Transcript

But I think as our speakers today will no doubt argue, the issue of governance reform at the Fund is much broader than simply voice and representation, and let me give you one example.

Some people have argued that the effectiveness of IMF decisions, including in a crisis lending context, would be considerably enhanced to a more rules-based approach to those decisions. But this raises the issue of, in the event that we were to adopt a more rules-based approach, how this could be reconciled with a model of political oversight where national capitals have a lot of influence on the day-to-day decisions of the IMF. And if we were to move away from that model of political oversight, what would replace it and how, in particular, would political accountability for the Fund's decisions, which clearly is essential, be achieved and hopefully be strengthened?

Fortunately, we have a group of speakers today that have thought long and hard about these issues, and we'll begin with a paper by Lisa Martin and Ngaire Woods, "Multiple-State Constituencies in the IMF: An Agency Model." Lisa Martin is the Clarence Dillon Professor of International Affairs at Harvard and author of "International Institutions in the New Global Economy," as well as a number of other books and papers on international political economy and international institutions.

Right after Lisa's presentation, we'll move to our panel where we have five speakers, and we'll begin with Jim Boughton, who is the IMF's historian and author of "Silent Revolution," a history of the Fund during the 1980s.

Charles Calomiris, the Henry Kaufman Professor of Financial Institutions at Columbia, who has written extensively on the need for a new global financial architecture and IMF reform. Many of you will know Professor Calomiris' views from his contributions to the report of the Meltzer Commission on which he served, but long before that he was a strong advocate of replacing ex post negotiations over conditions for IMF lending with ex ante rules to constrain the circumstances under which the IMF provides assistance.

Following Charles, we'll have Carlo Cottarelli, Deputy Director of the Fund's Policy Development and Review Department, who's the author of a recent paper on the trade-offs inherent in IMF governance reform.

Then we will have Jong Nam Oh, IMF Executive Directors for the constituency that includes Korea, Australia, New Zealand, the Philippines, and ten other countries. Dr. Oh was formerly Secretary for Finance and Economy to the President of the Republic of Korea.

And, finally, we will have Tom Scholar, the United Kingdom's Executive Director at both the Fund and the Bank. Prior to this, Tom Scholar was private secretary to U.K. Chancellor of the Exchequer Gordon Brown.

So over to you, Lisa, for 15 minutes, please.

MS. MARTIN: Thanks. It's great to be here.

The topic generally that we are working on is obviously these multiple-state constituencies within the IMF, and we're adopting a principal agent framework to think about the dynamics and functioning of these constituencies.

What drew me in particular to this topic is that I study international relations, international organizations generally, and the IMF structure of representation is somewhat unique in that different member states are, at least in a formal sense, represented through different mechanisms, and that isn't all that common among IOs. So I am interested in sort of the implications of this particular structure of representation generally. So as I'm sure all of you know, within the IMF some Executive Directors represent just one state while others represent multiple states. And this is the unusual feature that, in particular, drew me to this particular issue.

What we do in this paper is first we use a principal agent approach to develop some general propositions about the dynamics of multiple-state constituencies. Then we provide some descriptive background on multiple-state constituencies which actually is not all that well known. I didn't know much about it until I started working on this quite recently. And then what we do is we move on to concentrate on the interaction of voting rights, basically, and interest within these multiple-state constituencies to see how they actually might function in practice. The work is fairly preliminary, so I'm looking forward to getting some feedback on it.

Let me start, first of all, just by describing the situation, which I think some of you know extremely well, but others might not be quite as familiar with. The Executive Board at the IMF obviously represents all member states, but in different ways. Five countries, those with the largest quotas, each appoint a Director, and they may not waive this right; that is, the five largest quota holders can't say, "I don't want to appoint my own Director. I want to be part of a multiple-state constituency instead." They don't have that right. But the other members all participate in elections that then decide the remaining members of the Board.

Now, the Articles of Agreement of the IMF do specify the total number of Executive Directors so you know how many seats you're voting for when you participate in these elections. But other than that, there is a great deal of flexibility for how these constituencies might form, how these elections might actually turn out, and, in fact, there has been a fair amount of flux over time in the arrangement of these constituencies.

Just a little more descriptive detail. In these multiple-state constituencies, the elected Executive Director wields the votes of all those who elected him, and the Executive Director must cast those votes as a whole without dividing them. So if, for example, the constituency ere divided, say, on the stringency of conditions in a particular program, the ED cannot split his votes and say, you know, I'm going to vote this way. He has to cast those votes as a whole.

So, currently, five of the 24 Executive Directors are appointed, the five largest quota holders. Three other Executive Directors represent just a single country: Saudi Arabia, Russia, and China. And that leaves 16 of the EDs who are elected by groups of countries, these multiple-state constituencies. And these groups range in size from four countries making up a constituency to 24 countries in one of the African constituencies.

A little bit on how these multiple-state constituencies have evolved over time. Originally, the Articles of Agreement specified that voting power among the EDs had to be relatively evenly distributed. In fact, the original Articles had it very evenly distributed, with all EDs having been between 19 and 20 percent of total votes. But, over time, the distribution of votes among the EDs has become much more unequal, with the United States now having 17 percent of total votes on the Board, while the multiple-state constituency that has 24 African countries in it has just 1.4 percent of the votes on the Board. So that has been one interesting, just historical development over time—the great increase in inequality in terms of the number of votes wielded by each ED.

What's also happened over time with respect to the multiple-state constituencies is that fairly stable electoral groups have formed over time. You do have countries moving back and forth often in an attempt to increase their influence within a constituency, but now they're fairly stable electoral groups. We don't see a lot of change from year to year.

The groups have become more regionally coherent, so now most multiple-state constituencies have a specific regional focus. There's the Asian group with Australia and Korea in it.

When you look at countries that shift from one constituency to another, sometimes this has been a desire to create a more regionally cohesive group, but often it really is a competition for influence within the constituency. So if a state feels that its ED is consistently not representing its interest, either a threat to move to get more attention from the ED or actually moving to another constituency if you're promised a more powerful position within that other constituency is something that we've seen happening from time to time.

We've also seen that each constituency has developed its own internal decisionmaking procedures which vary quite widely, and this is something that I'm actually looking forward to doing more comparative research on as to actually document what the internal decisionmaking procedures are and how they vary from constituency to constituency.

Okay. So what we're trying to do now, moving on to the more analytical parts of the paper, we're trying to develop some typologies of these multiple-state constituencies and then think about, again, what the implications are for representation and influence.

First of all, what we do is we focus basically on voting power within the constituency, and looking at voting rights within each constituency, you can roughly divide these multiple-state constituencies into three categories. You have dominated constituencies that have one dominant member, one large state with a lot of smaller states. You have relatively balanced constituencies in which you have a small core of more powerful members, often just two that tend to balance one another, but sometimes a slightly bigger core. And then you have a third category of what we're calling egalitarian constituencies in which you have a relatively equal distribution of voting rights within the constituency.

And so one way to start looking at how would we develop this typology and figure out where do the different constituencies fit, a first very simple cut at this would just be to look at the share of votes held by the largest vote-holding member. So that's what LVM is, the largest vote-holding member.

First of all, we have a type where the largest vote-holding member has greater than 75 percent of the total votes within the constituency, and three of these multiple-state constituencies—Italy, Canada, and India—are of that type. Then you have a middle category in which the largest vote-holding member has between 40 and 75 percent, and obviously we have this group now of six constituencies that fall into that group. And then, finally, those with the largest vote-holding member have less than 40 percent of the total votes, and we have seven constituencies that fall into that group. So that's one just really sort of quick and dirty way to classify the different constituencies.

But another problem is that that doesn't really give you a very good picture of the distribution of votes within the constituency. It's just looking at the most powerful, the largest vote-holding member. So another thing that we can do—and here Ngaire and I are drawing on some work that she's done with another colleague—is to use the Gini coefficient just to look at the distribution of voting power within the constituency. And if you do this, you see that we get a slightly different, although not terribly different, picture of which would fall into these different groups.

So, again, we have these three constituencies with a very high Gini coefficient. We're going to label these the dominated constituencies and expect that, respectively, Canada, India, and Italy will play a really dominant role within their constituency.

Then you have this group of relatively balanced constituencies. For example, Australia and Korea hold a lot of the voting shares within the one in which Australia is the largest vote-holding member and so on.

