financial needs (not just credit) of small firms and lower income households are being served may require survey-based information, which is hardly practicable in the timescale over which the FSAP is conducted. FSAP teams need to look for other sources of information to fill these gaps.

Looking forward, we need to enhance our analysis of development issues in the FSAP. The extent to which we have not dealt adequately with development issues in the past does not, I think, reflect a lack of theories or an analytical framework. Partly it reflects trade-offs, in the face of a time constraint. More substantively, it may have more to do with the diverse nature of the counterparties involved or to the shortage of pre-assembled data. Indeed, a major problem is how to gather needed information and distill it in the short time typically available to an FSAP team.

This brings me to a concluding thought. The FSAP is just one diagnostic tool, albeit one that has become central to the work of both the IMF and the World Bank. It identifies issues that need to be addressed and helps national authorities set priorities for addressing them. We are currently engaged in an effort to improve the FSAP, so it can identify and prioritize more clearly a broader range of development issues. But it is in the follow-up to the FSAP that the real work needs to get done: the data gathering, the in-depth analytical work, the technical assistance that the World Bank, IMF, and others can provide, and -- most importantly -- the actions of national authorities and the private sector to strengthen their financial systems.

Thank you.

MR. HACCHE: Thank you, Larry.

Our final speaker this afternoon, Roberto Zahler, is president of the consultancy firm Zahler & Company based in Santiago, Chile. He is also a board member of Banco Santander-Chile Bank, chairman of the Advisory Board of Deutschebank America's Bond Fund, and a member of the Latin American Shadow Financial Regulatory Committee. Between December 1991 and June 1996, Roberto was president of the central bank of Chile, having previously been vice president. His tenure at the newly independent central bank of Chile was marked not only by important financial sector developments and reforms, but also by record economic growth and success in avoiding major spillovers from the tequila crisis of '94-95. Among his many other activities, he has been a consultant and visiting scholar at the IMF. He has a Ph.D. in economics from the University of Chicago.

Roberto?

MR. ZAHLER: Thank you very much.

Well, let me start by saying that FSAP in the short period of time since it existed, less than five years, I think has had a very good result since most countries are incorporating most of FSAP's findings and, furthermore, there is an excess demand for FSAP missions to go to the countries.

Now, actually the issues that they are dealing with, which is to try to, attempt to reduce the likelihood of severity of financial crises everywhere in the world, especially in emerging economies, I would say is extremely important.

Now, I will have some five or six general comments, and then I will have some five or six more specific comments.

My first general comment relates to the fact that when one reads some of the FSAP reports, one finds that there are really a very broad range of issues, some of them very much related to short-term macroeconomic stability sort of concerns, typically, let's say, from the IMF perspective, and others from what we just heard from a more developmental sort of approach.

Although in my opinion there's many aspects of synergies in both approaches, I really tend to believe that methodologically one has to distinguish very clearly both types of approaches, and I think there is a challenge there regarding in the future how to deal with even one alone sort of report or if you should divide it, because clearly the methodologies and analytical scheme to analyze these are, at least in my opinion, quite different.

Now, having said that, a second and perhaps more substantial comment is the fact that FSAP tends to consider as a given the international context in which the analysis takes place. For example, in one of the papers which we received it says very explicitly there, "High short-term borrowing in foreign currencies has been associated with many past crises in emerging economies. High readings on such indicators that have signaled crisis in the past may suggest the need for remedial measures."

But these remedial measures in the FSAP reports relate only and specifically to individual countries. In those reports there doesn't appear almost anything related to the international context. For example, the Basel Committee, the proposed Basel II capital accord requirements shortens even further the current, already present in the capital accord of Basel, bias in favor of short-term financing from industrial countries to non-OECD countries. This is clearly one element which tends to increase vulnerability, volatility, pro-cyclical behavior of banks in industrial countries, which clearly is to the detriment of emerging market economies. So I think that this issue of the international context is, I would say, very important to consider, and not only take it as a given.

Now, let me give you a second example. Take the case of Brazil between May of 2002 and November-December of 2002. That was the period when most surveys, most polls of what was going to happen in the presidential elections in Brazil pointed out that Lula would become President and markets became extremely nervous, especially international markets, and country risk for Brazil went from something like 600 points to 2,400 points in less than two months. And that required also very high increases in interest rates domestically. And we are today seeing the problems in terms of very, very low growth in Brazil, which hopefully will increase substantially next year.

