IMF Conference

International Monetary Fund
Multipurpose Room
Washington D.C.
Wednesday, December 19, 2001



Masood Ahmed, Deputy Director, IMF Policy Development and Review Department

James Boughton, Assistant Director, IMF Policy Development and Review Department

Joanne Salop, Vice President, Operations Policy and Country Services, World Bank

John Williamson, Senior Fellow, Institute for International Economics

MR. AHMED: Today's topic is IMF conditionality. It's a topic, of course, that's been on the table for a long time in different ways, but more recently, I think, it's been particularly interesting in the context of looking at aid effectiveness work.

And, another strand has been looking at it in terms of the relationship between countries and international institutions. Does conditionality infringe on or undermine the ownership of programs that countries develop and implement? Does the way in which conditionality is structured affect the likelihood of success of those programs?

So there is a whole debate over the last three or four years, coming out first from the crisis in Asia, which then spread out to other countries in '97-98, but also coming out of the work done in many low-income countries where there were ongoing programs in Africa that the Fund and other agencies were supporting, where the same kinds of questions about ownership and about effectiveness kept cropping up. This led the Fund under the leadership of the new Managing Director to look at the experience of conditionality in Fund programs, but within the broader context of looking at conditionality as it's being developed and applied by other institutions. The result of that has been a review of the Fund's conditionality, which has been ongoing now for almost a year.

So we're at this point where we have drawn a lot of lessons from the first year of looking at our own experience, looking at other people's work, and we are ready to start thinking about how to draw and synthesize all these lessons in terms of now developing some guidelines for the Fund's conditionality going forward, building on a set of interim guidelines that we have.

At the same time, other people obviously have been looking at conditionality, too, and the World Bank in particular has also been looking at its experience in terms of the way in which conditionality has worked in the context of Bank-supported programs. And they, too, have been trying to draw some lessons in terms of how they could move forward to make that more effective in support of implementation of programs in client countries.

And while these agencies have been looking at their own experience and drawing on others, people who are academics or civil society organizations or researchers have also been examining this set of issues and contributing in many ways to the processes that have been underway in the institutions.

So at today's Economic Forum, we thought we'd start off with a panel and then open it up and get inputs and comments and questions from many of you. And I see in the audience a number of you who I know have been thinking and working on these issues. Some of you we have talked with in the past year as we've done our own work. To get us started, we have three people who've been thinking hard about this set of issue.

To my immediate left is Jim Boughton, who, in addition to being the author of the book that John was showing you, has also been leading our work on the conditionality review in the IMF. Jim's currently with the Policy Development and Review Department in the IMF, but has also been previously in the Research Department and during a period was the historian of the Fund, which is when he worked on his book.

Next to Jim is Joanne Salop. Joanne is currently Vice President of Operational Policy and Core Services in the World Bank. Her job is to head up the Bank's work in framing the set of operational policies that guide the Bank's work in adjustment lending and in its operations more broadly. And Joanne has also in a previous incarnation, before she went to the World Bank, been a staff member in the IMF, so she has a good perspective of how the work of the two institutions has evolved over the years. Other than her current job in the World Bank, she has done a number of other assignments, including being chief economist for South Asia in the World Bank.

Next to Joanne is John Williamson, who is currently a senior fellow at the IIE, the Institute of International Economics, but who has also over the course of his long and distinguished career been a staff member at the IMF some years ago and more recently at the World Bank where I believe he succeeded Joanne as the chief economist for the South Asia Region.

So as I was going through the bios of these people, it struck me that there is a certain overlap, and I hope that we'll be able to get sufficiently challenging perspectives from the three to start the discussion. But then I also very much hope that you all will come in with your perspectives, and not just questions but also your own ideas and advice on how we could explore this issue further.

I'm going to ask each panelist to take 10 or 12 minutes to give us their initial perspectives on the topic, and then we'll open it out, have some questions and answers. And before we end, which we will at 4 o'clock, I'd like to give each of them a couple of minutes to come in with any final thoughts that they have.

So that's how we might organize ourselves between now and 4 o'clock, and without further ado, let me turn to Jim.

MR. BOUGHTON: Thank you, Masood. Welcome, everybody, and thank you thank you, John Starrels, for plugging my book in my presentation. I've also slipped in a little subliminal plug for my book in my presentation. It looks like a standard IMF logo on the screen. If you look closely, you'll see that I've managed to slip the title of my book into the logo, in case people stare at it long enough.

I also have a less subliminal plug that I can give you, showing you the cover of the book. You can buy it on your way out in the Visitors Center.

So, having done that, let me turn to the topic for this afternoon. I've turned around the title of the event very slightly. The subtitle that had been announced in the fliers was how much is enough in terms of conditionality. I'm not sure that there's really an answer, and I certainly wouldn't want you all to be sitting there thinking that we're going to come up with a number for exactly how much conditionality is enough. So I've turned it around a little bit to say let's focus on what we're trying to achieve with conditionality, which is to enable borrowing countries to implement strong economic policies that they can, in fact, themselves take responsibility for, that they can own in some sense. And I will look at what the role of IMF conditionality might be in trying to bring that about.

Now, having written a work of history over the last several years here, I'm going to start by giving a little bit of history on just how we got to the point where we are now. In other words, we're going through this review to see where we should be--how we should be changing our approach to conditionality, but I think in order to do that, it's important to understand why we've been doing what we have been doing.

The Fund has been doing conditional lending, lending conditional on policy changes in countries, for almost half a century. The very first time that the Fund experimented with conditional lending was in a stand-by arrangement for Peru in 1954, which was made conditional on Peru promising not to change its policies with regard to intervention in foreign exchange markets, basically saying we thought Peru had a means of keeping its exchange rate stable and sustainable, and they had to promise to keep doing that. And so that was our very first effort at conditionality. It was a fairly modest beginning.

But over the next few years, the practice of conditionality became much more formalized, very largely as a result of some very forward, innovative thinking by a gentleman who is here in the audience with us today, Jacques Polak, who developed a model that had very clear policy implications in linking restoration of stability in the balance of payments to money leaking out into the balance of payments through excessively expansionary monetary and fiscal policies. So that instead of requiring countries to try to act directly on the balance of payments, which would be very difficult, we wanted them to act directly on policies that they could control, monetary and fiscal policy.

The first time this was actually applied in practice through what were called performance clauses or performance criteria was in a program, a stand-by arrangement for Paraguay in 1957. They actually had four conditions. They had to do what Peru had promised to do, which was to maintain a stable exchange rate policy. But they also had to promise not to exceed a ceiling on domestic credit creation and not to exceed a ceiling on spending on both the current and capital account of the government.

So that then became the basis, this basic application of a model linking the balance of payments to monetary, fiscal, exchange rate policies, that then became the basic way of applying conditionality. But it was done in a very haphazard way. If we thought it was necessary, we did it. If we thought it was not necessary to impose conditions, we didn't do it.

As late as 1967, when the Fund had a large stand-by arrangement with the United Kingdom, the United Kingdom was able to borrow from the Fund without any explicit performance criteria at all. And that was because of kind of a special relationship, if you will, having been one of the two major founding countries of the IMF. But it made a lot of people nervous. It raised questions of uniformity of treatment. A lot of these other countries who had been borrowing subject to policy conditionality said this wasn't fair. And so that led to the first effort for the Fund to write down a set of guidelines governing the policy conditionality on IMF lending.

Those guidelines established principles of uniformity of treatment among member countries, and said that any time a country was going to be borrowing in what we call the upper credit tranches, borrowing more than small amounts in relation to its quota, aside from a few well-defined exceptions through special facilities, that it would be subject to a standard set of performance criteria. So it was really 1968 that marked the regularizing, if you will, the institutionalizing of these conditions.

