Transcript of an IMF Economic Forum (panel discussion at the IMF's Third Annual Research Conference): Promoting Better National Institutions: The Role of the IMF

November 8, 2002

Panel discussion at the IMF's Third Annual Research Conference
( View this panel discussion using Windows Media Player.)
Washington DC
November 8, 2002

Eduardo Aninat, Deputy Managing Director, IMF

Guillermo Ortiz, Banco de Mexico
Nancy Birdsall, Center for Global Development
Jeffrey Frankel, Harvard University
Jeffrey Sachs, Columbia University

MR. SHANG-JIN WEI: Good morning, and welcome to the third IMF Annual Research Conference and this panel discussion on "Promoting Better National Institutions: The Role of the IMF."

I am Shang-Jin Wei from the IMF Research Department. I am also chairing this conference's Organizing Committee.

We have a very distinguished panel and a very distinguished panel chair. My job is to introduce to you the chair of the panel, Mr. Eduardo Aninat, who in turn will introduce the panelists and chair the panel discussion.

Mr. Eduardo Aninat is Deputy Managing Director of the International Monetary Fund and has been in this position since December 1999. Mr. Aninat has an M.A. and a Ph.D. in Economics from Harvard University. He has held many distinguished positions in the past, including being Finance Minister of Chile and Chairman of the Board of Governors of the IMF and the World Bank, and many, many other positions which would take me a day to list.

So, without further ado, Mr. Aninat, it is your forum.

MR. ANINAT: Thank you very, very much.

We have an exciting panel, and I think time will surely run short; and we want also to have the people here, important, interesting, thoughtful people, participating and interacting with the panelists as well. So I will make sure that we are time-efficient.

The order of this panel is we will start with Guillermo Ortiz—and I will introduce each of them— followed by Nancy Birdsall, then Jeffrey Frankel and, last but not least, our friend Jeff Sachs.

Basically, we have convened here in order to discuss an important institutional question—and I underline the words "institutional" and "institution."

The role of the IMF in promoting better national institutions, as you know, has been a point of contentious debate. If I can just point to one instance of such debate, it came immediately during and thereafter the Asian crisis. Critics of the IMF concluded in several panels in the past that the IMF should limit the scope of its policy advice and conditionality to members just of monetary, fiscal and exchange rate issues and financial sector policies, but we should not be advising in any form or putting conditionality on the issues of governance, accountability, quality of institutions, standards and codes, and so forth and so on.

So we have, of course, here at the IMF a very clear position that we do care about the quality and scope of institutions and that they are central to the results of economic development, growth, and very especially stability.

But it is not for me to do the talk; it is for the panelists. So without further ado, let me start by introducing here—and each of you will have 8 to 10 minutes; those who stick to 8 will get more time at the end, and those who abuse 10 will be severely constrained by later interventions, and I promise that.

Let's start with the views of our friend, a good friend of the Fund, Guillermo Ortiz.

As you know, Guillermo Ortiz has been now Governor of the Central Bank of Mexico since 1998. During 1994-1997—and I must say that in my own capacity then, I met him in that capacity—he was Secretary of Finance and Public Trade in Mexico.

Mr. Ortiz is a member of the Group of 30, and his numerous previous positions also include professorships at various universities and as Executive Director—guess where—at the IMF.

He has published, as we know, widely on economics and finance issues, both in Mexico and abroad.

Guillermo, welcome to the IMF. You have the floor—as I said, 8 to 10 minutes in the first round.

Thank you.

MR. ORTIZ: Thank you, Eduardo.

I think this is a very important topic, and let me begin by stating what Douglas North said about institutions. He said that, basically, they are basically the rules of the game of society, and they are the set of laws and practices sanctioned by custom and organizations that give a stable structure to the relationship between individuals and groups.

Now, it is very important to distinguish what institutions work and what is a good institution. Of course, good institutions are those that, in a mixed economy, provide a framework for private agents to do their transactions in an efficient manner. They protect property rights and false contracts, they provide a stable but competitive setting for investment, et cetera, et cetera. In short, they are the backbone of a market economy.

