Transcript

Leslie Lipschitz, Deputy Director of the IMF's Policy Development and Review Department, is also with us today, and has joined the briefing to handle any questions you may have about the staff's response to the External Review Panel's findings.

Tom Bernes, the Chairman of the Executive Evaluation Group will have some brief comments, followed by Mr. Crow's opening remarks, which will be distributed at the end of the briefing.

The briefing and the contents of the report are embargoed until 11:00 a.m., Eastern Daylight Time, today.

In addition to the external review of IMF surveillance, we are releasing today an "External Evaluation of IMF Economic Research Activities." Advance copies of that report, which is also embargoed for an 11:00 a.m. release, are available outside the auditorium and will also be posted on the web site.

Thank you very much. Tom?

MR. BERNES: Thank you very much, and good morning.

Today's press conference is a continuation of a process begun in the fall of 1996 when the Executive Board decided to establish a process of independent external evaluation, and at that time or shortly thereafter launched the first review which was an external evaluation of ESAF, conducted by a team of experts, led by Professor Botchwe, former Minister of Finance of Ghana. The results of that review, as well as recommendations for changes in the way in which the Fund designed and implemented programs in low-income countries, was made available to the public the following year, last year. While the process of responding to these recommendations is still underway, IMF staff prepared a stock-taking of progress in implementing the recommendations during the last year, and this is expected to be released shortly.

On the advice of the Board Committee on External Evaluation, the Board authorized two additional external evaluations. Today, I have the pleasure of participating in the release of the Fund's second external evaluation, this on the timely issue of IMF surveillance.

I would just underline that in selecting this topic, the Board wanted to look at what is at the heart of the Fund's mandate, the bread-and-butter issue. It is not, and it was not, designed to look at programs in individual countries. Instead, it was meant to examine how the process of review of individual economies was carried out, which is designed to keep countries--or hopefully help to keep countries--out of trouble as opposed to what one does when one encounters trouble in a particular country.

As Tom Dawson said, the results of the third external evaluation on the content and scope of IMF research, which was conducted by a team of experts headed by Professor Rick Mishkin of Columbia University and former Director of Research with the New York Fed, will be available later today. This is obviously on a much more focussed topic.

With respect to the surveillance evaluation, the Fund's Executive Board has been extraordinarily lucky in being able to enlist the services of three eminent individuals who collectively possess a rather impressive depth of experience. As a Canadian, of course, I am not referring to my colleague, Mr. Crow. I will leave it to him to discuss the contents of their evaluation, and to Mr.Thygesen.

Let me just say a few words about where the Fund intends to go from here following the Board discussion. After discussing the recommendations of the evaluators, the Executive Board has requested Fund management to prepare an action plan to respond to a number of the issues raised. The Board expects to see this action plan before the end of the year. Once approved, management will begin to implement whatever changes are deemed to be appropriate, and a date will be set 12 months out at which time progress in the implementation of the recommendations will be assessed.

On a more general note, you may be aware that when the Executive Board agreed to the present system of conducting independent evaluations, it was agreed that the system would be reviewed following the first three evaluations to determine its adequacy and to decide whether or not adjustments should be made to that approach. We will begin to address this question shortly after this year's Annual Meetings, and I hope will be in a position to present formal proposals certainly by the time of the spring Interim Committee meeting.

With those comments, let me turn the chair over to John.

MR. CROW: Well, thank you very much, Tom.

I was chairing the evaluation exercise with the able participation of Niels Thygesen and Ricardo Arriazu, and I would like to make some opening remarks about what we thought we were doing.

You have the report, and I trust you have had a chance to look at it. So I will try to be brief. I will be brief. I will make three sets of introductory comments, one on definitions, one on process, and the third on issues.

On definitions, while Tom Bernes has pointed to one area, I want to talk a little bit about surveillance, what it is and what it is not. It is loosely, but fairly accurately, to be thought of as advice-giving. It is not program lending, the point that Tom made. That distinction determined the kinds of situations and country experiences that we could or could not look at.