The point I want to mention is the fact that Brazil was complying with all sorts of IMF conditionality. It even over complied in terms of primary fiscal surplus and other indicators. And in spite of that, they received a huge problem related to the way in which international financial investors behaved.

My reading of this is that there is something missing in the way in which international financial institutions are dealing with the international capital markets in relation to countries that individually do comply with most of the criteria of conditionality and programs of the Fund, Bank, and other international financial institutions. So I think that in general terms there is something to do relating to the working of international financial markets which also are very, very relevant to financial vulnerabilities at the individual level.

The third general comment I have is that I tend to see that in the aftermath of the Mexico, Asian and Russian crises, many people tend to believe that the way to try to reduce vulnerability in emerging market economies is by having a more solvent and more liquid and more healthy domestic financial system. I, of course, have no disagreement with the fact that a better financial system is much better for the country. But I would not overdo the argument that with that we are almost at the other side of the road, because, first of all, not all international capital coming into a country is intermediated by the domestic financial system. In many cases, most of this capital coming into a country goes directly to firms, to non-financial firms, and some of them go directly to the stock exchange. So even though you may have a quite solid and well-functioning financial system, you may still have for different reasons vulnerability at the macro level because too much capital may be coming in of a short-term nature, and because there is a wrong pricing of exchange rate or other sorts of risks in the way in which international investors are pricing the risk in domestic--excuse me, in emerging market economies, and, therefore, that can generate a problem in a total independent way of what's happening in the financial system.

So I think that one has to be aware that, for example, the way in which one looks at macroeconomic equilibrium has to take care of indicators that still are very much related to the way in which a country can become vulnerable to foreign shocks.

Perhaps the most important thing I would like to mention here is that when you look at some crisis like the Chilean crisis in 1982 or the Mexican crisis in 1994, the fiscal sector appeared to be quite healthy in both cases, but there were clear contingent fiscal spending associated to serious deficiencies in some areas of the economy which were highlighted once the crises erupted, and the crises erupted for macroeconomic reasons and not directly related to the way in which the financial system was behaving.

A fourth general comment which I would like to make relates to the fact that in one of the papers that we were given, they say that crisis countries are not deemed appropriate countries for FSAP. And I tend to agree clearly that that's not the moment where you have to deal with--you're not going to go like for an emergency sort of situation. You have to deal with it in a precautionary way. But I do think that countries that are in a crisis are the typical case where my thinking is that the type of development sort of approach of reforms that you require in a country, that's a very good moment to deal with them. Sometimes it's very, very difficult to try really to deal with reforms that are needed in countries, especially in the capital markets, in privatization, if the country is not in a real, real mess. So I would agree with arguments that say the IMF approach but not from the point of view of what would be the World Bank.

Finally, I have two more general comments. One relates to the fact that because of the very different nature and very ad hoc nature of financial sectors in different countries, what happens is that when you try to use FSAP reports to compare among countries--and at least until today, that is not very easy to do because of these ad hoc characteristics, and so it's really for an international comparison--it's still not very useful. And although I think that is a structural problem there, I hope that in the future that can be proved.

And the final point which I would make, at least in many emerging market economies, is that there has been a recent trend in the last, I would say, ten years of a very important component of foreign participation in capital markets and in the banking system. And that by itself I think introduces some new positive aspects and also some challenges, which until now, at least in the reports I've seen in the FSAP, don't appear very explicitly, which I think should be considered because that's a trend that is going on heavily, and that would--I think that has some important systemic implications for emerging market economies.

Now, let me concentrate briefly on a couple of specific comments I have on FSAP. One has to do with capital market developments. I would say that one, perhaps the main problem--there are two main problems one finds practically across the board in emerging market economies related to capital markets. One is the fact that you don't find domestic long-term capital markets in these types of economy, and that is a major, major problem which one country which has dealt with, I would say, in a very successful way is Chile, and that has been related basically to the financial--excuse me, the pension fund reform and some generalized indexation in the financial system. I'm not saying that these two things should work together always, but I'm saying that very clear-cut reforms or institutional characteristics have been the--are in the source, are in the basis of these long-term domestic capital markets, which I would say are one of the major, major challenges for economic development in these type of countries.