In the 1970s, economic coordination got more complicated. By this time the fixed exchange rate system had broken down, and countries were using lots of different exchange rate systems. We had the two oil shocks, the huge increases in oil prices, and other major instabilities in the economy. And the Fund had started experimenting with more extensive policy conditionality, a longer laundry list, if you will, of conditions that countries had to meet. And so then an effort was made in the late '70s for the Fund to say let's try to cut back a little bit, let's establish principles of parsimony, making conditionality less burdensome, again, trying to restore something of this uniformity of treatment among countries.

So those guidelines were codified in 1979, and those are the guidelines that we still operate under today. So for those of you who are schooled in advanced mathematics, you'll see it's been more than two decades that we've been operating under these guidelines. And just as we slipped away from some of the earlier agreements during the'70s, during the '80s and '90s we slipped away from some of the principles that were enshrined in these guidelines in 1979.

The point I want to emphasize is that the drift in the nature of conditionality over the last 20 years has been made for some very, very good reasons, that there were limitations to the guidelines that had been agreed on, reacting to an earlier age, that became apparent in the '80s and the '90s that led the Fund to expand the nature of its conditionality. And there were two things that I would mention:

One was that by focusing on overall aggregates, the overall fiscal deficit, total level of domestic credit creation and so on, you leave a lot of scope for the government to fix its own policies within those overall guidelines, that overall conditionality. That's a good thing up to a point if the government makes the right choices. If the government makes the wrong choices, then you get bad outcomes. And what we were finding was that there are a lot of cases where you got some pretty bad outcomes; that if the Fund just said you have to cut down on your fiscal deficit, in many cases the easiest way to cut down the fiscal deficit is to cut back on subsidies on bread prices. And that means that you may have millions of people out there who can't afford to buy bread anymore. And so the protests against the Fund and against countries' governments for making bad choices led the Fund to be more specific about the nature of its conditionality.

Secondly, countries started experiencing very long-lasting structural problems. We were dealing with more low-income countries, countries with really deep-seated economic problems. It became natural for the Fund to try to help those countries deal with those problems. And so we started broadening conditionality into more structural issues, and that particularly became apparent in the 1990s. Even though the 1979 guidelines said that conditionality shall be on macroeconomic policies not on structural policies, except insofar as they influence the macro, that was an exception that you could drive a truck through, and we kept driving bigger and bigger trucks through it as time went on.

This came to a head by the late 1990s when, in the course of trying to solve a financial crisis in Indonesia, Indonesia submitted a Letter of Intent on its policy intentions to the IMF in April of 1998 that listed 117 specific structural actions that the government intended to take during the course of this program, covering a very, very wide range of policy issues, including six conditions that were related to the environment, not a traditional area of concern for the IMF at all.

Now, all of these conditions or nearly all of these conditions were put in there for very good reasons. Only a small portion of these conditions were related in any very specific way to IMF conditionality. In other words, these were not performance criteria that the country had to meet in order to qualify for the next drawing. But they were promises that were going to be evaluated over time, and before the Fund would agree to the next stage of the program, before it would agree to the next round of disbursements to the country, these items would be subject to a review.

Now, obviously, nobody expected Indonesia to fulfill all 117 of these promises. It was impossible, and everybody recognized it was impossible. So the problem was that even though these were not specific policy conditions, there was no real assurance, there was no point at which Indonesia could say, okay, if we do these things, then we will be assured of drawing the money three months from now, or whenever. So there was an ambiguity that had crept into the whole process over time that was, I think people realize now, beginning to get out of hand. It was serving very useful purposes because Indonesia had to do these things, but it was making IMF conditionality very difficult to apply.

So that brings us to where we are now, as Masood summarized at the outset. We're trying to develop new guidelines for the 21st century that will enable us to pull back, restore some of those old principles, but still meet the problems of today.

There are three goals to this current review: streamlining, focusing, and fostering ownership. I'm just going to say a very few words about each one of these three before I finish.

We all know what streamlining means. It means that instead of having a clunky old 19th century steam train, you have a nice streamlined train. On the screen now you see a steam train, a French steam train of the 1930s. It was one of the very first efforts at streamlining. But what does streamlining mean in this current context? It means, first of all, less extensive conditionality. It means you're not going to write down a program that has 117 structural conditions in it. That's going to require a lot more collaboration with the World Bank, which my friend and colleague Joanne Salop is going to talk about in a few minutes.

It also means less detail. It means that if we think, for example, that a country needs to impose a value-added tax, whereas in the recent past we might have specified every step that the government needs to take in order to get that value-added tax in place and working properly, now we might have a much more general condition on what they need to do.

We're not trying to go back to what I would call and characterize as the bad old days of just saying we're going to impose overall conditions on fiscal deficits and so on, which we know led to a lot of bad outcomes. But we are going to try to cut back in those two areas.

Another thing we're not trying to do is we're not trying to weaken programs. We're not trying to get to a situation where we're just going to say countries can do whatever they want and all we're going to do is lend to them, because we also know that that can lead to some very bad outcomes. So that's what streamlining is.

Now, we also all know what focusing means. Here we have a photograph of a very large Japanese telescope focused on outer space. You probably can't see it, but there are two very small human figures down there, which shows you this thing must be hundreds of feet high. But, anyway, that is what focusing is.

Now, what is focusing in the context of this review? First of all, it means you have a clear view of goals. What are the goals of Fund conditionality? First of all, the primary goal has always been and still is to help countries alleviate their balance of payments problems. But at the end of this program, the country has to be able to finance its balance of payments. If you don't solve that problem, you haven't solved any problems at all. So that's goal number one.

The second goal--and this is language taken directly from Article I of the IMF's Articles of Agreement--is that the country should not be taking measures destructive of national or international prosperity in order to solve its balance of payments problems. That was originally a code for saying you don't have a competitive devaluation, you don't try to solve your problems at the expense of your neighbors. And that's still a principle that's enshrined very clearly in everything that we do.

But, third--and this is a goal that has become extremely important over time; it was always there, but it's risen to the fore--is that the Fund has to help the country improve its sustainable growth rate, that solving your balance of payments problem through compressing your imports, squeezing people's incomes, hurting the poor--none of these things solve any problems in the long run, and so you have to stay focused on that goal as well.

The idea, then, is to have the minimum--you want to end up having the minimum amount of conditionality that enables you to help the country meet those goals. That's what the whole exercise is all about, and that's really the answer to the question that John posed in setting up this forum of saying how much is enough.

The third element, ownership. Again, we know what ownership is. What does it mean in this context?

First of all, it means that the country has to take responsibility for strong policies. It does not mean that the country does whatever it would have done in the absence of getting into this mess in the first place. It means that the country's expected to reform, but it's also expected to design its own program. And the idea of the Fund coming in and telling the country what it has to do, which is an idea that has crept in over time, has to be completely disposed of to the fullest possible extent.

What will it take to achieve this? That's a very tall order, by the way. If we're going to achieve that, first of all, we need more flexibility on the IMF's part. If a country comes to the IMF and says we want to borrow from you, we need your financing, we're prepared to change our policies in order to try to solve our problems, but we also need your financing, instead of having the Fund go in and saying then in order to qualify for our financing you need to do X, Y, and Z, the Fund has to be prepared to say to the country, okay, then tell us what policy changes you're prepared to make and then we'll have a dialogue about it. Let the country put its program on the table first. That's the kind of flexibility that this is going to take.

Secondly, then, you have to have active participation in the process by the country. That means that in the course of having this dialogue you need to have as broad a participation as you can. We've been doing this for the last few years through the process of--what we call the PRSP process, the Poverty Reduction Strategy Paper process. And we're trying to generalize that process in order to encourage broad participation in the country.