For example, in our terms of reference, we were explicitly asked to limit our sample to "countries that receive no or at most sporadic use of Fund resources." Specifically, while we did, for example, look closely at Thailand, our professional interests for this report ended in July 1997 when Thailand entered into standby negotiations with the Fund. The same would be true of a number of other countries in our sample. About a dozen countries were in the sample, in fact. Not all had standbys during this period.

As regards process, we had very full access, both in our designated--designated by us, that is--sample of countries and in the Fund where we probably talked to close to 100 people, rather than the "more than 50" that is mentioned in the report. We also met, as you would be able to see in the report, in the list of people we spoke to, with quite a few third parties. And also with regard to the Fund, we should emphasize that we also had full access to its internal documents and not just documents to the Board which also, of course, are, by and large, internal.

As regards the use of all this written and oral material, we tried to synthesize and report upon it in Chapters 2, Conduct and Methods; 3, Substance; and 4, Impact. We thought it important to report the range of views that we heard from what we thought were interested and informed people, even if we did not necessarily agree with all those views. We also added commentary of our own where we thought necessary. Our conclusions, the conclusions of the team, and the recommendations are in Chapter 5.

Let me just make a few remarks about the issues. Now, there are 29 recommendations, but let me highlight here three central sets of issues or questions for surveillance.

One, the scope of surveillance. What should surveillance focus on, and indeed, is it in danger of becoming unfocussed? We concluded that it should focus, above all, on the important systemic and international issues in the world financial picture, matters where the Fund has a significant comparative expertise. We also think that it is in danger of losing that focus without gaining so very much in compensation.

Transparency and confidentiality. What surveillance material should be published? We think that on grounds of principle, such considerations as accountability, the peer review process that is at the heart of surveillance, and therefore its political legitimacy, and for reasons of practicality--the consultation reports are leaked, anyway--the consultation reports should be published. This does not mean, however, there should be no confidential exchange between a member and the Fund in surveillance, but this exchange would have to do more with hypothetical situations, these issues that we have discussed at length in the report. I will not go further, but I would just point to some of the features of them here.

Finally, methods and governance. We address a range of issues in staff surveillance practice, and also discuss, as we were invited to, by the way, the role of the Board.

As regards the Board, the point worth making here is that there are two, somewhat different, sets of questions that we deal with. One is operational--how to lessen Board overload by making its overview more focussed through committee work. This question of focus leads into another-- deeper, in our view--governance question. Given that surveillance consultation is in the end peer review rather than staff review, this has to do with how the Board might best exercise ownership of the process and thereby ensure that each consultation exercise is focussed on what is truly important.

Thank you.

MR. DAWSON: Questions?

A QUESTIONER: I'd just like it if someone could try to sum up in a sentence or two whether at this point you think your surveillance is doing a good enough job and if there is anything specifically that might be done to change it.

MR. CROW: I'll just start on that because we were evaluating it.

We think that in terms of multilateral surveillance, that is, the material in the World Economic Outlook and the International Capital Markets Report, it does a pretty good job. We have some criticisms, but it is well regarded and well--let me leave it at that.

On bilateral, I think there are some more complex and difficult issues. One issue we see is bilateral and multilateral not being as integrated as well as they might be in the Fund.

As I noted in my opening remarks, we also worry about the scope of surveillance. We think it is a particularly important issue, how many things should surveillance reach into, where we have some concerns which we express in the report.

Beyond that, we raise questions about the oversight from the Board in terms of ensuring focus. We raise questions about how the staff could do it better. It is not that it is done badly, but we find areas where there is room for improvement.

Just let me leave it at that, I think.

MR. THYGESEN: In addition to discussing what could be done better in surveillance, of course, we do also reflect on the fact that we were able to establish in the countries we visited that surveillance is only one among many elements informing economic policy and not usually a central one. If it is good advice, it can slightly tilt the balance of the domestic political debate, but in our study of countries that were approaching crisis, we were struck by the fact that even when good advice was given, it was not followed in a number of cases.