The second major problem that we still have--and we have it in Chile, and I would say practically all countries--is the lack of access to financing, formal financing from the point of view of small and medium-sized enterprises, and there honestly I don't have any suggestion of how to deal with that because, really, there are many, many problems related to how to evaluate risk of these type of enterprises, and I still think that we are very far away from how to deal with that problem.

Now, a couple of other elements which I would like to mention here which perhaps could be considered in future FSAPs.

On stress testing, for example, which I think is a very, very important component of FSAPs, stress testing takes as a given foreign shocks, for example, exchange rate shocks. Now, you have to be very careful when in some countries the domestic financial institutions are becoming important players in the sense of being more open to the world economy, and they themselves may be a source of exchange rate vulnerability or exchange rate changes.

To give you a very concrete example, in Chile pension funds, which can invest nearly 20, 25 percent of their total assets abroad, were a significant source of exchange rate vulnerability in 1998 and 1999 because they themselves, because of their size and their importance in both the financial and exchange rate markets in Chile, they tended to generate a very important appreciation of the local currency due to their behavior, a rational behavior, let's say, from the point of view of the institutions themselves, but a complicated behavior from the point of view of their macroeconomic impact regarding some very crucial key macro prices.

So I think that when one does stress testing, it's very important not just to consider the variable exchange rate or the interest rate or the asset prices that they change only by themselves, but some of them could be induced by some problematic working of their own domestic financial institutions.

Another element more specific which I think could be analyzed from the point of view of FSAP is the fact that the international financial institutions by themselves generally lend in foreign currency to emerging market economies. The point is which--or one element which could be analyzed and which would really be very important to improve the possibility of having domestic capital markets, profound domestic capital markets, is if the international financial institutions could lend in domestic currencies, in currencies of the emerging market economies. There is an issue of cost there, of pricing, but clearly there is--apparently there exists in the markets today some desire to buy assets denominated in local currency of some emerging economies, and that perhaps could be improved and could be quickened by an appropriate change in the way in which IFIs are dealing with this sort of lending.

Perhaps the final comment I would like to make--I have two final comments. One relates to the fact that when one reads some of these reports, there is a clear concern regarding the lack of liquidity and the lack of profoundness of stock markets and fixed-income markets in emerging market economies.

Now, one thing that has been happening in many of these markets is that because firms are being allowed to issue bonds and issue stocks abroad, increasingly most of these transactions are being done in the New York Stock Exchange market or outside the emerging market economies, and that by definition reduces liquidity and reduces the size of the domestic markets, and that perhaps is a positive development. The point I want to mention is that it may be the case that not for regulatory reasons but because of economies of scale or for other efficient ways in which markets are working today and the increasing internationalization, we do not need to have specific, let's say, as I mentioned, liquidity and profoundness in certain types of markets which may be better functioning from abroad.

The final comment I would like to make is the fact that regulatory arbitrage is something that is complicating the lives in many emerging market economies. As was mentioned here, some of the supervisors don't talk among themselves, but I would say that as serious as that is, the criteria with which some of the supervisors allow a bank to come in or allow a pension fund to come in or allow a stock to be issued are different among the different supervisory or regulatory agencies. Then that tends to create regulatory arbitrage, and in some cases that has been a source of major complications in the working of the financial system.

Thank you very much.

MR. HACCHE: Thank you very much.

I'd like to give the members of the panel a chance later on maybe, if there is time, for them to react to what their colleagues have said. But, first, we'd like to take questions from the floor. Those of you who ask questions, I would ask you to make sure that your microphone is switched on, and I would ask you to state your name and affiliation, please.

MR. HACCHE: Questions? At the back there, please?

QUESTION: I'm interested in Basel Core Principle 1.2, which means banking supervisory agency independence. And I would like to ask you how to measure this banking supervisor independence across countries, because sometimes developed countries, even legally, are not going to have the authority, but they never intervene, in fact, in terms of banking supervisor independence. But for emerging markets, even they have, for example, central bank independence on the paper, but they may be intervened by a lot of political influence. And then one of the divisions of central bank is in charge of this banking supervisor, they might be necessarily independent, not from a legal perspective but in fact they may not have independence. So how do you measure when you do FSAP analysis in terms of this dimension, and to the extent current evaluation is going to be comparable across countries? And this is my first question.