And then we have the egalitarian constituencies again, where you have a fairly diffuse pattern of vote holding within these groups.

One final thing we can do, again, just in terms of thinking about patterns of voting power, we can look at what we're labeling runner-up power, and here we're just asking about if you look within each constituency at the number two player, basically, what is the state that has the second highest percentage of votes within that constituency, and what percent of the largest vote-holding members' votes do they hold. So we find that there are eight multiple-state constituencies in which the second largest vote holder has more than half the vote share of the largest vote-holding member.

So, for example, in the South African constituency—actually, Nigeria holds the chair right now, I believe—South Africa is the largest vote-holding member, but Nigeria has 93 percent of South Africa's votes, so it's relatively balanced. And you can also see down the list. So of these eight constituencies, you could argue that there is not likely to be just one dominant member but a second player who will often offset or have an interesting dynamic with the largest vote-holding member. So that's looking at, again, just using vote shares to think about a typology of different sorts of constituencies.

Now, what do we do with this? Well, what we are ultimately interested in doing is thinking about what are the constraints on the Executive Director and how is the Executive Director going to represent the votes of the interests of these constituencies? And to do so, we use the principal agent framework to develop some general propositions. What we need to assume to make this all work is to assume that the group of states that is represented by a single ED is a collective principal. This collective principal delegates its right to vote on IMF policy to the ED. So it makes sense then to think of the Executive Director as part of a chain of delegation. And if you wanted to play out the whole chain of delegation, it's obviously quite long. You could go back to voters in each member state, to the government of that member state, then to the ED, then from the ED to further on within the IMF, but, specifically, the link of the chain that we're focusing on here is that from member states to the Executive Director. So what we're doing here is to treat the Executive Director as an agent of member states.

If you draw generally on a principal agent framework, you can think about the factors that are likely to influence the effectiveness of representation. First, obviously, principals and agents' fundamental interests, obviously how much heterogeneity or conflict of interest is there within a constituency, and then between the constituency potentially and the ED. Decisionmaking procedures within these collective principals will matter a great deal. If there are very clear decisionmaking procedures for giving the ED instructions and so on, that could make a big difference as opposed to have a relatively ad hoc procedure. Monitoring capabilities are going to be a big deal. For relatively small states, it may be quite difficult to keep close tabs on what is actually going on.

The competence of an agent is always a major factor in the willingness to delegate authority within this agency framework. And, finally, you'd want to think about the incentive structure that the agent faces, and here we're thinking about things like, for example, future career prospects once you're done being an ED and that sort of thing.

So these are general categories of factors that we think will influence the effectiveness of representation. What we're focusing on first is the first one of these, that is, the interest dimension, because we think that's going to give us a great deal of explanatory leverage.

So what are some of the implications of an agency framework for thinking about these multiple-state constituencies? Well, the influence of each of these factors, of these explanatory factors, will depend on the type of constituency. In some, there will be a lot of conflict of interest; in others, there might be minimal conflict of interest. So it really depends on the type.

We do expect that in the three dominated constituencies, it's unlikely for the Executive Director to have a great deal of autonomy since there will be a close tie between the dominant state and the ED. However, in the more egalitarian constituencies, especially the two that do not have runner-up power, the ED's autonomy will probably depend quite heavily on these internal decisionmaking procedures. The ED could potentially have a substantial amount of autonomy if he didn't have clear decisionmaking procedures in these types of constituencies.

On the other hand, if you look at the seven constituencies that we classified as balanced or the four that have runner-up power so you could designate two, you know, sort of dominant members of those more egalitarian constituencies, then we think that the degree of autonomy that the ED has is likely to depend very much on the divergence or convergence of interests between these two most powerful states. If they agree on things, they're likely to constrain the ED quite heavily. On the other hand, if they disagree quite a bit, that will give the ED the room to play them off against one another potentially.

Okay. So just—I know we want to turn to the rest of the panel, but just to tell you about some of the more specific types of propositions that we're developing and starting to collect some empirical evidence on, what we want to look at first of all is the interaction between the typology developed on voting shares and patterns of interests. And so what we're arguing is that patterns of interests are largely determined at a fundamental level by whether the member is a borrower or likely to be a borrower from the IMF or not. And so we can characterize along this interest dimension then three types of multiple-state constituency: those that are made up entirely of non-borrowers, rich states; those made up entirely of borrowers; and those that are mixed, that include both borrowing and non-borrowing states.

So what we can do—and I think this is sort of what I wanted to lead up to here—is to put together now these two dimensions: the typology of the constituency based on voting shares and interests on the vertical dimension. And so when we get here, if you look at the first row here, these are constituencies made up entirely of borrowing countries. And what we expect in these is, again, there are not likely to be terribly deep conflicts of interest. Obviously, these constituencies, their members will disagree on various issues, but because they're in sort of the same fundamental relationship to the IMF, there aren't likely to be very deep conflicts of interest there. Likely, in the one multiple-state constituency that has no borrowing countries in it, we don't expect terribly deep conflicts of interest. And, therefore, these problems of effective representation might not be huge.

On the other hand, if we go down to the third row, these constituencies that mix borrowing and non-borrowing states, here's where we see the more potential for failures of effective representation, basically, and the degree to which representation is going to be effective will depend on this distribution of voting shares, internal decisionmaking procedures, and so on.

So looking at this bottom row in the mixed constituencies, we would expect that Italy and Canada would have a great deal of influence over the ED in their constituency. In the others, again, we're going to have to—and so the bottom two cells here, we're going to have to look very closely at internal decisionmaking procedures, monitoring capacities and so on, to understand exactly how the dynamics between the ED and the rest of the constituency play out.

I'm out of time.

MR. OSTRY: Thank you very much, Lisa.

I'm going to ask that questions be held off until the end, and we'll move straight to the panel. So, Jim Boughton?

MR. BOUGHTON: Thanks, Jonathan.

I think Lisa and Ngaire have given us a very interesting, excellent analysis, really, of some of the complications for IMF governance that are presented by this multi-state election process for Executive Directors. And it's a nice complement to sort of what you might think of as the main strand of literature in this field, which focuses usually much more on sort of the maldistribution of voting power and of quotas across the entire membership. So I think it's a nice complement to the rest of that literature.

If I can characterize this literature, Jonathan mentioned in his introduction to the session that the terminology we usually use inside the Fund for describing this is voice and representation of different countries in the Fund, how that's decided and what effects it has. Ted Truman has a nice paper that he presented at a conference last month in which he uses sort of a pithier expression of shares and chairs, which is, I think, both easy to remember and probably a little bit more descriptive.

Yesterday, Simon Johnson in his remarks at this conference in the morning suggested that every paper should have a movie tie-in, and he proceeded to talk about that a little bit. But I think it's more logical that every paper should have a tie tie-in. So since I'm going to be talking about chairs this afternoon, I decided to wear a tie that has chairs all over it. So that's my tie tie-in for the day. I'm not sure I'm going to try to apply that to every paper I discuss, but it worked today.

So I'm going to, in doing this, try to sort of pull our thinking back a little bit toward this main strand of the literature and look at this general question. How are voting shares and quotas distributed? Is there a problem? What's the problem? What can we do about it?

The first point I'd like to make is that it's not a simple problem of saying there exists out there an ideal distribution of voting power and what we have to do is to find the political solution, a political way to get there. The problem is actually much deeper than that. The problem is that there is no ideal or generally agreed upon distribution of voting power. And that's because there are—there's not one gap here between reality and the ideal. There are two gaps.

First of all, there's a gap between the actual structure of the Executive Board, the actual structure of interests in the Fund, and the structure that would accurately represent economic importance and influence in the world economy. That's one gap, and that's the gap that we usually focus on.

There's also a second gap, and that's the gap between the actual structure of voting power in the Fund and the distribution that would accurately represent the global welfare, the interests of the world's people. And if the Fund is an institution that is supposed to promote the global welfare, then we would like to have influence within the Fund accurately represent that global welfare.

So you have two gaps, and there might be conflicts between the actions that you would take, the changes you would make in order to close one gap and the actions that you would take to close another gap. I think it might be useful to think of these as, first of all, there's a power gap between actuality and the ideal, and then there's a legitimacy gap, if you will, or a welfare gap or a people gap, or whatever you might want to think about that.

So I'm going to—in the few minutes I have your attention this afternoon, I want to do two things. I'm going to, first of all, put on my historian's hat and try to explain a little bit more of what this problem is, and then I'm going to try to pretend to be an outside observer from the Fund and suggest some more or less radical ways to deal with it.