Thirdly--and this is in a way the tallest order of all--is you need selectivity by the Fund. The Fund has to be prepared to wait until this whole process of flexibility, participation, and dialogue has got us to a point where the country is ready rather than going in and saying we're going to lend to this country just because the country needs our help.

Will we succeed in doing this? Some success is already evident. The streamlining process is underway. If you compare programs being approved by the Fund's Executive Board now with programs being approved two or three years ago, on average you'll see some major changes.

Collaboration with the World Bank has been strengthened, as Joanne is going to tell you in a few minutes. But the other things, these other--these tall orders I've been talking about, the real challenges of this, trying to achieve more flexibility, trying to get more participation by countries, trying to get selectivity on the part of the Fund, these things are going to take some time. We're really at the early stages now, and I think although those of you out there who may be skeptical about how far we're going to get, I think your skepticism based on past performance is probably well founded. But hopefully when we have another forum along these lines two years from now or three years from now, you'll have less reason to be skeptical.

So stay tuned, wait for the movie. You can see we have--this is an advanced poster for the movie about streamlined IMF conditionality, and so when it comes out, we'll see you all here again in a couple of years.

Thank you very much.

[Applause.]

MR. AHMED: Thank you very much, Jim, for the show.

Let me turn it over to Joanne, and I'm hoping that Joanne will talk about not only the Bank-Fund collaboration I mentioned, but also how thinking about conditionality is evolving in the context of the Bank's own work.

MS. SALOP: Actually, I would have preferred it if Masood had suggested that earlier, but I'll try to do my best.

Anyway, clearly, from Jim's presentation, this is not the Fund that I left in 1986. And I wanted to say at the outset a couple of things. One, I'm very happy to be here, very happy to be back at the Fund, and just to let you know, before we on the panel came here today, we had a very nice lunch that John [Starrels] and Masood hosted. And it was very interesting, and clearly the focus of the lunch was really on history, and, of course, largely led by Jim, not to plug his book, although we were all instructed to plug is book. But, nevertheless, at the table a lot of historians, John Williamson, of course, on my left, and myself. I've been a history of the international financial system for many years. I'm delighted, of course, to see Mr. Polak in the audience who hired me into the Fund many, many years ago.

So a couple things in terms of history, I think, that I wanted to put on the table from the Bank's perspective, and particularly following up on what Jim had said in terms of the Bank-Fund perspective. A little bit of history from where I sit and what I see in that long history.

I think many of the people in the audience are probably Fund staff. Is that correct? I don't see the little yellow tags. Does that mean Fund staff? Yeah? Because I don't see Bank staff, but just in terms of what your--what you know.

Well, you know that the Bank was created at the same time as the Fund way back when in the late 1940s, and the Bank first was called the International Bank for Reconstruction and Development. And the reconstruction was an important part of the Bank. Our first loans were to Europe. Our first loan was to France and then to Netherlands, Luxembourg.

Interestingly, though, the Bank was created really as an investment bank. Engineers, et cetera, were really there. It was--our first loans were program loans, so in some sense, the Bank's return to work on what has come to be called structural adjustment is in some sense, in some ways a return to our roots. But we were basically a project--institutional lending, institution for physical investment projects. We didn't even start lending for education I think until the '70s, et cetera, and even then the people who did education loans in the Bank were architects because what we built were schools. We didn't worry about the curriculum. Today we do education, of course, it's different. We do the curriculum. We do the business strategy for running the schools. We do sustainability issues.

But the World Bank has changed a lot over its more than 50 years, although we retain some of those things from the beginning. And, similarly, Bank-Fund relations, of course, have changed over that period. You can imagine during the period in which the Bank was an investment lending bank--and I was not there then--that we didn't talk a whole lot to the Fund. I guess we--well, I guess we did talk a whole lot because we probably shared the same building and we certainly shared the library and other things. But in terms of our work, the overlap in the work is much smaller.

But, nevertheless, clearly, macroeconomic issues, the central issue for the Fund, was important in some project issues, and I'm sure there was a dialogue throughout the '60s and the '70s. Clearly, in the beginning of the 1980s, that changed a lot. The Bank moved into structural adjustment lending with the debt crisis, the oil crisis, et cetera. And there the overlap between the Bank and the Fund, the mutuality of interest in terms of day-to-day, became much more center stage.

And so this was really what I would characterize in this era as very much the Bank moving into an area where it had previously been very much the Fund's preserve in terms of macroeconomic issues, and then sorting out Bank-Fund collaboration in that era and conditionality was very much focused on that.

How did we in the Bank take into account the Fund's views on macro? Could we go ahead with a program if the Fund was unhappy with the macro? How did we take those things into account? Did we do in the Bank our own macro work, or did we rely on the Fund's macro work? So those were the kinds of issues that needed to be sorted out on a day-to-day basis.

And there were some highly publicized cases where things did not go so right and which changed then the way we dealt with those in the late 1980s, Argentina at that time.

Now, during the '90s--and we saw this in Jim's slide--I see the transfer being very much the other way, and that it was during the '90s that the Fund started to move into areas of the Bank's concern, and Jim laid that out, in terms of the Fund became much more interested in structural issues. So just as the Bank moved into the Fund's areas in the '80s, taking up macro, worrying about macro much more, building up its macro people, in fact, I was part of the exodus from the Fund to the Bank in the '80s as part of that, when, in fact, staff were moving from the Fund to the Bank, macro staff, as the Bank built up on the macro side.

In the '90s, as the Fund worried about structural issues, actually, you probably saw the reverse movement in terms of Bank staff moving to the Fund in terms of structural expertise. And what's behind this? Well, what's behind this is the fact that we all have a common, increasingly common model, which is everything matters, and that it's not just macro matters for stabilization programs, et cetera, and that structural policies matter for structural programs.

Basically, the story that we all have is that for country development, it's good policies, and those are the mutuality of structural and macroeconomic issues. So that, not surprisingly, the Fund--the Bank worries about structural and macro issues, the Fund worries about structural and macro issues.

Now, this brings us institutionally, I think, globally and for students of the international financial system and shareholders, well, what kind of system do you want to have? You know, if you're designing the overall system of institutions, do you want to have a set-up where you have a Bank that looks just like the Fund and that maybe they compete against each other or maybe they work together, but that the Fund has the same kind of staff, structural, social, environmental, macro, as the Bank, structural, social, macro, environmental? And that you move to--do you want to have a set-up of the Bretton Woods institutions in which you have maybe more generalists everywhere, or maybe a whole lot of specialists everywhere? But do you want to have two institutions that basically look just like each other, twin sisters, if you will? Or do you want to have a situation where you have more specialist institutions who cooperate? So that you have deep within the Fund maybe real expertise on the macro stuff, as indeed you do, and then in the Bank real expertise on the structural issues and that you cooperate with each other and share.

It's broader than the Bank and the Fund. It goes to the multilateral development banks, Inter-American Development Bank, Asian Development Bank, African Development Bank. It goes to bilateral donors as well. And, in fact, throughout the 1990s, we all know partnership became one of the bywords of the day, and partnership among donor agencies really recognizing the issue of we need to drill down, get strong, be deep in our areas, and share expertise across, that model being accepted rather than a model, if you would, of dilution where we all replicate within our agencies all of the expertise and maybe at not such a degree of depth.

Now, what I see is where we are now is on that--basically the choice has been made by our shareholders that we ought to go for the drilling down, the deep drilling within the institutions in the partnership model, and that's how I interpret in part the streamlining discussion here, and particularly the Bank-Fund role. So an attempt now to, for the Fund, pull back into areas of core competency, for the Bank to pull back into areas of core competency, and to talk.