And my second question is now I know that a lot of emerging markets are trying to create a new type of financial supervisory banking agency, and based on your experience, do you have any suggestions what type of agency, for example, one banking supervisor agency like the U.K. or Japan going to be more appropriate for emerging markets or like U.S. type? A lot of institutions are evaluating financial systems in charge of different functions going to be more suitable for emerging markets, and do you have any concrete policy suggestion for this, a lot of ongoing financial reforms in this respect?

MR. HACCHE: Thank you. Perhaps we could collect another one or two questions, if there are any.

QUESTION: I chair a coalition of academics and NGOs called New Rules for Global Finance, and I wanted to go back to some of the rules that the Fund implements, and I wanted to ask what role and what transparency there is in the process of setting rules, say in the Basel Committee, how many developing countries have a voice there? What about the Financial Stability Forum? How are especially the smaller, poorer developing countries represented, comparable rulemaking bodies? I'm asking about their democracy especially from the perspective of the smaller developing countries.

MR. HACCHE: Thanks very much.

Another question? In the back?

QUESTION: How would you trade off the need for evaluation of investment opportunities, say, you know, in the stock market or even venture capital type of opportunities moving abroad when perhaps they're best made locally, domestically? For instance, you might have the best ability to evaluate projects say in Chile or in some other emerging market economy, and the fact that there is this tendency that you noted for stock markets, at any rate, to become more and more international. How do you trade that off, those two things off?

MR. HACCHE: Thanks very much. Should we take those questions? Who would like to start?

MR. INGVES: Thank you. Let me give it a try.

On the independence of supervisory agencies, it's one thing to read the laws and see what is in there, but when we do the work in the field, that also includes discussing with all sorts of market participants and the banks and the authorities themselves. In the end, when we produce the assessments, what comes out of it is a combined opinion, a combination of all the information that we get. But in doing so, there is, of course, always an element of judgment when you do that.

On the issue of what kind of advice we give when it comes to supervisory agencies, one reflection I have, having dealt with a large number of countries in crisis of various types, is that regardless of what organizational structure you have inherited, people always want to change things after a crisis. So if you happen to have supervision in the central bank, there is the tendency to move it out of the central bank. And if it wasn't in the central bank, you have a tendency to move it into the central bank.

Having said this, this has led us to the conclusion that it's not possible today to come up with a kind of one-size-fits-all advice saying that there is one and only one way of doing this. This really varies from country to country, and it depends very much on the preconditions in the countries in question and how they have done--what they have done in the past, whether the authorities are willing to talk to each other or not.

What also matters quite a lot is the number of markets that you happen to have, because if you only have a banking sector, then it's likely that the insurance supervision is going to be tiny. But if the insurance sector is equivalent in size to the banking sector and if it's a very large country, then you end up with a very different organizational structure.

In addition, when it comes to developing countries, in some of them the central bank is one of the few institutions from an organizational point of view with the capacity to carry out supervision, and also with the capacity to finance the supervision. And in those cases, it usually pays and it makes more sense to have supervision within the central bank. And this basically means that you just have to look at what you find in the field, and then you discuss, we discuss with the authorities, and come up with a solution which seems reasonable at that moment in time in that particular country. It just simply does not work to say that there is one sort of best practice that covers everything because then that's not the way the world works.

On the issue of the Basel Committee Financial Stability Forum and how they produce their standards, it may be up to the Basel Committee more to answer that question. What I know is that they do much more than perhaps in the past in circulating their documents and have them commented on extensively, much, much more extensively than in the past. And that also holds for the work of the Financial Stability Forum and the way they are dealing with financial stability matters.

When it comes to the projects that we deal with within the Fund, the way the Fund works, the papers we write always go to the Board, and they are endorsed by the Board. Now, since our membership covers practically the entire world, all countries get a chance to comment on the papers, if they want to do so, as part of the process. And one example I can give you in which I was deeply, deeply involved, was when we produced what is called the Central Bank Code on Transparency, which came into existence about four or five years ago, and that is a document which was extensively discussed by everybody, and it was also distributed to all our members, and all those who wanted to comment on it had a chance to do so.

Thank you.

MR. HACCHE: Does anybody else have anything to add on those questions before we turn to--yes, Larry?

MR. PROMISEL: Yes, let me just add one thing on two of those questions. On the representation in the policy-making, the standards-making bodies of the Financial Stability Forum, the World Bank and the IMF tend to participate in many of those bodies--not all of them, I think, and not always as full members--and we view it as part of our function. And one of the reasons we do participate is to represent the interests of a much broader constituency than is represented directly on those bodies. So that when we comment to the Basel Committee on their proposals, we do so from the perspective, in part, of the countries we represent, and we represent virtually all countries in the world. And, arguably, it's in the interest of these bodies to use that rather than having such a large group of countries present or people present that it makes it difficult to operate effectively.