There actually are three problems we're trying to overcome. The first problem is that the existing distribution of quotas and voting power in the Fund does not just represent economic calculations. It represents both economic and political calculations. If we go back to Bretton Woods when these decisions were first taken, the formal input was a formula that was actually developed within the U.S. delegation at Bretton Woods primarily by Raymond Mikesell, who was working for Harry White at the Treasury. That's a purely economic concept. But, unfortunately, Mikesell had left his laptop at home when he went to Bretton Woods, so he didn't actually econometrically estimate this equation based on trying to minimize some loss function. What he actually did was he was given a set of relativities by his bosses, by Harry White representing the Roosevelt administration, influenced by other people and so on. So he had a set of relative quotas that he was trying to achieve, and then he calibrated the numbers in his formula, which we are still using today, to try to meet those relativities.

So what were these important political considerations? Well, first of all, the United States was the dominant country. It held 75 percent of the world's gold reserves and was clearly the dominant economy in the world. Secondly, the United Kingdom was clearly the dominant partner, the number two country, no question about that. Restoring prosperity in France and other Western European countries was a major goal of both of those countries, so giving them a high profile in the institution was important. The Soviet Union was a key ally. Remember, this was taking place in the middle of World War II. The Soviet Union was a key ally. Giving them an important role was important. China under Chiang Kai-shek, major U.S. ally in fighting the war in Japan. Again, giving them a big role was important. And the prosperity of Latin America was important to the United States.

So what we had as an outcome was a situation where the top two positions went to the United States and the United Kingdom. Then we had the Soviet Union in third place just below the United Kingdom; China and France then filled out the top five. And as Lisa mentioned, the top five got to appoint their own people and not have to participate in an election of Executive Directors.

Then Stalin decided that the Soviet Union would not join the Fund, so that enabled India to come up and take on the number five position, where they had initially been number six. And then, in addition to that, two seats on the Executive Board were set aside under the original Articles of Agreement for special seats of Latin America. So that was the way it started off.

Now, to say what the effect that this might have had, it brings me to the second problem in all this, the first problem being that it looks like an economic process, but it's actually a combination of economics and politics, and, therefore, it's a lot harder model.

The second problem is that there's a strong element of inertia in the way the IMF was structured through the Articles of Agreement and the way it was actually run. We've had 12 quota reviews over the last 60 or so years, eight of which have resulted in large increases in quotas. But none of these reviews has ever been zero based. They take the starting point and you have an increment in which most quotas just go out in lockstep over time. Now, there are a lot of exceptions to that, but that's the basic rule. So that today's quotas in the Fund still reflect these decisions that were made in the middle of World War II in 1944.

Now, to see how that might affect things, consider what would have happened if the Soviet Union had, in fact, joined the IMF in 1945 and then had received just average quota increases in the intervening years. At Bretton Woods, the Soviet Union got a quota of $1.2 billion, which, as I mentioned, was just below that of the United Kingdom.

Now, Raymond Mikesell, who was trying to make this formula work, the most he could come up with for the Soviet Union was about $750 million. So, you see, they actually got a much higher number than what we call their calculated quota. So it reflected their political importance.

Now, if they had gotten average increases since then, the actual percentage that they would have had would have declined a little bit, as other new members joined. But it wouldn't have declined all that much. So by the time the Soviet Union had split apart at the end of 1991, they would have had a quota share of around 10 percent of the total.

Now, that's important because, as many of you know, there's a veto system in the IMF in which the United States is the only single country to have a veto because it has more than 15 percent of the vote. And so major—the most important decisions, the really big decisions about finances and so on, require an 85-percent majority. So people think the U.S. has a veto, nobody else has a veto.

Well, the Soviet bloc plus China, if you think of China as their ally during much of this period, as a group would have been a veto player. Right? So we would have had a very different balance of power within the IMF if the Soviet Union had joined. But that's all fantasy because they never did join. The Soviet Union never did join the IMF.

So that when Russia came to join at the beginning of 1992, they had to start all over again. They didn't benefit from this inertial effect that they would have benefited from. So they said, well, gee, you know, it seems to us, since we're the biggest part of the Soviet Union, we should have something close to 10 percent. And everybody else said, whoa, wait a minute, that doesn't seem right to us. So what they actually got was 3 percent. So by not joining in 1944, they lost all the benefit of this inertia, and they ended up with a much smaller size. So instead of being number two, they were number nine in the hierarchy.

Now, compare that with Korea. I don't want to steal all the thunder from my colleague Mr. Oh here who is going to talk about Korea, I'm sure, a little bit later on. But Korea joined the IMF in 1955 with a quota of $12.5 million, not $1.2 billion like the Soviet Union. They got $12.5 million, which was 0.14 percent of the total voting power in the IMF in 1955.

Now, Korea has, in fact, gotten a little bit more than average quota increases over the years, but not all that much. So where they started off being just below Burma and Ceylon in 1955, they're now just below Denmark and Norway. Well, both the GDP of Korea and its amount of foreign trade that it does are three times as big as either Norway or Denmark, but it's still below them. So just as the Soviet Union lost out by not joining and not getting the benefit because it joined after its fortunes had fallen, Korea lost out by joining before its fortunes rose, and so we have this effect. So that's the second problem.

The third problem is that there are these different views as to how to overcome this inertial effect.

So the obvious solution to this would seem to be to have a zero-based quota exercise. Just forget about history, start all over again, and look at the current distribution of economic and political power in the world, and come up with that.

Well, why hasn't this been done? Well, actually, it was done. Five years ago, the Fund commissioned Richard Cooper of Harvard to do a zero-based quota exercise, and he came up and he did an excellent job by all accounts, wonderful study, careful study of how quotas should actually be distributed. The problem was nobody liked the results because the Cooper Commission showed that one problem was the U.S. quota was too small and we have to have a bigger quota for the United States because at 17.6 percent now, it's smaller than its actual share in the world economy.

African countries' quotas are way too large. They get 2 percent of the total. It should be more like 1 percent because Africa, you know, we're talking countries that are poor and they do very little international trade, so according to the Cooper report, their quotas are too big, we should cut them roughly in half. So this was a non-starter.

You know, you could go on and on about this, but basically because we have these two competing criteria, trying to close the power gap representing actual distribution of power in the world, and trying to close the welfare gap to try to come up with something that actually makes this a politically legitimate institution, they work in opposite directions.

So I have very little time left, but I am not going to take ten seconds each on seven things that I think can be done to actually—not try to close these gaps, because, as I'm trying to suggest here, that's really impossible, but to try to minimize the effects of these problems, and that I think is the way of the future. What I'm going to do is sort of an increasing order of difficulty.

The first thing is to delink the amounts that countries contribute to the Fund, break the link between that and their actual quotas. In other words, make quotas less important for what money people actually do. And that's easy to do because, in fact, we're already doing it because we have all these different ways that countries can contribute money to the Fund.

The second thing we can do is to delink the amounts that countries can borrow from their actual quotas. We do that through the Strategic Reserve Facility and so on. You know, Korea borrowed over—what was it?—1500 percent of their quota, I think it was, which is something that—20 times their quota.

The third thing we can do is if we have a poor distribution of voting power in the Fund, don't take votes. Right? So this is, again, something that is, in fact, often done, that to the extent that you can, you operate by consensus and just forget about voting power.

Fourth—and this is important because it's something that often gets left out of the debate, but I think it's extremely important. The fourth thing we can do is to try to form more effective coalitions among countries. The U.S. is not the only veto player in the IMF. The European Union, if it acts together, has a much higher vote than the United States. And so whenever they can agree and speak with one voice, the European Union is a veto player. The G-24, the developing countries, have a much higher aggregate vote than the United States. If they can agree on an issue and act together, then they're a very important veto player. So more effective coalitions is the fourth thing.

The fifth thing—and this is, again, something you might not think of in this context, but the fifth thing is to streamline and focus the work of the IMF. When all that the IMF did was to deal with monetary and fiscal policy in countries and say, gee, you guys are spending too much or your monetary growth is too high, nobody cared about a welfare gap. Nobody cared about democratic legitimacy for the institution. But if the IMF is going to tell countries how to structure their economies and whether to privatize the banking system and decisions of that nature, getting much more deeply into policymaking in countries, then we have a problem. We have to close that gap. So by streamlining and focusing our work and pulling that back in, which the institution has been trying to do for the last four years, is very important.