Now, that's not so easy because we talk, but lots of issues in talking. I want to tell you about three of them.

One, well, which are our areas of core competency in the Bank and which are the Fund's? Some are quite clear. Jim even alluded--environment I think everybody accepts, World Bank. Certain issues, I guess central bank, technical assistance, the Fund. And we could array all of the issues along a spectrum, some clearly Bank, some clearly Fund, some in the gray area in between. And there's been much debate and discussion about what those are, and some areas are clearly shared: financial sector, trade, tax, et cetera. So--and I think that is pretty well sorted out.

And then what's the protocol? How do we decide what's what and where's where? How do you resolve conflict? How do you--if one agency says, you know, it's pink and the other agency says it's blue, who decides? Which is it? Or is it, you know, purple? What is it? So how are those things--but I think that--a lot of progress has been made on that.

Two, the second, more serious and current issue is, well, how do you handle cross-conditionality? How do you handle issues with the Boards? How do you handle legal issues? Because if indeed the Fund is running macro, and with support from the Bank in structural areas, and the Bank is running, managing, helping, expertise in structural areas, with support from the Fund, and you have to take a program to the Board, you know, say the Fund, relying on support from Bank staff, well--well, what? I mean, what does the Board do? And do respective Boards then rely on a decision from--the Board of the Fund rely on a decision from the Board of the Bank?

Now, here the decision is also clear and been taken, no cross-conditionality and basically what we have set up is a system where--or that we're working towards in which it's quite clear individual Boards make their decisions, and make their decisions as always based on advice from management and their own considerations, but that the advice from management may well be informed by the views, et cetera, of staff of the other agency. So the so-called lead agency concept, and that is what we are pursuing.

So that I think is pretty much resolved. We have to play it out, but that's the second point, again, the first point being which areas do you specialize in, pretty much clear. How do you handle cross-conditionality? Pretty much clear. Third issue, major challenge, and this I think is where we are right now, and an issue that we're working on. And certainly an issue that I am personally involved in, you know, in terms of the operational management side in the Bank and working with Masood and his colleagues on this at the Fund. And, that is, how do you actually handle--how do you actually manage the work, and particularly if there are differences in culture between the Bank and the Fund?

Fund, emergency work, a country, you know, needs help right away, needs a program, needs stuff, needs fast action. The Fund, of course, has a strong reputation for very rapid and effective action, quick.

The Bank, on the other hand, does not have that reputation, and so how to square that circle. If the Fund goes out, gee, they need something, you know, tomorrow, very quickly, the Bank--where is the Bank on these issues? So, one, how do you handle the different perception of timing.

Two, on these issues of depth--and that's related to the question of timing--again, much of the discussion at our lunchtime was focused on privatization issues, where I think it was clear--and we were discussing what the lessons learned from history and research are, and clearly accepted that on privatization, you know, it's not a revenue-raising vehicle, it's really a structural issue, shouldn't be having conditionality on numbers, et cetera, in programs. But that question of the detail, the devil being in the detail in a lot of the structural issues requires then a lot of--oftentimes a lot of prep time to get ready, not to say that macro programs don't require prep time, but, nevertheless, if you're doing a continual Article IV, up-to-date, et cetera, and moving in on macro, monetary, and fiscal aggregates, it may be possible to do those more quickly than developing the detailed work that is needed on some of the structural issues. In addition, on the structural issues, breadth, lots of them, and they range from environment to social to infrastructure, regulatory framework for this, that, and the other thing, huge range in the structural area.

So consider, gee, Fund mission goes, needs help or needs support from the Bank on structural issues, are we ready? Have we done the prep work? Have we done the prep work in the required detail that we could give to the Fund, look, here, we've analyzed this sector, here's this, this, this, can we be ready every day of the week, every month of the year? The answer is no.

So how do we manage that? I mean, how do we manage these different kinds of time frames where, again, the Bank has been working on more of a--not the bullet train, not the fast train--by the way, we did finance the bullet train in Japan.

[Laughter.]

MS. SALOP: This is true. But we were on--you know, on a different train. How do we marry, how do we merge those two things? And that, again, is one thing, as I say, we're working on that now. One solution we are putting on the table is, well, fine, staff ought to talk to each other earlier, you know, and if, in fact, there are issues that are likely to come up that the Fund staff can be seeing, we in the Bank really need to know about those things earlier on, be talking about those, getting them into our work program so that we can have done the detailed work, indeed, in a timely manner should the issues come up.

So I think, road ahead, I think that the position here on the effectiveness of this work we're doing in terms of teaming up, trying, each of us, to specialize in our particular areas, work out the actual day-to-day work modalities in this stuff really will be critical for how successful we both are in streamlining of conditionality, and, in fact, making a stronger international financial system where we're all much more able to specialize at the same time that we partner across.

A final word, just to respond to the point that Masood had asked me at the outset, just much of the work we are doing in a parallel fashion along with the Fund staff on conditionality, we also have some--we don't have Jim's book, but we have some other things, and we have them on the website. If you give me your name, I can send it to you. But we have been doing work, of course, on conditionality, many of you know, and we've talked to some of you. We are doing this work to really revise our guidelines as well. The same stories coming out all over, as you know. The real issue is country ownership, country programs, et cetera. The question, the challenge for us all is how do we effectively influence countries. How do we get--inform the policies that they want to take. So I'd be happy to follow up on any questions.

MR. AHMED: Thank you very much, Joanne, and thank you for putting it also in this kind of historical context of Bank-Fund relations, because I think it is important when we talk about conditionality not to think of it as a disconnected phenomenon, but that it is part of an overall process of technical assistance, policy advice, program design, monitoring. It's one part of it. It's not some self-contained dimension of our work.

Let me now turn to John, who's looked at this from both the inside and the outside. John?

MR. WILLIAMSON: But who hasn't recently been looking at it in the same sort of detail as the two previous speakers, so my apologies.

I was sent some background reading, incidentally, by the organizers of this meeting, and it consisted of the 69-page chapter of Jim Boughton on this subject. Let me say, it was actually an excellent read. I thought it was a masterful weaving together of the story of IMF conditionality in the '80s, so I add my endorsement of this volume.

I want to--I interpreted our terms of reference as being to talk essentially about the role of the structural conditionality in IMF programs, and so that's what I plan to do. I understand why structural conditionality got there in addition to the traditional conditionality. I think the reasons, in fact, were touched on by Jim Boughton.

There was a lot of concern in the early '80s that IMF conditionality had a deflationary bias, that it was, therefore, bad for growth. There was a lot of concern that when governments were simply told to cut public expenditure, they very often cut the wrong things--things that were bad for long-run growth, things that were bad for short-run equity, and, therefore--and then they blamed the IMF for forcing them to do it, so the IMF tried to get out of that.

So I understand the pressures that led to the emergence of structural conditionality, but it seems clear that, in fact, this has bred its own set of problem. There have been far too many conditions. Regarding those 117 conditions in the Indonesia program, maybe the IMF didn't expect all 117 to be fulfilled, but listing 117 and then saying you have to satisfy who knows how many of them in order to get this released I think wasn't helpful. So there were too many conditions covering things that really were not critical to the immediate problems of the country.

It led the Fund into a series of areas where it doesn't have any comparative advantage, and it undermined country ownership. So it has bred its own set of problems, and the question is, therefore, today what should be done about this.