MR. LIPSCHITZ: The evaluators are perhaps more generous than the staff in the evaluation of surveillance. I think the staff view is that there are substantial deficits in the way we do surveillance, and that there is a lot of ground we need to make up.

Let me try and be specific. In the whole area of transparency, that is touched on in the report, there is a huge amount of ongoing work where we are trying to figure out just how much more we can give out without making the surveillance process subject to impediments in our interactions with governments--how transparent can we be without encumbering the process.

In the whole area of coverage that John alluded to, how can we cover more of the international dimensions than the capital market dimensions of surveillance? In some sense, the international dimensions are covered quite automatically. In any bilateral surveillance, we look at the market growth of the country which depends on all its trading partners. When we figure out its interest rates, we look at the interest rate arbitrage opportunities in the markets. So that is done.

The much tougher part is how to deal with the hypotheticals. Let me try and be concrete. If I am doing surveillance of a country in southern Africa, how do I take into account what the effects might be of a devaluation of the rand on that country, or devaluation of the Chinese currency in Asia? These hypotheticals are much, much more difficult to deal with.

The former Prime Minister of New Zealand said something along the lines of speculating about speculation is so damn speculative it is not worth speculating about, and it is that sort of thing.

Hypotheticals are very difficult to deal with, and yet, we have to think of that sort of thing when you are going into a country and trying to warn about what might come out. It is that, I think, that we are wrestling most with, those capital market issues, those hypotheticals, how to deal with them.

There is a lot of work, internal work, going on along those lines, all under the umbrella of early warning systems. We are looking at a variety of variables that might give an indication of the vulnerability of a country to a number of occurrences that might occur, but have not occurred. That is where I think we are grappling. I am not sure we have the answers, but that is the area I think we think we need work.

A QUESTIONER: Just one--or actually two--questions, the first being that the noise about Mexico and the fact that perhaps some of the lessons that were seen in Mexico weren't necessarily applied in some areas. Is that more of a regional thing? Is it something that perhaps a broad outline of what was about to happen in Asia had been seen in Mexico and perhaps enough steps weren't taken to be better prepared for what did happen eventually in Asia?

And secondly, the talk about IMF clout and the fact that when surveillance or advice is undertaken that the IMF really doesn't have much in the way of clout to back it up and make sure that it is followed more closely or implemented in a better way or recognized as being good advice to begin with. So just two issues.

MR. CROW: Okay. I should be clear for everybody that Mexico wasn't one of our sample countries. But we did, of course, review the Fund material, the post mortem, if you wish, on Mexico, and the post mortem on the Asian crisis, and so on. And you are right, there are some similarities with Asia. They weren't quite as clear at the time as they came to be later. For example, the banking system: the Fund had recognized the banking system issues, but I don't think the Fund was that quick to seize upon the importance of this and wasn't that quick in Asia either. It wasn't completely out to lunch, but it wasn't that quick, that's for sure. Getting up to speed, finding people who are knowledgeable about these kinds of issues and the interrelationship with capital flows, and with implicit guarantees, and the exchange rate system, pegged rates, that kind of thing, does take time.

We also identified, however, in our judgment, that even making allowance for the time it takes to get these things spread out across the Fund, the Fund isn't that good at spreading the news or spreading the information. Our feeling was that in Asia, the lessons of Mexico hadn't been that well absorbed for the most part, and I think that's one fairly clear criticism we make. In the jargon, we refer to the issue of knowledge transfer. That's all I'll say on that for the moment, but others may have something to say.

As regards IMF clout, we point--and I know exactly where we do this, at the beginning of the final chapter--to what I think is a disjunct between what the rest of the world thinks the IMF can do and what people on the inside think they can do with surveillance. Remember, we are talking about pure advice here. We are not talking about programs, which is a different order of exercise.