I fully agree with Stefan that no one-size-fits-all on the structure of a regulatory regime. It seems to me what's important is that the various supervisors and regulators involved cooperate with each other and talk to each other. That problem exists whether you're talking across agencies or within an agency. The fact that in the U.K. there is one body doesn't mean that the insurance people talk with the banking people any more than it would be the case necessarily in countries where they're separate. And these supervisors and regulators have different objectives. The objectives of a securities regulator are not the same as a bank supervisor, and it's a different perspective, different expertise.

So I think it really does depend on the whole host of historical, political, other factors as to which makes sense. But the fundamental problems of cooperation across sectors exist regardless of the structure of the supervisor.

MR. HACCHE: Roberto?

MR. ZAHLER: Perhaps to follow up what I mentioned before, what one tends to see is the fact that in most emerging market economies, the stock exchange, the liquidity, the number of transactions has been going down--at least during the '90s and in the recent years. And one finds basically three types of reasons for that. One is, as I mentioned, the fact that most of the listed enterprises in the domestic stock exchanges are the same that can issue abroad, what happens is that obviously the international capital markets are much more profound, much more liquid, even much less expensive. So, clearly, transactions tend to go outside.

There is a second reason that in many countries capital gains taxes have been changing and have in a certain sense changing for the worst in terms of not stimulating transactions domestically because of the capital gains tax.

And there is a third reason related to concentration. Both suppliers and demands of stocks have concentrated quite significantly in many of these countries due to different sorts of mergers and acquisitions both in the pension funds and the insurance company on the one side, and on big firms that issue stocks on the other.

So I would say that these are in a sense some sort of structural reasons which there may be a trade-off or not, but that's, I would say, part of life. One has to live with that.

But having said that, most of what I would call venture capital enterprises or small and medium enterprises, they are even not listed domestically. So they are in another league, and one can try--and there have been some experiences of capital market reforms in some countries that have gone in the direction of generating taxes and other sorts of--inducing firms to get listed and allowing, for example, for pension funds to relax some of the criteria with which they can buy certain instruments so that they can buy also this new more entrepreneurial, more modern sort of enterprise. And that is sort of a market that is starting to develop, although still very gradually.

MR. HACCHE: Thank you. If there are no more questions, maybe I can turn again to the table and ask if there are any last words in the last few minutes. Larry? George? Roberto? Stefan?

MR. INGVES: Very many good suggestions have been made when it comes to how to carry out this work, and that's a good thing because that shows that there is a keen interest in sort of thinking and arguing about how you do these things when you visit a country for two to three weeks and start digging into all the different technical aspects, of how you put together a financial sector in a country in such a way that that financial sector ultimately produces something for the people in that country. So there is more fascinating work to be done, and there is a huge demand out there.

Two comments among many I could pick--there were so many of them, but just let me pick two of them. Professor Kaufman pointed out the whole issue of planning and to think ahead, and what we do when we do the FSAPs is that we kind of produce a snapshot picture of what things look like in a country at a given moment in time. But that also means that everything cannot be fixed within that time period. And there really, really is high value in planning ahead, and particularly planning ahead thinking about how you deal with issues if things really go wrong. In too many countries--and this actually holds in my own country, and when I was doing the work there--having to sort things out during an acute crisis, when you have not planned ahead, is something which is not very pleasant. So you'll always be better off if you try to do it in advance.

Roberto talked about stress testing and how you do stress testing, and that also has to deal with planning ahead, because it is very important to do the numbers and to understand the numbers. And it also is very clear from this project that we have undertaken in the past five years or so that stress testing is a very powerful tool. If there is a problem out there, the first question is always: What do the numbers look like? And if you don't know what the numbers look like, then you don't know what to do, you don't know where you're going, and it takes very much more time to figure out what's happening compared to if you had the information. So it's really, really high-value in doing that at home in the countries.

Thank you.

MR. HACCHE: Thank you.

Finally, let me draw your attention to the documents--there are a couple of documents on the table outside the room here that are available to you if you haven't picked them up already. And thank you very much for coming.

[Applause.]