And then very quickly, two radical solutions. One is restructure the executive or restructure the IMFC by creating a council and so on. I think some of my colleagues may be saying more about that, but I just want to emphasize that getting that far is much more radical.

And the final thing I want to mention is even more radical. Rethink this whole role about special majorities in the Fund. The only reason we worry about the U.S. having a vote of more than 17 percent of the total is that that gives them a veto and nobody else has a veto.

You know, Tom Friedman in his most recent book about globalization has pointed out that the world is a lot more flat than it used to be. But the IMF is not a lot more flat than it used to be. You know, we're still—we live in a flat world, but the IMF is still a mountaintop with a bunch of Americans sitting up at the summit. And so if we can find some way to do that, we'd reduce the problem. But that's pretty radical, and it's not going to happen by 3:30. So I'll stop there.

[Laughter.]

MR. OSTRY: Thank you, Jim.

We have a deficit of about five minutes, close to that now, so let's call on Charles Calomiris now for 12 minutes, please.

MR. CALOMIRIS: Well, it's humbling to take the podium after Jim for a couple of reasons. First, he knows so much more about the IMF than I do. And, second, he thinks his proposals are radical. And that really makes me wonder where I'm going to come in, because I look at them as moving the chairs around on the deck of a certain ship.

I want to talk about how a New Yorker thinks about the IMF. I'm a New Yorker now, and I talk to people in New York, and to people in other countries, about the IMF, and I think that perhaps the most important thing I can tell people at the IMF is what these people outside of Washington think about you, and why that matters, and what that tells us about the need for governance reform.

I want to start with the proposition that when you think about governance, you have to first decide what you're trying to accomplish; the proper governance structure follows from what you are trying to do.

Let me give you an example that I think everyone who's been following the news in Germany can understand. The German codetermination system is a particular kind of corporate governance system, one that is designed to prevent shareholder value maximization by corporations in Germany. That system, however, is now being challenged. In the current globally competitive environment, German corporations cannot compete as the result of codetermination. Volkswagen is a perfect example. Codetermination, according to estimates, destroys about 45 percent of the value of Volkswagen. Thus, a corporate governance system that's hugely value-destroying, and which was chosen purposely in the 1960s and the 1970s in Germany exactly to achieve its objective, which was to give workers voice and to make corporations not pursue value maximization, is now being challenged.

Is the German codetermination system a bad corporate governance system? Of course it is if you want to pursue value maximization, but the real question is not about corporate governance, but rather about the appropriate goal to pursue. That is the real question for Germany, and for the IMF. I hope you'll agree with me that until we talk about what the IMF is supposed to be doing, we can't really decide what the governance system should look like. I personally want Germany to move away from a codetermination system, but that's because I think that it would be much better for the German people if their corporations pursued value maximization. So there we are. That's where we start.

I was at an event earlier this week sponsored by the IFE, an organization that brings together professionals at the pinnacle of the financial services industry, including hedge fund managers, risk managers, executives who run trillions of dollars in New York. I was on a panel with my old boss at the IMF, John Lipsky, who's the chief economist at JP Morgan Chase, and David Derosa, a critic of the IMF.

I was cast as one of the defenders of the IMF. That is something that should give you pause. Not only was I cast in that role, but I willingly served that role. My views about what the IMF should do and how it should be governed haven't really changed much over the past decade, but the rest of the world's views have. One of the biggest problems that we IMF reformers face right now is the view that the IMF is not worth fixing. I say this with sadness of heart. The IMF is losing support in the world, and many simply believe that the IMF does not have an important role to play anymore. My own view is different; I think that the IMF is not succeeding in playing its proper role, but that it does have a role.

To many of my New York friends a combination of forces—in particular, globalization, securitization, the communications revolution and deregulation - have changed the world so fundamentally that they believe that financial systems really don't need central banks much anymore, much less a global lender of last resort like the IMF.

I think that's dead wrong. What this view fails to appreciate is that the growth of securities markets and sophisticated new financial instruments have not meant the end of financial intermediation, and financial intermediation is inherently subject to the sudden seizing up of markets. I like to remind my New York friends about a certain institution called CFS Inc. that had done $1.6 billion of outstanding securitized debt, which had come to naught, and the securities holders managed to get about $70 million back from that. I reminded them of Long-Term Capital. Temporary disruptions to the orderly flows of financial assets happen when the sufficiency of the equity capital and credibility of financial intermediaries is suddenly called into question, leading to a liquidity crisis. Securities markets are increasingly vulnerable to such shocks. For example, securitization increases the complexity and the number of intermediaries involved in creating financial assets; it does not eliminate them. Financial intermediaries measure and manage risk, with the blocking and tackling of due diligence, covenants, collateral, exposure limits—all of which can become very tentative in certain market circumstances.

In short, financial intermediation is alive and well, and so are its risks. Despite our progress, under some circumstances—when the assumptions on which new financial innovations are based—are called into question we can find ourselves in a more vulnerable position.

I am concerned about this as more than a hypothetical possibility. First, one of the great engines of growth in the world right now, China, in my view has just sold the world a bill of goods with regard to the new IPOs for its large banks. I believe that there is a distinct possibility of a major financial upheaval in China sometime in the next several years if circumstances in those banks do not improve, and I do not see them improving sufficiently.

Now, I may be wrong about that; we'll wait and see. My concern, however, is that if I am right, we may soon find ourselves in need of some global agent able to help manage the fallout of a China-induced liquidity crisis. Now, I am aware that Japan, China, and other countries have established a mechanism for that purpose, but I doubt (particularly in light of recent political tensions between Japan and China) the sufficiency of that mechanism.

Something else worries me, too. The incredibly thin spreads in sovereign debt for emerging market countries right now, which can't last, particularly in a rising interest rate environment, particularly with the kind of global imbalances that we have right now.

So, I am not ready to concede that there's no potential systemic risks facing the global financial system, but the harder question is: What do all those risks mean about what the role of the IMF should be?

Some commentators think that the IMF should be a coordinator of G-7 fiscal and monetary policies. I think that is a pipe dream. The IMF has no hope of telling the G-7 what to do, or of coordinating exchange rate movements among them, and I think that it's a waste of time to even ponder it. We need to let the IMF do what it can do, and it's not going to tell the United States or Japan or the EU how to run their monetary policy or their fiscal policy.

How about the IMF's role as a coercer of underdeveloped countries? Well, I don't have much interest in that either. First, the evidence of its efficacy in that role is mixed at best. Second, in my opinion, coercion can be questioned on ethical grounds as an appropriate mission for multilateral institutions. Third, in today's polycentric world, where power is being shared by a variety of countries, I'm not sure that Japan, China, Brazil, Mexico, India, Korea, Russia, the United States, and the EU can all agree on what sort of coercion should be applied to, say, Pakistan. Therefore, I don't think it makes much sense to have a global coercer. We don't have the "consensus," so to speak, in Washington that we once had.

I believe that the IMF could play an important role, however, as a coordinator of liquidity to try to prevent and to try to mitigate emerging market financial crises, like the kind I think are likely to occur over the next decade. We've had a quiet six years. The quiet, I think, is coming to an end.

What's preventing the IMF from being able to play that role today? The basic problem is that the IMF lacks the size and the structure to be a credible provider of liquidity. Because of the lack of political consensus about the IMF's role and the rules necessary to achieve it, we are frozen in the past. Reform of the structure of lending has not happened. And the IMF has stayed small, while the financial markets have gotten big. And I'm not just talking about the sizes of current accounts, or about capital flows measured by issues of bonds or stocks or bank loans. I'm talking about derivative transactions, and securitizations across borders, which are huge and growing.

The sad truth is that the failure to reform has made the IMF irrelevant from the standpoint of emerging market financial crisis prevention and management. Let's be real. You don't matter right now. You're not even close to mattering.

What would it take for you to matter? Well, my compatriot Archimedes had something to say about levers. What you need, really, is to become a lever that can move the global financial system during crises. To be that, you need mechanisms that focus on working with, and harnessing the power of, markets, and you need to increase your capital footings substantially.

Right now, that is not going to happen because you don't have the political consensus about what your mission should be, and therefore, you are stuck in paralysis and irrelevance.