Our view was that on occasion, Fund advice could be significant. We tried to break it down among kinds of countries. We looked at countries all the way from so-called small states, micro states, all the way up to the United States, and tried to peel away the different kinds of impacts Fund advice could or could not have. It is never a primary influence, in our judgment, on a country's performance. It can play a role, on occasion, depending upon the stage of debate. We looked at four countries which went into crisis during this period, part of our sample group, and the obvious question is, well, if the Fund was giving advice, why did the country go into crisis--was it because the advice was not very good or was it because the Fund was not listened to? The advice may have been not good and still not listened to, or good and not listened to, or bad and listened to. So we looked at the various possibilities there.

I think the basic conclusion was that the advice was not perfect. It wasn't terrible either. It depended on which part of the advice you looked at, and I think we'd give them good marks, even if it wasn't--how can I put it?--the whole recipe. On the fiscal side, on the capital account, we think it is more questionable that the Fund advice--the Fund didn't press on the exchange rate as hard as it might have in some cases, although it was also arguing that if you were going to peg the exchange rate you have policies that try to support it--don't run inconsistent policies with your exchange rate, those kinds of things.

At the end of the day, however, it was not that clear to us that the authorities really paid that much attention to the Fund's advice, whether they wanted to or not. They are all in rather difficult political situations, and the domestic politics overrode the Fund advice, I would say, in every case, which points, I guess, to an important lesson in this regard, the question of how can Fund advice be made more transparent, more clear. It gets to an issue of publication, I think, pretty quickly.

Let me leave it there.

A QUESTIONER: I want to follow up on this publication issue because there was a remark there that you thought advice should not be vetted or edited for market-sensitive information. Would it make a difference if--and I want to relate that to the issue of contingent credit because I have heard speculation, I guess, that contingent credit has not been used because everybody is afraid that if they are the first one to sign up for it, they will get trampled in the markets. But would it help if after this greater transparency, you would sort of give them a credit rating by publishing this information? What is your comment on those two issues, market-sensitive information and the contingent credit?

MR. CROW: I am going to deal with one and leave it to someone else, maybe Niels and the others--contingent credits is a tricky, tricky issue, in my view. When we had some remarks upon it, it was still in the process of evolving.

But let me talk about the market issue, there. We focussed on the consultation reports, and as you know, there are arguments on both sides about publication: the question of public institutions providing information to the markets, make markets work better, accountability, and so on; and on the other side is that the very act of publication can cause a crisis, and would you like to be responsible for that, given the fragility in some situations? How far do you go?

Well, we looked and looked at this, and, of course, both arguments have some merit. We came down on the side of publication, above and beyond the broad issues of principle, because as we understand the situation--and we spoke to a lot of people--those consultation reports in fact do get leaked. They go to 182 countries. It would be surprising if they were not leaked, frankly, in my view. So, from that point of view, if the market is going to get them--and let's say anybody who is sufficiently interested and determined and informed will get it--the market will have done with it whatever it wishes to do. We also make the point that it is understood that a report might get leaked, and therefore the staff, in its report, doesn't convey, let's say, the full breadth of the kinds of issues that Leslie was talking about, the vulnerabilities and so on and so forth, in the staff report.

So, from the point of view of markets, we don't see any real issue in releasing the reports, and we don't see any need for vetting the reports because they go through a process of vetting already de facto, if not de jure. That is the essence of our argument.

Then the other arguments, in terms of accountability, which carries some weight here, and we think, in terms of quality maintenance or improvement, by allowing people to see them more generally, we have a comment upon these reports. We think they carry through as well.

The other question of contingent credit, I think, I am going to leave to someone else. Niels, you drafted a chunk on that. Why don't you talk to it?

MR. THYGESEN: Well, that may be a difficult task. The contingent credit line is a tricky issue.

It is clear in our mind that to the extent that this new facility becomes important and operational, it will strengthen some of our recommendations, including that on transparency. The contingent credit line, if it is anything, is, of course, an effort to make a public demonstration of confidence in a country's policies, and how can you develop that without making your views about that country transparent?

So, in that sense, the contingent credit line strengthens the recommendations of transparency and of clarity in Executive Board discussions about the prospects of a country.