Also, they should provide a framework for the government in its capacity as regulator, tax collector, et cetera, et cetera, to do its functions properly.

I would venture that most policymakers have a sense of what institutions work and what institutions don't work in their own countries. Then, we have to ask ourselves why is it that there is not a change either of agents, society, governments, and so on toward improving institutions. And I think there are three basic reasons for that.

One of them is the weakness of the organizations that could manage change. The knowledge base of these agents of change may be shallow, and there may be shortages of human capital, skills, and so on and so forth.

Another reason, a second reason, I think, is special interest groups. They are many times heavily entrenched in the institutions themselves, or they develop in the shadows of the contracts that embody those institutions among sectors of society, and they are actively pursuing their own interests and do not allow for change.

And the third reason is probably inertia. As you know very well, institutions develop from historical and cultural patterns. Some people would say—I hope I offend nobody—that the institutions that colonial Spain left in Latin America are to some extent responsible for some of the difficult times that we have had over the past two centuries.

Nonetheless, this is a fact of life. And then, we come to the question of how can international institutions and organizations help change domestic institutions. My conclusion is that they can be very helpful to overcome the first problem—knowledge base; they can be somewhat helpful in overcoming the second problem—the special interest question; and it is of absolutely no relevance to the third problem.

So let me try to organize these thoughts in the following manner. It is obvious that the IMF in particular, and the World Bank, and so on are like hubs of economic knowledge and experience. Every time an IMF or World Bank mission goes to a country to tackle a problem, there is experience acquired that can be processed and hopefully applied to other countries and other situations in similar experiences, then leading to a positive, let's say, outcome—hopefully, a positive outcome of change.

Also, the IMF and the World Bank have their own institutes that usually try to upgrade the skills of civil servants through courses, seminars, et cetera, et cetera. So I think that that is an important aspect.

Another one is that, for example, membership of international organizations or international [inaudible] can really help strengthen institutions, and let me give you an example in the case of Mexico.

NAFTA has been a tremendous force to create institutions in Mexico. The fact that you have tariffs, that you have a horizon for tariffs to be reduced, the fact that you have investment rules, the fact that you have a solution for controversies and mechanisms for solution of controversies—all this has helped to create institutions. For example, the new foreign investment law was enacted in 1993 when the NAFTA discussion was being negotiated, so that itself is important.

It is also a very useful example for the Fund in particular to strengthen institutions such as central banks and ministries of finance, in the sense that both central banks and finance ministries are effective only if their reputations are high. So there is a reputational component that is very important, and that can be strengthened by positive action on the part of the Fund. In many cases, Fund programs actually strengthen the hands of domestic authorities, particularly, again, central banks and ministries of finance, and they help them to be somewhat isolated from political pressures.

Another important impact, for example, is the role of international watchdogs—like Jeff here—he does this competitive report; he participates in this annual competitiveness report—this is really something which is important.

For example, in Mexico, the OECD recently published a report on education in member countries. We got pretty bad notes, so that spurred a national debate, and an Institute for the Evaluation of Education was formed with the broad participation of the union and so on.

So I think this is also another important aspect.

Another point that I would make is on the whole question of transparency. I think the Fund—and we had a seminar recently on standards and codes, and I could spend a few minutes here, but I will not—are extremely important. I just want to underline the usefulness of this whole concept of transparency that was started with the special dissemination—I don't remember the name—of information, and the work on standards and codes, financial sector assessment programs and so on. They have been extremely useful in many countries including our own.

Finally, of course, the Fund and international organizations can finance the domestic building of institutions.

Having said that, on occasion, international organizations can do more harm than good. There is a fine line between providing a useful spur to help mobilize domestic support for reform and meddling in domestic affairs on the other hand.

The difference between constructive incentives and meddling is a thin one. Like beauty, it is in the eye of the beholder. And of course, this discussion is not new. It relates to the whole question of ownership. So I think it is very important, both the substance and the form, in how the Fund and international organizations interact with domestic authorities.