What would meaningful reform look like? The IMF needs to focus on developing one assistance mechanism to replace all others, which would be a form of credit line used to deal with systemic problems. This new line of credit would be supplied under credibly enforced rules, defined in advance, that would limit the abuse of the line, in part by limiting the discretion of the IMF staff and board in determining the amount and terms of the credit that could be supplied.

Governance reform is a crucial part of this overall reform of goals and assistance mechanisms, and is crucial to restoring confidence in the IMF.

I will skip a detailed discussion of what it would mean to harness market discipline and how the IMF could work with the private financial system to create credible advanced lines of credit that could actually be useful for coordinating liquidity assistance in dealing with crises. I have written about it before. My views have evolved over time, but have not fundamentally changed, so there is little point in elaborating today.

By the way, some complementary reforms are necessary outside the IMF, too. The BIS needs to reform its policy mechanism, too, to get away from command-and-control capital standards for banks toward market-harnessing capital standards. But that's another topic for another day.

What governance reforms should accompany IMF policy reform? Governance reform is crucial to ensure the credibility of structural reform. First of all, if you're going to set up mechanisms that look more like credit lines, supplied with limited policy discretion, you need to have a governance structure that is not inimical to that approach. It makes little sense for such an IMF to be governed by the finance ministers of its member countries, who can be relied upon constantly to attempt to circumvent any attempts to limit discretionary interventions.

Our panelists today are talking quite a bit about voting rules. But the allocation of voting determines the sharing of rents, but does nothing to limit the pursuit of rents. The IMF board is very good at sharing rents via "log rolling." The G-7 Finance Ministers and all the other Finance Ministers can agree on one thing: Discretion is a good thing, and their discretion is a very good thing.

What governance structure for the IMF would be more compatible with its proper role as a source of global liquidity, supplied via policy rules that establish lines of credit for sovereigns? Perhaps the most obvious reform is to put the central banks in charge, not the Finance Ministers. And maybe we should move the IMF to Switzerland to create a better atmosphere for independent thought.

Of course, now you understand why I thought that my proposals might be more radical than Jim Boughton's.

In closing, I want to say that my criticisms of IMF policy and governance reflect my profound respect for this institution, and for the professionals who run it, as well as my belief that there is an urgent need to speak openly about what it will take to restore the influence of this institution in pursuit of its important mission in the coming years.

Thank you.

[Applause.]

MR. OSTRY: Thank you very much, Charles.

We'll have Carlo now for 12 minutes, please.

MR. COTTARELLI: I just realized how difficult it is to be called Carlo Cottarelli when you follow alphabetically such a good speaker as Charles Calomiris.

I will talk about governance, reforming the IMF governance at the Executive Board. In discussing the role of the Board and possible lines of reform, I think three issues need to be addressed: the first one is the voting rights issue; the second one is the role of the Board in the daily management of the Fund; and the third one is proper modalities for political supervision of the Fund.

These three issues—voting rights, the role of the Board in daily management, and the role of political supervision—have implications both for the legitimacy of the institution and for the effectiveness of the institution. I will address them in turn, focusing particularly on the implications for the effectiveness of the IMF. And just one caveat. I don't have much time because I'm just a panelist, and with more time I could nuance my views. I won't have time. I will have to be concise. So I will have to be blunt in my statements, and I hope nobody gets offended, particularly on that side of the table—and maybe also in front. I don't know.

Let me start with the issue that has perhaps attracted more attention in the last few years—the voting rights issue. I will not enter in the substance, who should get more, who should get less, obviously, but I would like to underscore one point that is frequently overlooked, namely, that addressing this voting rights issue is important. It's good not just for the legitimacy of the Fund, but also for its effectiveness, for two reasons.

The first one is obvious. A multilateral institution like the Fund has to be regarded as legitimate by its members to be effective. Otherwise, we will not be credible. We will not be regarded as objective. And eventually we will not receive the cooperation that we need from the member countries.

The second reason is perhaps less obvious, and it requires some elaboration. There has been over the last few years, in my view, an increased emphasis, an excessive emphasis by the Board on the use across all countries of the same form of procedures in conducting economic analysis. In my view, it's a sort of—what I'm saying is the one-size-fits-all approach has become more common. And my conjecture—it's a conjecture, but I will put it forward—is that in part this, consciously or unconsciously, is the response to sort of unhappiness and frustration about lack of progress on the voting rights issue.

Essentially, some Directors insist on the use of procedures aimed at ensuring formal equality of treatment across countries in our work practices because they feel to be underrepresented in any process that would allow a more flexible and discretionary approach to economic analysis. But, of course, this—I will call it this obsession for formal equality of that has impaired, is impairing effectiveness.

Let me give you just one example. Some years ago, the IMF, as you know, introduced a more standardized approach to assessing debt sustainability. It stands to reason that this approach is not needed for all countries. In some cases, it's obvious that a country does not have an issue, a problem with debt sustainability. In other cases, particularly advanced economies, issues of debt sustainability take a different form. They may be related to long-term demographic trends.

Yet some Directors insist that the debt—because of the equality of treatment, it was necessarily to apply the same approach in the same way to all countries, even if when this was not needed from a technical perspective.

Now, this is just one example among many that I could make, but it does in my view stress the point that I was trying to make, namely, that this unhappiness about the voting rights issue may have led to excessive focus in our operation on formalities—on formalities, equal treatment, doing the same for all countries, rather than on the substance of a more evenhanded treatment of countries.

My conclusion is that addressing the legitimacy issue through the proper instrument, the voting rights—better distribution of voting rights would also help the effectiveness of the Fund. We could be more flexible, selective, and eventually more effective.

Let me turn to the second issue: the role of the Board in the daily management of the Fund. There is in my view a need for a less direct involvement of the Board in the Fund's daily work. One issue is the size of the Board. Here I would like to quote the Crow Report, which is a report—the external report they did in 1999 to assess the effectiveness of surveillance.

The Crow Report stated this issue in very clear terms, and let me quote it. The report said that, "Everything we know about institutional governance indicates to us that the Group of 24 is, to put it mildly, extremely large for useful exchanges of views, discussion, and group decisionmaking." So size is one issue.

The second issue is the Board's difficulty in processing the massive amount of information that is involved in the daily management of the Fund. Lisa and Ngaire's paper noted that delegation is at the core of any principal agent relationship, and obviously I fully agree. It doesn't make any sense for the principal to try to replicate all the activities of the agent.

This situation in a way reminds me of a short story published in 1960 by Jorge Luis Borges. Borges described a group of cartographers who managed to prepare the perfect map of their own country. Let me quote. "In that country, the art of cartography reached such perfection that the cartographers set up a map of the country which was the size of the country itself, and coincided with it point by point. Successive generations understood that this widespread map was useless and abandoned it to the inclemency of the sun and of the winters."

The bottom line, the principal cannot expect to cover all the actions of the agent, as much as a map cannot be expected to cover every single point of a territory.

The Board supervision needs to be maybe more selective. The Board should focus more on the strategic country and policy matters, delegating to management and staff a large part of the daily work. Incidentally, increased delegation of routine matters from the Board to management and staff is one of the proposals included, as you may know, in the Medium-Term Strategy Paper put forward recently by the Managing Director.

Let me move to the third and last issue, perhaps the most difficult one: the role of political forces in the supervision of the Fund.

This issue affects the Board in two ways, because Executive Directors are both political representatives and they are also Fund officials. So when we ask to what extent the Fund should be independent from political forces, we're really asking two questions: first, to what extent management and staff should be independent from the Board as a political entity; and, second, to what extent the Board should be independent from the political forces of the capitals.

These are difficult questions. On the one hand, there is a fairly widespread notion, perception in the outside world that there have been cases of political interference in some major Fund decisions. This obviously would call for, it seems to me, more independence of staff, management, and the Board for political pressure.

On the other hand, it must be acknowledged that the role of political forces remains critical for the Fund, for the legitimacy of the Fund, and also for the effectiveness.

I can tell you, from my direct experience as mission chief, that when you are out there in the field in a mission and you have to take decisions that may displease the local political authorities, it is of great help to know that you have political principals that will scrutinize your actions and that these political principals will, so to speak, counterbalance the political pressures that may arise for staff dealing with sovereign countries. A staff that were abandoned to itself would lack the legitimacy to deal with and the strengthen probably to resist pressure from sovereign countries. Thus, the role of political supervision remains essential.

The question, however, is that I think it is in everybody's interest if the form of political supervision is exercised in a different way, with modalities that are more based on transparency and ex post accountability than on intervention on the daily Fund decisions.