There remains, nevertheless, the question of the hypothetical, as was mentioned also by Leslie Lipschitz, as the most difficult issue at all, and in principle, the contingent credit line also covers that.

We think there is still some residual scope for confidentiality in this area without undermining the general point of transparency in our recommendations.

MR. LIPSCHITZ: Let me come back, if I may, just briefly to the question of publication. First, let's say we have been publishing more and more with press information notices out now on nearly all surveillance cases. And as you probably know, we now have a pilot project on the publication of staff reports, and the critical question that we were trying to evaluate in this pilot project was whether, if authorities know that this report will be published, the discussions will be a great deal less candid.

We heard from both sides before starting. People said, "Well, the OECD publishes its reports and the reports are negotiated down to the last comma and full stop." We didn't want to get into that situation. So, in the pilot project, we have said to authorities right at the beginning of a mission, "Tell us whether you want to publish this or not, but don't say you want to publish it and then hold off on the discussion."

Now, the results that are coming in are very interesting. One of the reports that has just been released, or will be soon, is extraordinary, where we have a country which, on a lot of the early warning system indicators, looks extremely vulnerable, but the assessment of the report is that it is not, because there are so many mitigating factors that you cannot simply use a template and go in and make an evaluation. This country was confident enough of that position that they said, "Go ahead and be as candid as you want and publish it." But I'm not sure that's going to be the norm.

So, while I have a lot of sympathy with what John was saying on publication--it certainly will enhance our clout--we are wrestling with this issue, and I think we have to see fully the results of this pilot project before we decide on which way to go.

A QUESTIONER: Isn't it true that countries would have to be transparent to qualify for the CCL?

MR. LIPSCHITZ: Absolutely, absolutely. That is exactly what the concern is. The notion is that the CCL will be a very exclusive club. To be part of the CCL, the country will have had to have met a whole lot of very strict criteria on its own transparency, on its own statistical output. Besides that, the Fund will have to warrant that its policies are pretty much perfectly appropriate, and the CCL will be there to guard against a contagion-type spillover affecting the country--so that, by the time we go into a situation like that, we won't have to go in and say, "Well, you've got to change Policies A through Z." If the policies are right, we would go in just to tide you over difficult circumstances that originate in mechanisms beyond your control.

So I do not think the Fund is out there canvassing who will join the CCL, more looking at who appropriately could be included in this club where we can really warrant that policies are going right.

A QUESTIONER: You seem to be putting across the picture of the Fund as a bureaucracy where one department does not talk to another, where people are too timid about actually saying what they ought to be saying because they don't want to upset their bosses or they don't want to upset the government. How do you create a culture where people are actually prepared to discuss the vulnerabilities, which is one thing you are asking them to do?

Then, on a similar point, you say that the Fund is doing too much, is biting off more than it can chew by expanding its surveillance. But at the same time, that is what its members are demanding. That was part of the request from Congress when the U.S. approved extra money.

Does the Fund have to go to its biggest members and say we can't do it? How do you get around that one?

MR. CROW: A number of questions

Well, the Fund is a bureaucracy and has departments,and departments are important creatures in the Fund and department chiefs are very important. We think there is a bit of an issue there in terms of how information gets transferred. I am not sure that I would say it is a matter of any kind of antipathies among departments. But everybody has a lot to do, and getting your own work done is extremely important. The staff work hard. So there isn't as much room, I would say, for cross-fertilization or sitting back as there could and, in our judgment, there should be. We criticize front offices for this, I think, which, if they should be doing anything, it should be taking the broader view. So that is one set of issues.

Timidity. The Fund is known as being pretty straightforward often, but the kinds of issues they deal with are such that sometimes they do require challenging governments in terms of what the issues are. The pressures on the organization are, I would say, enormously to hold back because of the nature of its governance and being a bureaucracy, not having political power. So, again, our judgment is that this has to be resisted if it is going to do the kind of job that is expected of it.