Now, what has surprised me about the discussion of the two previous participants is that they didn't distinguish between the different facilities of the Fund, which, to my mind, pose some very different issues. The Fund's traditional stand-by facilities and the new Supplemental Reserve Facility. The crisis one-the SRF, right? These ones are there to deal with the balance of payments problem, and that may be a traditional balance of payments problem of the current account type, or it may be the new capital account type of crisis where there's a capital account reversal. I think that is a meaningful distinction. It's obviously one where there's no sharp dividing line between the categories. But there are polar types, and it's useful to make that distinction.

But in both of those cases, one really looks to the Fund's stand-bys or the supplementary reserve facility or, in principle, the contingent credit line, if any country ever uses it, as the instruments that are appropriate. And it seems to me that in that context the need for structural conditions is really -- [tape ends].

-- yeah, okay, you may want to pay some attention to how you safeguard the poor in the immediate course of adjustment or, for that matter, safeguarding the environment, conceivably. But it's the traditional macroeconomic instruments, some combination of devaluation and monetary and fiscal deflation that are really relevant to correcting the balance of payments problem. And the objective is to let the country correct that and then get on with life as before, as soon as possible.

As I see it, the big need in this area hasn't anything to do with conditionality. It's to give the Fund the new power to agree to standstills, the sort of thing that was discussed by Anne Krueger in her speech two weeks ago, so that especially in a capital account crisis it has an instrument with which it can take the immediate pressure off the Fund and allow its rather traditional type of measures to take effect.

I think where one gets into much more need for structural measures is when one comes to the Poverty Reduction and Growth Facility, as it's now called. The new program there is to get the country to draw up its Poverty Reduction Strategy Paper, and basically it lays out its intended program, and that's supposed to be done in consultation with its own population, even more than with the international institutions, the multilaterals but also the bilateral donors. This is indeed envisaged as a partnership between the various groups, but primarily it's the country itself that's expected to draw up the program and take ownership of it.

And then the sort of role, the natural role of the IMF, it seems to me, is to make sure that the macroeconomic part of the program hangs together, that this isn't a program that is heading for--that is inconsistent and is, therefore, going to run into problems in the short run because of macroeconomic imbalances. And as I see it, there is one major problem in the way that the IMF is set up to go about that, and that is that it does its accounting wrong.

When it looks at the fiscal accounts of the country, which is a crucial thing, then for aid-dependent countries it actually puts the aid receipts below the line. They're current account receipts. They don't create indebtedness. They should be above the line. And in some of the aid-dependent countries this creates problems that the IMF does appear to have an excessive zeal for fiscal contraction in circumstances where it is not appropriate.

I don't believe that the Fund always does that. In fact, I read this morning the Argentines are now saying that the IMF was irresponsible for having endorsed the previous administration's excessive fiscal expansion. And I have a lot of sympathy with the present Argentine Government in that respect. So I'm not saying this always is a problem, but I think it is a problem in the aid-dependent countries, and it's a problem for a very fundamental reason, namely, that the accounting is done wrong, and that should be remedied.

Now, when it comes to all the other issues--well, there are one or two marginal issues which could be argued to be either Fund or Bank responsibilities, and I think Joanne mentioned in that context the financial sector, trade, and the tax system. But even if you count those as marginal issues, there is still a whole range of other issues that are absolutely fundamental to the sort of growth programs that the Poverty Reduction Growth Facility is intending to promote, things from governance, from the institutional infrastructure for a market economy, liberalization, privatization, social expenditures, financial sector, infrastructure, various other sector programs, environment. You can go on and on. And all of those, it seems to me, are the areas in which the Bank has had--traditionally it has built up a comparative advantage, and I indeed cannot see the sense of the Fund starting to compete over that vast range of areas.

So the question then becomes how does one coordinate between the two institutions in this type of program? As I see it, there are three basic possibilities. One is for the IMF to lend simply on macro conditions and for the World Bank to lend its poverty reduction support credits on structural conditions and for them to go along more or less independent of one another. And I can't say that seems to make much sense because if you don't get both of them roughly right, then the country is going to run into trouble not too far down the road. So I don't think that would be a good solution.

A second one would be for each of them to lend with their own set of conditions, but to look for some sort of approval by the other, and that is cross-conditionality, and I'm a little bit not clear as to--it seems to me that is basically what is happening, but Joanne says there's no cross-conditionality so I guess I don't quite understand how this is being organized, and maybe she or Masood can come back and clarify this for us.

The third solution, of course, is to consolidate the lending into one organization, and that's the solution which the report to which I was signatory last year of the Overseas Development Council suggested would be appropriate, and we argued that the right--given that most of the issues were ones with which the World Bank had the primary responsibility, that was the natural place to concentrate that type of lending, and so the PRGF should be moved from the Fund to the Bank, and that still seems to me to be a logical type of reform to look at. And then the Fund would continue to be a lender--it would continue to have a role in this type of program in certifying that the macroeconomic conditions held together and made sense, but it would continue to have a role in these countries when they ran into balance of payments crises, but much more the traditional type of role in these low-income countries, and doubtless one would need to have cheaper credits for them, but they would be given in response to the same type of problems, and there wouldn't be an attempt to get into this longer-term development lending, which is what the PRGF basically does.

Now, you might argue isn't there a fourth possibility, which is to abandon conditionality and have both organizations lead in saying whether the country strategy is appropriate as regards the outcomes in a particular area. This is sort of--this is looking at the common-pool propose of Ravi Kanbur which says that in the new conditionality, it would make sense for really all the donors to do the same type of thing that is now happening with the Poverty Reduction Strategy Papers, that the country draws up its program and puts it before the donors, and then the donors say how much money they're prepared to put in, and then the Fund provides guidance on whether the program is appropriate on the macro side and the Bank on these series of structural issues.

It seems to me it's still necessary there to decide whether IMF approval depends just on the macro or whether it also depends on Bank approval. So I don't see it really gets one away from that dilemma, although I am very much in favor of moving in the direction of the common-pool proposal for lending to these low-income countries. But I don't think it gets away from that particular dilemma.

So let me--finally, there's one Fund facility--I'm not sure how much it's still used these days, but where maybe the structural conditions have a more central role, and that's the Extended Fund Facility. This was a facility created in the '70s on the argument that some countries had balance of payments problems that were never going to be resolved without structural change in the economy, and so the IMF should be prepared to lend longer term and to get into structural issues of those types of countries. And maybe there's a case there for retaining structural conditionality more or less as it seems to have evolved, but as regards both of the facilities where the bulk of the Fund lending goes on nowadays, it seems to me that one ought to move back to a much sharper division of labor.

I take Joanne's point that everything matters to everybody, but, nevertheless, I think making progress really does depend on trying to establish a fruitful division of labor, and one shouldn't simply allow a relapse into each organization doing everything, which I don't think would be productive.

MR. AHMED: Thank you very much, Joanne, and thank you for also introducing this point about the distinction across different kinds of Fund facilities and, more importantly, different kinds of Fund roles in different groups of countries. So I think that will be something that no doubt people will want to pick up on.

I see a number of people who want to contribute, and could I just get, to get a sense of the time, a show of hands of the number of people who are likely to want to say something or to raise a question? Okay. So we have seven or eight people so far, and we have about half an hour. So I just encourage you to use that time accordingly.

Let me start over there, and could I ask you to do what John asked, which is to go up to the microphone and identify yourself, please.

MR. LEVINSON: Jerry Levinson. It seems to me that the discussion has largely revolved around organization and process, and the underlying assumption is that the Fund and the Bank know, to use Boughton's observations, good policies and Joanne's idea of a model. And that's precisely the issue, what is at issue. Do you know? And is it reasonable to expect the Bank and the Fund to know?