To conclude, just a few ideas that I put forward in a working paper issued a few months ago in which I discussed three possible ways on how to achieve this kind of political—more detached political supervision.

First is to give the Fund more operational independence while enhancing the ex post accountability. Keynes, you may know, when the Fund was established, advocated a non-resident Board, a Board in charge of broad oversight but not in charge of specific decisions. Now, this may not be necessarily the best approach, but I think that some form of more detached supervision is needed.

Second, I underscored in this paper the need to enhance the protection of staff from political pressure. The Crow Report recommended altering the incentive structure faced by staff by making it clear that management will back up staff members who give frank advice in the course of our work.

Third, and last point—and I will conclude—I also stressed the need to reassess the role of the Board vis-a-vis the capitals. There is a widespread feeling that over the last few years, the role of the capitals has increased vis-a-vis the role of the Board, and this has perhaps increased the weight of political consideration.

Thank you very much.

[Applause.]

MR. OSTRY: Thank you very much, Carlo.

We now have Jong Nam Oh.

MR. OH: I'd like to thank the organizing committee for inviting me to express my views on reforming the governance of the IMF and how to enhance the effectiveness of the Executive Board. I would like to thank Professors Martin and Woods for the helpful paper, and I would like to start by looking at the issue of governance a little differently from the others and return to some issues discussed in the paper.

Let me start by offering this perspective: When I am introduced to my Korean friends, their first response is quite often, "Ah, you work for that bad IMF." The question that lingers in my mind is whether that characterization is fair. The answer may be yes or it may be no. If yes, we at the IMF may really be doing something wrong. If no, then the comments may well be influenced by Korea's involvement with the IMF and the negative sentiment that has not been adequately addressed.

In reality, the answer is more complex than both of these simple responses. To explain this view, I would like to comment on Korea's relationship with the IMF over the past five decades since Korea joined the Fund in 1955.

In 1955, Korea's GDP per capita was about $67, and it was one of the poorest members of the Fund. I played golf with one of the founders, initiators of those joining the IMF last Saturday, and I heard that the priority of the negotiation team was how to minimize the quota size in order to make the burden lighter. That was our history some 50 years ago.

In 2004, in the last year, GDP per capita was around US$14,000, and the economic size being number 11 in the world economy. And as we all know, in 1997 Korea experienced a financial crisis and received assistance from the IMF in the form of finance and, more importantly, a program which contributed to the economic success that Korea has enjoyed over the past eight years.

However, the structural adjustment required in the domestic economy does not lend all Koreans to that acknowledgement of the IMF's assistance. In that regard, there is a role to make the work of the IMF better understood as the perception that the IMF is bad is not necessarily an accurate one.

I'm also pleased to say that following the IMF program, Korea's relationship with the IMF has developed from strength to strength. However, the development of the relationship is incomplete as the governance arrangements at the IMF have not taken into account the very significant change in Korea's economic size and its contribution to the world economy. In this regard, my Korean friends, who say to me, "You work for that bad IMF," may have a valid point in their own observation.

Korea adopted the IMF's advice, grew as an economy but did not gain recognition from the institution. Korea is not alone. Many emerging market economies have a significantly greater economic size than they did, say, some 10 years ago. Quota may be reviewed every five years, but the level of representation has not materially changed. The under-representation of countries such as Korea, Turkey and Mexico, clearly shows how their actual quota shares are significantly out-of-line relative to their economic importance.

I wonder how many of you are aware that I am the first Korean Executive Director of the IMF, while the economic size of Korea is number 11 in the world, and our quota size is 28 out of 184 countries as of now.

Further, the idea of Asian Monetary Fund, the IMF does not like to hear that, but the idea of an AMF reveals the discontent of those countries with a huge gap between actual and calculated quota based on the formula.

To that end, the representation at the IMF is a challenge. Globalization is bringing countries closer together, and the demand for IMF advice remains strong. However, the legitimacy of that advice is weakened when its source does not adequately represent the membership that forms the mandate for delivering the advice.

Yesterday the Managing Director raised this point in his speech at lunch, and I agree with his argument. The issue should not be regarded as a zero-sum game. The legitimacy of the IMF is challenged as membership no longer fairly represents members. The proposal to address the issue at the 2006 Annual Meetings in Singapore is a step forward on this issue. In that regard, leadership by Australia and other countries through the G-20 and similar groups, provides an important path for developing options and building consensus for a decision that is to be taken by all 184 member countries, not G-X countries. This is where the challenge of governance lies.

As a panelist, my comments would not be complete without discussing points raised by Professors Martin and Woods. The paper joins one of many pieces of work that discuss how the role of Executive Directors can be improved. It is a useful contribution because it identifies differences between the dynamics that exist for Executive Directors who represent a single country, and those who represent multiple countries like myself. As an Executive Director, who represents 14 heterogeneous countries, I am pleased to express my views based on my own experiences of holding Executive Director and Alternate Executive Director positions.

I do not intend to comment on all five areas in the paper, but I would like to make particular mention of the following observations. On interests, most Executive Directors are not in a strict principal-agent relationship. Indeed, they are in a more complex and constrained relationship. Therefore, any application of the principal-agent model needs to take into account this complexity that limits the autonomy of EDs.

I agree that constituencies could be largely categorized as set out in the paper, namely, dominated, balanced and egalitarian. However, one should not assert that in relatively balanced constituencies, a diversity of interests would give the ED the autonomy as he or she played members up against one another. I do not consider myself I have any more autonomy than other colleagues.

On monitoring, I agree with the view that the monitoring of an elected ED by member states will depend on the transparency of the processes in which the ED takes part, and the capacity of the country and its specialized agencies to collect, digest and use information on what the ED is doing.

In our constituency, we have made efforts to enhance transparency in our work in the ED's office. In formulating our gray/buff statements, we always seek input from member countries, particularly on the policy issues. We provide member states with (1) a report on Board discussions where they capture our interventions in the Board meetings; and (2) a monthly constituency update which can provide a summary of key Board discussions and a summary of items that may be of specific interest to members.

And finally, we hold constituency meetings twice a year during the Annual and Spring Meetings. We also hold bilateral meetings with each member country to report performance and to seek views. Our office recently surveyed our members on surveillance issues, which is a further example of monitoring the needs of our constituency member countries. I also recognize that monitoring can take place through advisors and staff situated within the ED's office.

On competence, while recognizing that longstanding EDs tend to have more capacity to exercise influence, more emphasis should be placed on the level of competence of EDs to be able to pursue the constituency's interests effectively within the Executive Board.

I would also like to point out that a country's foreign policy is a significant determinant of the country's competency at the Board. The foreign policy impact can be at several levels. For example, instructions on policy issues, or broader foreign policy such as the leadership demonstrated by the U.K. on debt relief issues.

In concluding, I would like to emphasize one fundamental point. When we discuss reform, what that means is, there is something that needs to be fixed and if all of us agree that we need to reform the governance of the IMF, then I would like to see action as early as possible instead of talks.

Thank you so much.

[Applause.]

MR. OSTRY: Thank you very much.

MR. SCHOLAR: Thank you, Jonathan. Since one of the themes of the day is chairs, I'm going to stay in mine if you don't mind.

[Laughter.]

MR. SCHOLAR: In case anyone takes it while I'm gone.

[Laughter.]

MR. SCHOLAR: When I was first appointed to the Board four years ago, a friend of mine wrote to me to congratulate me, and sent me an article which described the IMF Executive Board as "A retirement home for second-rate bankers and third-rate bureaucrats," or maybe it was third-rate bankers and second-rate bureaucrats, but anyway, it wasn't very complimentary.

When I arrived here some months later, I found that the Board was not in fact at all like that, but I did find it a rather confusing place, and I spent my early months here wondering what on earth was the point of this institution that I found myself a member of.

My early experience was of a huge number of very vibrant debates amongst an impressive and rather daunting group of people, and this seemed to me to be like a series of rather intense university seminars. But the link between that and a function producing an outcome as part of the work of the institution was not immediately apparent to me. But I've been here four years now, and I've gradually worked my way to an answer to this question, which is maybe not the answer, but it's my answer. So I thought I would use my 12 minutes this afternoon to offer some thoughts on that general question of what the purpose of the Board is.