The final question you raised is what kind of job. Well, if it is going to do the kind of job that it is being asked to do, it is going to have to perform miracles, obviously. As indicated in our report, and this is one part where I put a paper clip in, they are required to promote, besides market-oriented reform, trade liberalization, and economic growth, also "democratic governance and social stability"; doing this would involve, among other things, "establishing or strengthening elements of a social safety net" and "the maintenance and improvement of core labor standards," as well as pursue "macroeconomic stability, while promoting environmental protection."

[Laughter.]

From another source, give "more attention" to "the development of sound social policy and infrastructure in developing countries" and "give particular priority to core budgets such as basic health, education, and training to the extent possible, even during periods of fiscal consolidation."

[Laughter.]

Well, good luck. I think there is a lot of buck-passing going on here. There are a whole series of issues here, some in terms of the extent you get away from fiscal and monetary policy and the banking system into questions of accounting or governance, how can banks collect on credits: if a bank can't realize on its assets, can it function? That kind of thing gets you away from strict fiscal and monetary policy. But I guess our view is that you should start from fiscal and monetary policy, the exchange rate, and the financial system, and then move out.

This looks like a leap into the unknown. It is not moving out, I would suggest. That may be over-dramatic, but I don't think it is unfair.

I think there is an issue there. I would even wonder whether all the members want it. When we went around speaking to people in the various countries, we found a lot of criticism of this kind of stuff.

MR. THYGESEN: Our concern is not so much that these tasks listed by the U.S. Congress and others are unimportant, but whether they should, could be done by the Fund. There are other institutions, and some division of work has to be established. The Fund could possibly show a little bit more reticence in being ready to take on new subjects in Article IV consultations and other contexts, despite the strong pressure to which it is subjected by its largest shareholder and by a number of other members.

To take a couple of the examples that John listed from the U.S. Congress--the labor standards--we have an International Labor Organization, and we have the World Bank which is more equipped to look at some of the issues of social policy than is the Fund. So it is a matter, really, of organizing work somewhat differently, rather than tacking on and becoming an amorphous, large institution.

MR.HACCHE: Leslie, departments don't talk to each other and staff are reluctant to give their views?

MR. LIPSCHITZ: I did like this image of the timid bureaucracy, and I think it's very different, looked at from the inside and the outside.

I think, certainly, from the inside, there are very many relatively informal processes where departments get together, and these are gloves-off debates. I mean, there are very, very clear differences that emerge and that are argued about, and that are taken to management in an attempt at reconciliation. We figure out where the institution is going to come out. It is not in any sense the timid institution, to my mind, in its internal workings.

Externally, I think it is true that the Fund's interactions with the press and the markets are characterized by a certain reticence at times, perhaps because we are a monetary institution and this notion of discretion is so important for an institution like this, and there is a concern about violating the trust of our interlocutors in some of these countries. So there is a degree of discretion, but, certainly, in the internal processes, in coming to a position, it is gloves-off and pretty wild at times.

Now, on the second question--on core coverage--I thought John's quote was absolutely terrific. One of the things that the staff has been very grateful for in this report is: please don't ask us to do everything all the time with no more resources--it is quite impossible. Having said that, don't expect us to be able to be relevant and to be able to answer the questions of our membership by focussing only on a few small, core, traditional areas.

Let me give you an example. If we are going to a country that has a savings ratio of 26 percent and a current account deficit of 8 percent of GDP, quite clearly there are a lot of resources available for investment. If that investment is not happening, the question is where is the money. We cannot go to the Board and say, "Well, we don't know." There are issues of governance that have to be addressed. Governance is not our core area. We cannot possibly have an internal FBI-type operation--we don't have expertise. But these are issues that have to be raised, and they have become core to the macroeconomic assessment of the country.

You can't possibly go to Europe now with high unemployment and a monetary union and not look at issues of labor markets. They are core in the current circumstances to a macroeconomic assessment.

So, in each case, there are issues outside of our core concerns that are nevertheless core to the set of considerations we have to address, and I think as long as the case for relevance can be made on each of those issues, they ought to be brought to the table.