And my question is--first, just let me back up, and I won't take any more time. I'm not sure that Boughton's characterization of the bad old days is too--or I am sure it's too simplistic because the bad old days had one virtue, that is to say, I was the Assistant Director of the aid mission in Brazil during the '60s, and the Fund and the Bank would come down, and they would concentrate on just what Joanne said, that is to say, the macroeconomic conditions most directly related to the balance of payments. And essentially it said to the country how you distribute the burden of adjustment is a problem, a political problem within your own society--it's for you to solve.

But it also left the countries room for different roads to development, different paths to development, different balances between the public and private sector, different modes of collective bargaining arrangements arising out of the particular history of the countries.

So my question to the panel is this: Do the Fund and the Bank believe there are different paths to development which will indeed involve in different societies a different balance between the private and public sector, different emphases upon privatization, different collective bargaining arrangements, rather than what seems to be the case now, which is an almost mindless insistence upon privatization under any and all circumstances, regardless of how it's carried out, and a single collective bargaining arrangement that, in reality, is a shorthand for labor market flexibility? In effect, launching the Fund in the '80s into structural reform really derived from Secretary Baker's 1985 Seoul, Korea speech in which effectively he laid out an American model of capitalism, and that seems to be the model that the Fund and the Bank are promoting.

MR. AHMED: Thank you very much.

Let's take three or four questions.

My name is K.K. Samanta. I'm a consultant. I will raise here the same basic structural points which I raised in the meeting of the World Bank Annual Conference on Development Economics. Now, your method of dealing--you have all--all the speakers have done something very unconstitutional by referring to countries and peoples. Your constitution only allows you to deal with the governments and speak with the governments.

Now, when you speak of Argentina, certainly you are not speaking about the country or people of Argentina. You are dealing with the little Finance Ministry there, and your country department and maybe your [inaudible]. You have not consulted the chambers of commerce of Argentina. You have not consulted the professional bodies of Argentina which have functioned maybe for hundreds of years, and they have a huge reservoir of expertise, which you never care to tap or to take their views or their suggestions. This you are doing with every country, and every country is very much frustrated that something is going on between the Finance Ministry and the World Bank and the IMF which is not serving the purpose of the people at all.

The government wants to survive. He wants to take a loan. He comes to IMF. Their Finance Ministry goes to the World Bank, takes some loan and survives. What good does it do to the country?

So what I suggested to Mr. Wolfensohn is this: that now the time has come to get out of the old World Bank and the Finance--the International Finance Corporation from being an old relic. Go into the next century. Involve the people, their representatives, the chambers of commerce, the professional bodies and NGOs, and reduce the participation of the governments. Most of them are corrupt and inefficient; otherwise, they wouldn't have come to you.

[Laughter.]

MR. AHMED: Okay, great.

MR. Samanta [continuing] And I want to ask the same question. When will you let in in your organization the professional bodies and the chambers of commerce with full power and facilities as you offer to the governments?

MR. AHMED: Good, and I'm sure my colleagues will want to answer both on the facts and on the implications.

I'm going to work my way down here and then come back here. I have two people here, and then in the back.

I'm David Roodman with the World Watch Institute. I want to first just thank--congratulate the IMF on the relative transparency with which it has run this review of conditionality. I think that's something of a break with the past and is a welcome development, and I hope that that sort of development will be forthcoming more in the future.

That said, I am somewhat concerned about the nature of this inquiry. There seems to be a great deal of focus on streamlining and also focus, which is certainly warranted, but I think it also prematurely narrows a whole discussion of the strengths and weaknesses of conditionality and how to fix it. The subtitle of this seminar is how much is enough. I think there's a bigger question, which is why hasn't conditionality worked very well on its own terms and what is to be done about it, and there are many answers to that. One is quite likely for streamlining and focus, but there are also a lot of other problems: pressure to lend within the institution and pressures applied to the institution, which maybe the IMF cannot really solve, the fact that providing finance to a borrower government can actually bolster forces within the government that are opposed to reform, and so on.

James Boughton acknowledges this at the end, talking about the need for more selectivity, which I very much support. But it seems to me mostly to have been dismissed or relegated to footnotes in the IMF documents.

So I think I should just stop there and say I'm concerned that the IMF still isn't demonstrating capacity to really honestly criticize itself in a way that I think is ultimately needed for its own health.

MR. AHMED: Thank you very much.

I'm Jo Marie Griesgraber. I'm Director of Policy for Oxfam America. And I'd like to second the first three set of questions, but also add that I'm sure the Fund is not supportive of monopolies, and I think there's a monopoly on policy prescriptions in the approach that's taken, even when there's the target set and the government is invited to design its own policies to arrive at the desired outcome. Ordinarily there is an acceptable route to that end and those that are not.

And so I guess I'm very concerned that even in the capacity building and the training that's offered at the IMF, it's more of the same or the same. I don't think there's a diversity welcomed in how to achieve the desired outcomes.

I'd also want to second Jerry Levinson's concern on the fact that it tends to be contractionary rather than growth-promoting policies.

And, finally, kind of the chronic theme of the non-governmental organizations is how do you know what works, because the Fund is so reluctant and apparently doesn't know how to do the ex ante impact assessments. What do your policy prescriptions mean to the poor, the poor disaggregated by sectors and by regions of the country, and what does it do to the environment? And I think until that task is accomplished, you probably should just put a freeze on your policy recommendations because you don't know what you're doing if you don't know how to anticipate the outcome and demonstrate that.

Sorry to be so harsh, but I have limited time.

MR. AHMED: Thank you very much.

I think I had three people here, so I've got Jacques Polak, the gentleman behind, and then there was the--yes?

MR. POLAK: Jacques Polak I would like the speakers to address this question: Conditionality in the Bank and conditionality in the Fund are two entirely different things. There was a story in the press this morning that the Bank had lent $400 million to Colombia to finance its fiscal reform. Now, fiscal reform doesn't cost any money. You have to hire--it certainly doesn't cost $400 million to organize fiscal reform. The loan by the Bank of $400 million for fiscal reform is essentially a bribe to introduce--to induce the Colombians to engage in fiscal reform.

The distinction I would like to make is that in the Fund conditionality is used to support its loans. In the Bank it's the opposite. The loans are introduced to support its conditionality.

I'd like the speakers to address this question.

MR. AHMED: Thank you very much, Jacques.

The gentleman here?

My name is Maxim Kishmir, and I'm an economist at International Equity Management. My question to you all is: What are you (?) cooperation with the capital market is part of your work plan for the next decade or two, or are you simply trying to cooperate with each other across the street and ignore the capital market? If yes, then what is the mechanism you are contemplating, and how are you going to make it work? If not, then no, just say no.

MR. AHMED: All right. Is that pretty much it--or I have--oh, I'm sorry. You were right between me and the screen here.

My name is Brett Schaefer. I'm with the Heritage Foundation. I guess I would disagree with a couple of the commentators already on emphasis. I heard the quote up there saying that everything matters. My response would be that if everything matters, nothing matters, and I would ask you to maybe take a step back and prioritize. What is your top goal? What is the thing you want to accomplish most?

And I found the comment on selectivity to be very interesting, and I was wondering if you could answer very briefly, just name four or five countries that in your opinion don't deserve loans.That's it.

MR. AHMED: And I think we have the last two, and then I'd like to ask the panelists to come back. The lady over there.

I'm Elizabeth Drake from the AFL-CIO. I have a question, first of all, about specialization and how, if the Fund only focuses on macroeconomic conditions, which I think is appropriate for it to do, it can also ensure that the social impacts of their macroeconomic policy prescriptions are positive rather than negative. I wonder if this is a question of Bank-Fund collaboration. I think in the context of the PRSP there is a commitment to do social impact assessments of both the policies in the PRSP and the lending that comes out of the PRSP. And I just wonder how you see your role in terms of the social impact of your programs if we do have a more clear division of labor, which I think is desirable, but carries difficulties as well.