First of all, I think I should say that my conclusion is that the basic structure established by the founders of the institution has stood the test of time, and I think the structure of an Executive Board which is responsible for the affairs of the institution has turned out to be very farsighted as a model.

The Fund is an intensely political institution. It's true that its mandate is economic and social. It's true that its work is heavily informed by technical, analytical, economic, technocratic expertise, but the Fund's advice and the actions that the institution takes touch on the most intensely political affairs of its membership, and in particular, obviously, issues such as taxation and government spending (including decisions on both the levels and the distribution). This is the stuff of revolutions if things go wrong. So the Fund is, I think we should recognize, a political actor in the end.

The Board, in that context, provides the political legitimacy and accountability, without which I don't think the institution would have survived. It's a vehicle which connects the staff and the management with the political shareholding, and the governments with the Fund as their own institution. The Fund and the Bank often feel that they don't get as much support from the governments of the world as they would like when the going gets tough, and maybe sometimes there's a case to be made there, but I think without an Executive Board which connects the two, the Fund would find life a great deal lonelier, because after all, even when a government criticizes what the Fund has done, nevertheless that government was a part of the decisionmaking process through its representation at the Board.

Finally, I think the Board is the only global vehicle for reaching global consensus on issues of economic policy, and country issues. There's no other forum for doing that.

The process through which the Board does all these things is quite a subtle one and quite a slow one, and it often means that in a particular discussion on a particular country or a particular policy issue, what is said doesn't result in any immediate impact, but does gradually over time move the center of gravity.

We should judge the Board by looking at the outcome, and what we have seen from this basic governance structure is an institution which has been able to adapt to enormous change in the environment over the last 60 years, and has proved robust to those changes, and has survived a series of major crises which I think would have destroyed the institution without this link between the institution and the governments that established it and that own it.

This brings me to the issue of independence and whether there should be more of it, and here I've read over and over again the sentence in the introductory material circulated before this discussion, which poses the following question: "Does involvement of national capitals in decisionmaking result in undue political interference that undercuts the global mandate of the IMF?" And one often hears such questions posed.

And I've read that over and over again. I'm sorry, I simply don't understand the question. I don't understand this tension which is posed between political so-called interference and the global mandate of the IMF. I mean, where did the IMF get its mandate from, and how is it going to keep its mandate if the countries and the governments that own it don't continue to have faith and confidence in the way it works? And I don't think that I would characterize that as interference. I think I would characterize that as a way of providing support.

I think the analogy which one often hears with the notion of central bank independence really doesn't hold in the case of the Fund.

If you look at the classic model of inflation targeting for a central bank, you have first of all a government that sets an objective. It's a clear, single objective, an inflation outcome. The connection between the actions of the agent, the bank, and the outcome is not perfectly well established in the literature, and of course it's a complicated business, but there's a fairly direct connection between the two. And it is very easy to monitor, and there's high frequency data which allows you to monitor the success of the agent in performing this. You have inflation figures every month.

Now, none of this at all holds in the case of the Fund, and I really don't see how you can apply this analogy to an institution like the Fund.

And let me say also a word on the rules versus discretion debate. There is one extreme here - the argument that the whole problem with the Fund is that it has far too much discretion, and is subject to political interference, and that what you really need is a much more independent and technical institution that can apply clear rules and then follow them. I really can't see that working.

The first question is: who will write these rules? Well, maybe the institution itself or the staff of the institution can write the rules, in which case I don't think they would have much legitimacy in the world at large, and once the going got tough I think we would also find they didn't have a great deal of support.

Could the governments, the shareholders write the rules? Well, they can, and every six months the shareholders write a text. It's called the IMFC Communique. I don't think it's easy to imagine such a communique ever setting out some rules that were sufficiently clear to provide a clear guidance to difficult judgments in the case of a crisis. And so I think what one would end up with in a pure rules framework is one where the rules are dreamt up in-house.

And the first time that the rule pointed to a genuinely tough decision, where, for example, the Fund decided not to support a country facing a financial crisis - perhaps for very good reasons, and where perhaps on the technical merits of the case, that would be the right decision - on that occassion I think we would find that the Fund would not survive the criticism that would follow from that, and people would say, "It's impossible to have this unaccountable organization making these decisions. We have to have a better way. We have to have some oversight."

So although I quite understand all of the criticisms about discretion, and in the case of central banking we think the world has found a good answer to them, in the case of the Fund I don't see how we would get there.

Let me say a few words though on how I envisage the Board operating and how it might operate in a different way. Here I have to say I have a lot of sympathy with what Carlo Cottarelli said. I think the basic model of the Board's engagement, particularly on country issues, is about 30 years out of date. It belongs to a world in which the guiding principles were one of peer review and confidential advice, and the Board was a forum in which a country could get advice from its peers, and this advice would remain within the hallowed doors of the 12th floor.

In today's world of mass communication, availability of information and transparency, I think this makes no sense at all. The real judge of what the staff say in their Article IV reports is not the Board, but the world that reads them (two-thirds of them are published). I think that is correctly where the test lies.

I would see an improvement in the functioning of the Board if in its surveillance work we could move to a situation where the Board set the staff a remit for surveillance, setting out the objectives, issues to be covered, and priorities, the Fund carried out its surveillance, published the results of that, and then every year or so we would have an assessment as to how well the Fund was carrying out its mandate.

On programs, I think there's scope for improvement here as well. There's a tendency for us all to gloss over the tensions that are inherent in difficult decisions and difficult judgments in programs. You see this both within the economic sphere, where there's always pressure to make programs "add up" in some way, so that over the lifetime of the program, everything comes back smoothly and stability is restored. I think partly as a result of that pressure, we often see over-optimistic assumptions in programs.

But there's also—and let's be frank—often a tension between economic analysis and political decisionmaking, and I think it's a mistake to try to twist the analysis to justify what seems to be the likely outcome. I would prefer to see a much harder edge to the economic analysis presented in program documents. That should also be accompanied by a much more realistic, probabilistic assessment with fan charts and all the other analytical tools that central banks love to produce. And then if the Board wants to reach a different view from the view which the staff in its best judgment presents, well, that's fine as well, but let's at least be clear what the decisionmaking process is and what the factors that have come into it are.

So I would see scope for improving the functioning of the Board in those two areas A third area is that there should be much greater public accountability of the Executive Board. The Board is the decisionmaking body of the institution. It's more or less unknown in the wider world. I think each of us individually should be more visible in our own countries and in particular with our parliaments and with our publics; but also, more generally, the Board as a body ought to be more visible and more accountable.

I was going to say a few words about quotas, but I see that my 12 minutes are up, and you're holding up the sign too. Thank you.

[Applause.]

MR. OSTRY: I'd like to thank all of our panelists. Charles has to leave promptly. I would like to still open up for questions from the floor. If Charles has to leave, you may be able to get him to answer a question on his way out, but please step up to a mike and identify yourselves, and direct your question to either the panel as a whole or to a particular panelist.

QUESTIONER: My question is for Mr. Boughton. I really like the way you frame your proposals as proposals that would minimize rather than trying to come to grips with these welfare and power gaps, which seems very politically difficult. Do they minimize the impacts or the negative impacts, and in particular the proposal for streamlining the work of the Fund?

I really think you're on the right track there because this seems to be two pieces of an equation. The more intervention in the country's affair that you have on one side of the equation, the more you have to rethink the way the Fund is structured, the Fund's governance is structured. And if there is no hope for reforming the Fund's governance, then you have to take a look at the others, and reduce the level of intervention. However, any time the discussion comes up at the Board it seems it's met with resistance by the larger shareholders on the grounds that this would mean supporting (?) the financial integrity of institution.

Do you see any way you can get around that objection in terms of your proposal? Thank you.

MR. OSTRY: Let me try and collect a few questions first, and then we'll turn it back to the panelists. Any other questions, please?

QUESTIONER: [Off microphone.]

MR. OSTRY: Thank you.

Other questions?

QUESTIONER: [Off microphone, inaudible]—on the approach of, you know, resulting governance [inaudible]. I agree with Mr. Boughton that we should resolve [inaudible] because I think if we don't—we resolve only the economic gap, then [inaudible].

The second is just a comment on the [inaudible]. [Inaudible] officially we—it is too [inaudible] for us. We have what we call [inaudible], some of developing countries, and the way it works for us is almost like [inaudible] from the leader which is the [inaudible], and also the factor [inaudible].

MR. OSTRY: Thank you.