MR. CROW: Could I add something on this?

I am not sure whether we are disagreeing or not here, but let me emphasize the evaluators' view. It is not that you just discuss the exchange rate, fiscal policy, monetary policy, and leave everything else, but it is that the Fund is primarily a financial institution with financial expertise, and it starts from international concerns about exchange rates, overlap, spillover, et cetera, and so on and so forth.

There are issues that extend beyond that, depending upon the country that you have to deal with. Even in the labor market in Europe, we say there are certain issues the Fund may wish to deal with, some which it probably should leave to others to deal with, for example, and that is an example which we discuss in the report. But the case for extending beyond that has to be demonstrated each time. It cannot and should not be assumed if the Fund is to maintain focus. I think that is the essential point.

Yes, if you can make a good case as regards relevance--and its relevance to the core, not relevance in some abstract sense, I would say--yes, do so, but if you can't... That case has to be made each time. Otherwise, the waters get very, very muddied, in my view.

MR. BERNES: In the peer review, you do have the summing-up of the Board discussion, and certainly, this issue of core issues is one which the Board identified as the critical issue, which is very complex, and which it wants to return to.

I think there are a couple of perspectives. One, there has been an evolution of the range of issues which the Fund has looked at, partly because our understanding of those other policies that can affect macro performance has grown; and as we have seen that issues such as governance, economic governance, can have a significant impact, the mandate has expanded to cover those, and it has put tremendous strains on the resources of the Fund at a busy time. It also calls, in some cases, for types of expertise which the Fund doesn't necessarily have and which, therefore, needs to build up, or as Niels Thygesen said, to work find new ways to work more collaboratively with other organizations.

It is a complex, difficult issue, but it is one which we are going to have to address as an institution which the Board has flagged as one in which it has to come back to. There is some very careful thought and some direction to staff, but I think the broad sentiment was that while there are certain core issues that John Crow mentioned that one should always look at, other issues which have potentially a significant impact on macro performance also have to be considered.

On the comment on timidity and discussions between departments, both this report and the other report on research activities which you will see today both commented on the question of knowledge transfer and identified some problems. There are some problems there just because resources are being strained and people for the last couple of years have been running to stand still, and may not have seized upon other areas of research, or knowledge in other departments, to the extent they should have. That is clearly something which is going to have to be addressed.

In terms of language and being precise, we have to recognize that these are complex issues that we are dealing with, with potentially significant impacts on people as well as on financial flows, and it is difficult to be short and concise. The problem you confront was shown in the first question at the press conference today: in one sentence can you tell us whether surveillance is good or bad? Journalists like to--because they want to communicate--bring it down into a very narrow sentence. When you are dealing with a country's macro performance, yes, it would be nice to say your policies are great or your policies stink. Reality is a lot more complex than that, and as you try and describe a situation, the interaction of policies, the vulnerabilities, you are driven into a complex and guarded statement of the facts and the possible evolution, and it is necessary to bear that in mind.

MR. LIPSCHITZ: I would add one small addendum to what John said and what Niels said.

On the issue of expertise and drawing on outside expertise, this is a message from the report that we take wholeheartedly. It is absolutely nuts for us to be using resources looking into some complexity of the labor markets if the OECD has already done the job and we can simply draw on the report. There are a whole bunch of areas on financial markets where we have a great deal less expertise than the BIS and other institutions. Let us draw on that expertise rather than reinventing the wheel. So that is absolutely dead right, and we take that point from the evaluation very seriously.

The argument for relevance when we go outside core issues, I think, is interesting and important. The one area where I think the staff has been way too timid is vis-à-vis the Executive Board. I recall an instance where on an Article IV surveillance with Switzerland, a Director was quite incensed because the staff didn't know the infant mortality rate. Oh, my God, Switzerland, the infant mortality rate, whatever it is, it is low. Forget about it.

[Laughter.]