But also, one of the main risks that runs in international organizations is the flip side of this expertise that the Fund has accumulated in its, let's say, connection with all the countries. It has expertise in a vast variety of countries, but it really lacks in-depth expertise in a single country. And since institutions often interlock, it is very difficult to see how pushing, for example, in one direction will affect this.

Let me give you an example—for example, the Fund's insistency on this bankruptcy court in Argentina, and the bankruptcy law in Argentina. Well, it is all very well, because bankruptcy laws look good in practice. But you have to make sure that the judicial system will actually put in place a bankruptcy framework that works. Otherwise, the law means absolutely nothing. I can give you examples in the case of Mexico from the banking system and so on and so forth.

I am getting close to my 8 minutes, so let me conclude with two positive points.

The color of this discussion, I think, is that we should not have unrealistic expectations of the ability of international institutions to revolutionize national institutions. Often, we should think about building upon what already exists; the creation of new institutions may sometimes be necessary, but the development is a slow process.

The two points I want to make—the first is related to the SDRM. I think that it is very laudable, of course, that the Fund has engaged in this collective action process and the SDRM mechanism to try to smooth the relations between different agents, creditors and debtors, in a country, but I think that the form and the timing of these efforts in the framework of pretty strong opposition on the part of issuers and the international financial community is something odd, and one has to think about it.

The final point is that it is as important to create institutions as it is to preserve them, so one of the important roles of the Fund when it assists a country is to avoid a breakdown of the institutions. The tragedy of Argentina, I think, more than anything is the total breakdown of existing institutions. In countries where the Fund has been effective—Mexico, Brazil, and so on—it has helped to preserve and strengthen institutions. In the case of Mexico, I can tell you that as part of our 1995 program and so on, we took this issue of transparency very seriously at the central bank, and this was an important element in enhancing, let's say, the working and the credibility of our institutions.

I'll stop here.

MR. ANINAT: Thank you very much, Guillermo, for a very insightful first presentation. I think you have reminded the audience about the importance of the work on the 12 broad areas of standards and goals that the IMF along with the World Bank has been developing over these recent years, and the role in stability, transparency, and as you put it, ownership also, of improving the quality and the evolution of institutions—I saw you hesitate there with the acronym, and since it happens to me, just in order to fill the gap, it is SDDS, DDDS, ROSC, and FSAP and all that—but we'll forgive you for that.

If I can also pick up one little controversial aspect of your presentation for the audience later, it is the issue of degree of involvement in these institutional areas in each particular country.

With that, let me go to Nancy Birdsall, who is well-known to all of you. She is the founding President of the Center for Global Development, a rather new, emerging, but very strong center, if I may add my own remarks. Prior to launching this center, Nancy Birdsall served for three years as Senior Associate and Director of the Economic Reform Project at the Carnegie Endowment for International Peace.

Also, as you know, and very important for us at the IMF and for Latin America and the Caribbean as a whole, from 1993 to 1998, Nancy Birdsall was Executive Vice President for the Inter-American Development Bank. Prior to that, she was Director of the World Bank's Policy Research Department.

I will say in passing that she is also co-founder with myself of a little group about social equity that we run here and there.

Nancy, the floor is yours.

MS. BIRDSALL: Thank you very much, Eduardo. I want you to tell me when I have taken seven minutes, okay, because I always go too long.

I want to talk about—and this will not surprise some of you who know me—the social contract in an open economy and IMF conditionality. And what I want you to remember is the expression "open economy social contract."

So what I will do is try to define crudely what I mean by that, and then I want to ask what does that have to do with the IMF, and what does that have to do with IMF conditionality.

So, what do I mean by an "open economy social contract"? Well, it is a critical national institution, so I was inspired by the title and purpose of this panel to decide to talk about the concept of a social contract.