QUESTIONER: I have a basic question. I want to know if you consider that redistributing the votes is possible, how it can be done, and what countries are most opposed or they're opposing it?

MR. OSTRY: Thank you very much. Perhaps one more question.

QUESTIONER: [Off microphone, inaudible] satisfaction index on the numbers, and whether the election [inaudible] is a sort of subject for a mechanism in terms of fairness, in terms of [inaudible] aspect? If all the numbers are really the same [inaudible].

The other [inaudible] point of view [inaudible], how successful has the IMF been historically in predicting [inaudible]?

MR. OSTRY: Okay. Let me turn it back to the panelists now. Why don't we start with Tom Scholar, and then move this way?

MR. SCHOLAR: Thank you very much. Jeromin's question. Yes, of course, Article I sets out the mandate, and if you could just look at Article I in each case and deduce from what it says there what the Fund should or should not do, or assess the actions of the Fund against that mandate, and if there were a simple agreed objective, and a technical way of reaching that judgment, then I would agree with you. But I think in practice, on any difficult issue, it's a matter of judgment what is the correct course of action which will lead the Fund to act in accordance with its mandate.

And on any such issue, you can argue it all sorts of different ways. And what the Fund needs, where it's taken some difficult or controversial stand, is support for that stand.

Now, saying that is not the same thing as saying the Fund's actions should be nothing more than the political preferences of its members. Its members need to provide an interpretation of how to achieve that mandate. It's an interpretation which I strongly believe should be based on social and economic considerations rather than geopolitical or strategic considerations. But in the end it's a question of judgment, and political judgment has to enter into that.

I think the solution to the dilemma is the one I suggested, of much greater transparency about the decisionmaking process, in which it would be very clear to anybody that examined the issue why the outcome was whatever it was, and if there's a contrast between the advice of staff and the decision of the Board.

And I should also say, as part of this, we need much greater transparency within the institution about the arguments within the staff, because it's not as if it's plausible that there's some tablet which says this is what staff thinks should be the case. We know that there are often huge internal arguments, and that's the sign of a healthy and effective institution, but I think as an institution we should have the confidence also to expose some of these arguments to the world at large.

The last thing I'll say is the IEO wrote a very interesting report on the Argentina crisis, and I thought the most interesting part of the report was the section on decision taking. And it was only about half a page long, perhaps for understandable reasons, but there were some very interesting comments in there on the role of capitals, and in particular G-7 capitals in the decisions that were reached.

I really think that that provides some very interesting material which hasn't been examined enough, and a greater transparency is what I think might solve some of these dilemmas.

Thank you.

MR. OSTRY: Thank you, Tom.

MR. OH: I will try to answer how effective has the Fund been in terms of predicting the crises, and all of us know, two pillars of the IMF are, number one, crisis prevention through surveillance functions, and number two, if a crisis comes out, then crisis resolution through emergence financing, as well as a program. I think nobody was able to predict the Asian financial crisis until the last moment. Not only that, at the beginning of the resolution process, the prescriptions were not necessarily correct. That's one of the points, the Fund has been criticized for by outsiders.

But the Fund was quick enough to learn how to fix it. At the beginning they made some mistakes because of the lack of the experience. They did have experience in Latin American countries, and they tried to apply those prescriptions to the Asian financial crisis. Some of the prescriptions went wrong, but later on the Fund was quick enough to change the directions of some of the programs. So in that sense, nobody could have predicted.

But now it's becoming better and better in terms of predicting, and now it's becoming better and better in terms of preparing the prescriptions. That's what I have in mind.

Thank you.

MR. OSTRY: Thank you, Mr. Oh.

Lisa?

MS. MARTIN: Just a quick response to the question about developing a succession index. I think that's a good idea. We haven't done that yet, but I think the general picture is very much, as this gentleman over here described it, which is that in the constituencies that have one country that has many more votes than the others, basically that country always—the chair always is the ED. In those in which there are two countries that are bigger than the others, it basically alternates back and forth. I think the most interesting group will be the more egalitarian constituencies, and to look at how the ED moves around in those, because there you do get a lot of movement.

For example, I think Equatorial Guinea holds the—is the ED right now for the African constituency, so how that moves around I think actually could be quite interesting. It's a good idea.

MR. OSTRY: Carlo?

MR. COTTARELLI: Just a couple of points. I really need to have (?) understanding the point on the political (?), but I will need to make some example. I'm too chicken to make an empiric, a real example, so I'll make a fictitious example based on my experience.

I was Mission Chief of the United Kingdom for a while.

[Laughter.]

MR. COTTARELLI: And if I come to the UK, I give you, and then because my view hurt the government, (?) picks up the phone, calls the manager (?) and asks for the replacement of the mission chief. That is what I would regard as an inappropriate political pressure. He never did this. Nor did I say anything that could hurt the government. Maybe was I self-centered I myself, being afraid that there would be such a call. It's a complex picture.

Second point, on the analogy that Tom, or lack of analogy with respect to the central bank independence, I don't think one should really push the argument of complete independence of the Fund, but you made, Tom, the point that we are in a very difficult situation, and central bank independence is something that has a very specific purpose, and only in that case it is legitimate or proper to have delegation to technocrats.

There are other cases like antitrust authorities, where you have very complex decisions that are delegated to independent agencies, things that are difficult to manage, and these are very different from central bank and do not prevent a politician from delegating to a technical agency.

MR. OSTRY: Thanks very much, Carlo.

MR. BOUGHTON: Thanks, Jonathan. Since we're running late I'll give oversimplified answers to a few of these questions.

On the last question that was raised about how the Fund has done in forecasting crises. Crises cannot be forecast. You can forecast problems, but if you can forecast a crisis, then you can prevent the crisis, and then you'll never know whether you foresaw it or not. So that's oversimplified, but I think there's a lot of reality to it. But the point is to try to give good advice to countries.

I think generally the Fund has given good pre-crisis advice to countries, and obviously with some major exceptions.

The question about how we actually could ever go about rearranging voting shares, I think it is possible. And there have been numerous cases of it. And the question is why should any country give up part of its voting power? And the answer is, it would give up part of its voting power if it can get something else in return.

One of the most prominent cases was the United Kingdom, which voluntarily gave up the number two slot after half a century in the early 1990s, and moved down to number four. I'm sure that Tom could tell us some stories about what the U.K. got in return for agreeing to do that. But negotiation is a matter of give and take is the main point, so I think it can be done.

The question before that had to do with how do we reconcile the need to maintain the incentive for creditor countries to support the Fund with the idea of balancing our shares more, and so that they would be sacrificing something, and that's an extremely part of the problem, but again, I think that people will be prepared to give something up if they can get something else in return, and that's the kind of political negotiation that has to go on.

Finally, just to respond very briefly to a point that Tom Scholar made about whether there is a conflict between political oversight and technical objectivity on the part of the staff. And in general, I think he's absolutely right, that there is no inherent conflict between those two. But in practice there can be, and Carlo gave one example in his own work in the U.K., but, you know, I—

MR. COTTARELLI: No, it was a fictitious example.

MR. BOUGHTON: Well, I'll give you a real example. I've spent a lot of time talking to people who are very severe critics of the IMF, the kind of people who make Charles Calomiris look like a defender of the faith. One of the criticisms that I heard a lot, and that others of us have heard a lot, had to do with the Korea program, where the U.S. Treasury was very actively involved on the ground. And the way this criticism was put to me on more than one occasion was that every time David Lipton would get off the plane in Seoul, there would be five more conditions in the IMF program.

Now, I think there's a legitimate question as to whether that was true as a cause and effect, and to find out whether it's true, you'll have to wait for my book to be finished that I'm writing right now.

[Laughter.]

MR. BOUGHTON: But if it's true. then I think that is the kind of conflict that people have in mind when they say we need to reduce political oversight, because I think there's a very legitimate question as to whether it's genuinely in the U.S. interest to interfere at that kind of technical level. I've asked some people in the U.S. Treasury or people who used to be in the U.S. Treasury, and they say, well, gee, the U.S. had very strong national interests at stake in trying to promote a quick and strong recovery in Korea in 1998. And I think that's true, but I think the U.S. and other countries also have a strong interest in maintaining the technical expertise and objectivity of the staff.

MR. OSTRY: Well, I think we owe a debt of gratitude to all of our panelists. We've run over by about 20 minutes, but I'd like to thank all of you for coming and participating in this forum.

[Applause.]