It's not an issue. And it certainly isn't an issue on which we ought to be focussing, but that is not what the staff would say. They would scurry around, all embarrassed, and say, "My gosh, we have to know the infant mortality rate for Switzerland, the Netherlands...." Totally idiotic! Let's focus on core and not get sidetracked by silly issues.

MR. HACCHE: A couple more questions?

A QUESTIONER: Is there any discussion in your report on the publication of Article IV's?

MR. CROW: I'll talk on the report, and maybe I will make a comparison with the pilot project.

We propose they all be published--i.e. mandatory--for the reasons that we talked about earlier.

I might just say a word about the pilot project now that was noted. This is self-selection: a country volunteers for it. It is not mandatory. You call for volunteers, and the Fund has been hoping for what they call a "robust sample." I don't know what the sample is. So I don't know whether anybody would judge it robust. No doubt, you can call it--

MR. THYGESEN: 20 countries.

MR. CROW: 20 countries, 20 robust countries, let me put it that way. But it is a self-selection process and has the questions relating to self-selection bias involved in it, but the view is that that will provide some information and it may provide some on what it's able to do. But in terms of our recommendations, we say publish them.

MR. BERNES: From the Board's perspective, on a lot of these issues there have been ongoing adjustments to policies while the evaluators were evaluating, and this, indeed, is one of those cases where there has been a healthy debate in the Board between those who favor mandatory publication--I happen to be one of those--and others who are concerned about the sort of concerns that Leslie mentioned earlier, about the impact on the policy dialogue with the country.

The conclusion we came to was to launch an 18-month exercise--the pilot project--where countries can volunteer to have their Article IV published, and we will assess the results of that, and the impact on the policy dialogue where the country is concerned, and come to conclusions.

This has been going for about 4 months now. It has got another year or so to run before we begin to draw the conclusions in terms of whether we should continue with it as a voluntary system, go to a mandatory system, or go look at something else.

MR. LIPSCHITZ: Movement on things like this is necessarily slow, because we're an international institution, not a supranational institution.

So you have a division within the membership of the Fund, you do a pilot project, you see how it works, and eventually, there is movement toward creating a consensus at the Board to go ahead with publication, or not, but necessarily in an institution where there are 182 member countries, it takes a while to get everyone on board with any initiative.

A QUESTIONER: I have a technical question. I just wondered if you could clarify. You both used the term "countries approaching crisis," and we have the list of those countries, obviously Brazil, Korea, and Thailand. When does your study--I mean your review--actually end relative to the point at which they are approaching crisis? MR. CROW: Well, in terms of the one instance which I pointed to and where I can give you an exact date--

Thailand--I say July 1997 because in July 1997, the crisis completely broke out in spots, and there was a Fund mission to negotiate a standby arrangement. So, from that point on, surveillance becomes something else.

A QUESTIONER: What about in the other two?

MR. CROW: We looked at Thailand, for example, from, say, the early '90s up until July '97. And the kind of questions we asked were: well, the Fund was giving advice for this period, to what extent were they warning about a potential crisis? To what extent were the Thais listening? To what extent were they focussing on what proved in hindsight to be the right things--either side, for that matter? To what extent were the Thais already supplying the Fund with information--is another issue there.

We looked at the same situation for Korea, which is a bit later in the year, like December, I think, you might say. It was up until then, but we looked at the evolution of events before. When is Brazil? January 1999. Well, we looked that far. The Czech Republic--we looked at the period when the Czechs went into a...

MR. THYGESEN: Spring of '97.

MR. CROW: Spring of '97. So those four cases were looked at in terms of the evolution up to the point where they asked for Fund financial assistance. Then it became a different ball game for us and outside our mandate.

MR. THYGESEN: One might mention also the case of Sweden, where we looked at the whole period of the '90s. Sweden did not seek assistance from the Fund, but it did have intensive contacts with the Fund. There, we studied certainly also advice-giving up to the outbreak of the crisis and immediately following it.

MR. HACCHE: Thank you very much for coming.

[End of press conference.]