And on the question of flexibility, on the Fund being flexible in the conditions that it applies to countries, I guess I would also like an example. Do you have an example of when a country has said we cannot comply with that condition or we would not like to and the Fund has changed its mind and said, okay, instead of privatizing 87 corporations in the next year, you can only privatize 42, or something. Just a concrete example of what sort of things you would be flexible on and what things you wouldn't be flexible on, where you would say, no, you absolutely cannot have a non-transparent budget, it has to be transparent. I mean, I think you cannot just say the word flexibility without getting a little bit into the substance of what things you'd be flexible on and what things you wouldn't be.

Thank you.

MR. AHMED: Last gentleman.

My name is Jerry Nisenson. I formerly worked with the Treasury through the '80s that reviewed many of the programs, and I do some teaching now. I just had one question. I'd like the panel to address the subject of exchange rate, what the Fund's position on exchange rates for the individual countries and conditionalities, in particular the prevalence of using a fixed exchange rate as an anchor of the whole program. The most recent example is in Turkey, which failed.

Thank you.

MR. AHMED: Okay. So we have a large number of questions, and let's try and see if we can deal with this. I think that they fall into some groups, and perhaps if you'll allow me, I'd like to ask my colleagues to deal with them in that manner.

There is a set of questions about how much flexibility there is in the approach. Do we accept that there are different paths to the same outcome, or do we have a sort of single model which we try to rigidly apply? Are there limits to what countries can come up with in terms of their programs? And, in some sense, do we know what we're doing? I think it was Jo Marie Griesgraber who sort of said let's put a freeze on all policy prescriptions, and I'm not sure where that would lead one to. But, in any event, does one have a sense of what it is that we are doing? Just one block of questions.

Then there's a second set of questions that's related to the work--the review itself. Have we too quickly narrowed down the focus on looking at streamlining and focusing of the review? Have we looked at the experience on conditionality more broadly, what works, what doesn't? And in that sense, it would be useful to perhaps say one word on that.

Then I think there was also a question about whether, if everything mattered, do we have a sense of putting priorities into our work--how do we prioritize in different country contexts?

Maybe if I could ask my colleagues to just start on those, and then if any of you feel that your questions haven't been addressed--and then there was Jacques' question as well, which was: Is the link between conditionality and lending the flip in the two institutions? And it would be good to get a perspective from both on that.

Why don't I ask to do this in reverse order, and ask John--and John, in particular, as the sort of author of the Washington consensus, and now sort of where we are today. What's your read on how much flexibility you see in the two institutions in terms of accepting different approaches, different paths to development? And then I'll whether Joanne and Jim from their perspective agree with that or not as well.

John?

MR. WILLIAMSON: Clearly, I don't believe that every country ought to be doing exactly the same thing at the same time. Even if one has a reasonably coherent view of what's the best way to go about economic reform, it's going to be highly--what you think a country should be doing at a particular point in time is going to be highly dependent on its initial position. And then I think there is indeed something in the point that Jerry Levinson made that one wants to leave a certain amount of scope for countries to reflect their own choices.

But I think in the past we may have left too much--been too liberal on that. For all the talk about the Washington consensus having dominated the policies of these institutions in the last ten years, it's not that long ago that they were lending to Nyerere's Tanzania and Kaunda's Zambia. You know, I pick those two because they're two who, I think, individually one has to regard them as fine, upstanding people who believed in certain principles, but the principles were disastrous for their countries. And let's be clear about it. We have no business to help countries--help rulers run their countries into the ground by lending to rulers who are going to do that.

So I think one can take the--one can indeed be too willing to lend to rulers, if one simply rules out the Mobutus, I mean, they're easy. Yeah, of course, you can name five countries that you wouldn't lend to, you know, Afghanistan and North Korea and Iraq and--Afghanistan until about two weeks ago, and Angola and so on. But that's not the point. The difficult cases are not the Mobutus. They're the Kaundas and the Nyereres. And do you lend to countries where the rules are well intentioned but are doing things that are leading their countries into really big problems? So I think you want to give--you certainly want to give a certain amount--you want to allow a certain--you recognize local circumstances, and you want to allow a certain amount of latitude for countries to do as they think. Yes, we do sometimes talk about countries and confute countries and governments, and, of course, the Fund and the Bank deal with governments and not with countries.

But here I would say that I think the PRSP really is a move forward at least that works, and my reading is that it's something that an attempt is being made to make it work. And so a country can set out what it thinks its policies are, and unless the international institutions and the donors think that they have got it wrong, then the lending goes ahead. There has to be a point, I think, at which you don't throw good money into a country, but that--and the essence of that is supposed to be consulting with the country's own population and getting away from simply talking to the government. But you can't cut the government out of it. The government, after all, especially if it's a democratically elected government, does have more claim to represent the people than the chamber of commerce or the NGOs or anybody else.

Do I have another two minutes, or am I--I think Jacques Polak's contrast is indeed an interesting one. I'm not sure that--again, I think that the PRSP process really gets around this, doesn't it? I mean, in both cases, then, the institutions are lending in support of a set of policies, and the policies--but I guess that really is the Bank approach, isn't it? That you are lending in support of a given set of policies and not simply putting conditionality into place so that you make sure you get repaid, which, in any case, with IDA is not a major consideration, and if some of us had our way and turned IDA loans into grants, it would be even less of a consideration.

I've now provoked Joanne, and so I'll hand over to her.

MR. AHMED: Joanne, in addition to responding to John, I thought there was actually two points in particular. One, this question of--John, do you want to come back with a double-hander before you pass on?

MR. WILLIAMSON: I'm sorry. There is one other thing that I do want to comment on, and that's the exchange rate issues raised right at the very end. Because, I mean, the Fund's traditional position has been countries can pick their own exchange rate regime, and then we help them make it work. At least that's the position as regards big countries like Argentina. And I really think that the Argentine case means that we ought to consider whether that is responsible.

MR. AHMED: Okay. Joanne? And I was going to say that there are two points that I thought particularly relevant. One was this issue of prioritizing clearly is very important for the Bank for the reasons that you were outlining and how does one set those priorities. And, second, the question that was raised about social impact analysis and how would we do that going forward, but anything else also that you want to pick up on.

MS. SALOP: Thank you, Masood. Let me start with Jacques Polak's question, which I was glad that John answered, because I was sitting here trying to figure out which way, and I found it to be quite a mind teaser in terms of whether the loan is supporting conditionality or conditionality is supporting the loan. So I want to come back to you afterwards and see if we can sit together and figure out--very timely in terms of the work we're doing on adjustment lending to really focus on what is the sizing of adjustment lending operations. So it's very much relevant also to our own work going forward.

Two, I wanted to respond also to a point that John had raised previously when he was provoking me, which was on cross-conditionality. He said, I think, it didn't--it sounded like cross-conditionality to him. So let me just restate what our position is. It is that in terms of the--in terms of taking things, items, loans, things that need to be approved by the Board to the Board, basically we don't put clauses in those proposals that say dependent on approval of something by the other institution. So in some sense, it's a legal matter, that indeed at the end of the day, if there are going to be conditions or that the Board is going to have to approve certain recommendations, they're going to have to be on the merits and all the merits included in it, so no references to the--no references in the conditions, et cetera, the contract to the other institution.

And if indeed there are conditions which say the Fund believes are essential to the success of its program, critical there, and those very conditions are critical also to the Bank staff--perceived by the Bank staff to be critical to the success of its program, we have agreed generally to move in the direction of trying to harmonize the wording of those so that we don't cause headaches for borrowers and to have it, you know, we put a condition it's every three months and they put the condition it's every four months, so how does one square this circle. So to really move in the direction of a more harmonized approach so as to reduce the transactions costs for borrowers in those cases.

On the prioritization question and everything matters--and here I plead guilty of loose lips. Of course, I don't really mean everything matters equally. John picked me up on that as well. But what I did mean is that throughout the 1980s, realization, say, in the Bank that clearly macro matters, you can't do good projects and have good, successful development outcomes in a climate in which the macro is a mess. It just doesn't work.

And, similarly, in some cases for the Fund, they have found they need structural conditions, whether it's institutions, whether--whatever, certainly on the social side, et cetera. These things are essential for the sustainability of the macro, whether it's the political economy or whatever, but that those things take the point. It's not everything matters everywhere equally, et cetera. You have to prioritize. That's the business. And so--but, nevertheless, the idea that you can't be in a silo, here's the macro, here's structural, and the two don't talk, no, not that kind of world.

And that is really, of course, in some sense the recognition of the new development paradigm of the late 1990s, which is, you know, basically all agencies, and all are recognizing that it's both sides of the house that matter, the macro and the structural, taking the point that there are particulars that matter.

Now, to prioritization and how do we work, also to some of the questions is there choice, can countries--what do they put up or what and who decides and all of those, that whole set of questions. A word about how we work in principle.

In principle, a country puts its issues on the table, their vision, and, in fact, that's what we're seeing carried out now in the PRSP program. Country articulates its priorities, articulates hopefully informed by all the learnings and advice and all of our two cents, et cetera, but it's the country priorities they put on the table.

We, the Bank, presumably the Fund, too, it's our job to do diagnosis, a professional diagnosis of whether we think all the programs set out there are going to work and to advise the country, and then on the basis of a dialogue on these things, figure out what indeed we think we can support in that program, and then to worry about results in terms of results, in terms of on the ground the social results, growth results, et cetera, and feedback loops then into the--to start the process again.

Of course, it's not one size fits all and, of course, it's country issues starting at the beginning, country priorities, but the job of Bank and Fund staff is really to do this professional assessment and to advise the country as well as to our Board in terms of whether what they put on the table adds up to a program that will be sustainable in getting the countries the results that they actually want. I think there's no shrinking from that.

I'm looking at my list. I think I've covered it.

MR. AHMED: Jim?

MR. BOUGHTON: Thank you, Masood. I'll be very brief because we're running short of time. You've all been very patient.

I'd just like to focus primarily on this question that David Roodman raised about why doesn't conditionality work, and, you know, that is the key question that's driving everybody's thinking about these issues, both inside the Fund and among the critics of the Fund.

The answer is essentially, if I had to formulate an answer in one word, I'd formulate the answer as ownership, that we have not always formulated programs in a way that fosters ownership of the right kinds of policies.

Jerry Levinson asked: Does the Fund recognize that there are many paths to development? Yes, I think the Fund does recognize that, but it also recognizes that there are many paths to un-development or non-development. There are many ways that a country can go wrong. There are many ways a country can go right. And I think the task of the Fund in this whole process has to be to separate A from B, that you recognize that there is a whole range of solutions to a problem. But there is also a whole range of non-solutions. And it's the non-solutions that the country's been following up until now or they wouldn't be here in the first place.

So we narrow it down to a range of feasible options. We ask the country which among these feasible options it prefers. That is precisely what's going on in Argentina right now, where Argentina has been in an unsustainable situation, it clearly needs to change something, and the elements of the change have been widely discussed. They would include fiscal adjustment. They might include exchange rate adjustment. There are lots of things that could be done.

The whole range of policies that's adopted has to add up. It's not the function of the Fund to go in and say you have to do it this way rather than that way. I mean, in a way I think this is the answer to the question Elizabeth Drake raised about, you know, what's an example of the Fund being flexible. Right here as we speak is the right kind of flexibility. It's up to the authorities to come to the Fund with a program that will work, and the Argentine authorities are doing their best to get a program through their own government, through their congress, through acceptance by the provinces, through acceptance by the public. It illustrates more clearly than anything that I could say up here the importance of a broad range of ownership within the country for a set of policies.

You're not going to get that by having the Fund go in and say, well, if you did it this way rather than the other way, then this would work.

Now, maybe there are people both inside and outside the building, people in this room today who would have the view, gee, I have a pretty good idea what would work in Argentina, I could go in and write down a program. You could read the pages of the Financial Times today and find people who are saying there is one way to do it, it's the way I've been telling them they should do it, and that's what they should do. But those days are gone, and that's not the direction we're trying to move in.

What I'm encouraged by in the kind of responses that I've heard from the questioners today is that I think you've all made some very valid criticisms of specific cases or of instances in the past, but those aren't the directions we're moving in. When Mr. Samanta said that the Fund should go out and talk to a wide range of people, that's exactly what we're doing. I spent an hour and a half last week in a very productive meeting with representatives from international trade unions, from the IFCTU. These are people from developing countries, from transition countries, who are trying to deal with serious labor issues in their own countries. Those people are also talking to Fund missions when they go to those countries.

So I think a lot of the concerns that you've raised are very valid concerns, but they don't necessarily reflect what's happening now, and I hope that, as I said at the end of my initial remarks, that a few years from now that it will be even less reflective of what's happening on the ground.

Thanks.

MR. AHMED: Thank you very much, Jim, and I'd like to bring this to a close. I just want to say one word, which is that we've really been talking a lot about process, which is about conditionality and how it's applied and should it be detailed or should it be less detailed and how do the Bank and the Fund work together.

There are, I think, two complementary dimensions which we haven't talked about today, but on which equally people are working and need to think through. One is the substantive content of policy advice in different circumstances, and we got into that in some ways, you know, questions about exchange rate, questions about privatization, I think somebody said "mindless privatization." I'm not sure anybody's doing that, but, clearly, there is a parallel discussion that we need to continue to have on the substantive content of advice on what works and what doesn't work and what knowledge does one have about what works under what circumstances.

I would disagree, for example, with what Jo Marie Griesgraber's contention that we have no idea about what works. I think we have pretty good ideas about some things that work. We have pretty good ideas about some things that work in some circumstances, but also areas where we don't have a very good idea of how things work and what the links are and what the lags are. So I think that's a different parallel discussion which is -- [tape ends].

And the second point I want to make is that the other complementary part of it is that process matters in the following sense. It's not simply the content. It's not simply the conditionality. It's how you interact with countries and how interaction institutions work and whether they're decentralized or centralized, or the way in which their staff interact with counterparts and with other agencies. And so process in that sense is very important, as important, maybe, in terms of generating the kind of ownership of programs as the precise content or the precise modalities of conditionality. And that's a more difficult discussion to have because it's a more amorphous discussion. You can't get your hands around those issues as well.

But I would recommend thatif you have the time and the interest, take a look at the paper on ownership that we just put on the website. It tries to begin to deal with some of these issues of how one can interact differently. And if any of you have reactions to it or advice stemming from reading it or thinking about the same kinds of issues, feed those back to us, too, because I do see these three pillars: the content; the modalities of conditionality; and then the process of interaction, which all have to work in the same direction and without which you really aren't going to get the maximum impact from any of them.

So I just wanted to add these brief comments and thank all the panelists and, of course, I would like to thank all of you for taking the time to be here and for sharing your ideas and concerns. Thank you very much.

Mr. STARRELS: One final commercial interruption before we say goodbye. For an audience of this size, I can't believe this is all the cards we have! We really want to hear from you again.So put them in this basket and thank you again for joining us today and we look forward with pleasure to seeing you again on January 10, 2002 to discuss the Euro.

MR. AHMED: And duplicate copies of the same cards are very welcome here.

[Applause.]