And, as many of you will know in general, the idea of a social contract is that it is the outcome of a collective decision, usually through a political process, by which citizens and a society arrange to mitigate some of the cruelties of the unfettered market to provide some protection for themselves through the collective decisionmaking process that government provides and to sometimes provide for some redistribution so there is a sense of equal opportunity, or at least, less unequal opportunities, in a society.

Because it is political, it is important that it reflect the needs not just of the poor—and we often think social poor in poverty—but of the large majority of the citizens in a particular collectivity.

And of course, as Dani Rodrik pointed out some years ago very well, a social contract is absolutely critical in an open economy, and of course, it is very critical in emerging market economies and other developing countries that are subject to a lot of volatility and instability in part because they are open economies.

He pointed out that it is the most economies in Western societies that tend to have the deepest social contracts.

What does this have to do with the IMF? Is this IMF business? Doesn't it sound like World Bank business or IDB business? It is all about redistribution, and the IMF is wary of redistribution. It is awfully political-sounding, and then again, what does it have to do with conditionality.

Let me start with is it IMF business, and let me suggest three reasons which are not all of the reasons, but three reasons why I believe it is IMF business.

The first reason is that good fiscal policy is the basic ingredient of a healthy, open economy social contract. Obviously, good fiscal policy affects job creation, and in most societies, it is all about jobs for the great majority of people. It affects job creation most obviously through the interest rate insofar as bad fiscal policy tends to crowd out the private sector, drive up interest rates, and impose on monetary authorities the burden of maintaining stability.

But it also matters because good fiscal policy clearly affects the ability of governments to be countercyclical and to protect various groups in the population, not just the poor, but the very vulnerable middle-income groups, when times are bad.

I think one of the big differences between East Asia and Latin America in the last, say, five to ten years in the late 1990s, has been that difference that East Asia went into its crisis, most of the countries on average with better fiscal policy.

My second reason is the same. Good fiscal policy is the basic ingredient of a healthy social contract. But I made it a second point because I want to emphasize here the issue of the composition of the tax burden and expenditure benefits.

I think here we have a problem in that on composition issues and on taxes and expenditures, there is a certain incoherence in the mandate that the shareholders have given to the World Bank and the IMF. An example is Argentina in the late 1990s. Let me explain what I mean.

In the late 1990s, when Argentina was growing rapidly, there was a kind of sin of omission on the part of the international financial institutions. Who was there, insisting in a transparent way—or perhaps not insisting but creating information and transparency around the composition of expenditures and the formal burden of taxes in Argentina—who was there, making transparent and more visible to a larger public where a lot of the expenditures were going—and they were going sometimes to patronage, and they were going, of course, to political ends. And who was there clarifying the extent to which the richest 10 percent of households in Argentina weren't paying very much in taxes? Indeed, it is an IMF paper that suggested some years ago that the average tax burden of those households was about 8 percent in the 1990s.

The IMF wasn't there because the IMF leaves to the World Bank—I guess—the issue of when you get beyond the aggregates. But the World Bank wasn't there because the World Bank leaves to the IMF macroeconomic policy.

And the World Bank and the IDB were busy with social programs—forgetting that the social contract, writ large, has a lot to do with healthy fiscal policy.

Of course, there are a lot of other ingredients—exchange rate policy, labor market issues, social protection and safety nets—but let me go to my third reason why social contract has to be, at least in some respects, IMF business. And that is that the financial sector in open economies is IMF business, and what happens in the financial sector affects the capacity of societies, particularly in open economies and particularly in emerging markets, to manage the social contract.

And of course, the structural problem is that the financial sectors in emerging markets, because they are emerging markets, tend to be shallower and thus less resilient and less able to help manage what happens if there is either an external shock or some sort of internal policy shock or a natural disaster or whatever.

Now, I wanted to raise the financial sector in this context to reemphasize that the social contract is not just about poverty and the poor. The middle-income working class households are particularly vulnerable to shocks that arise because of this problem of a relatively shallow financial sector. And it is the middle-income households and working class households who are in the end the political bedrock of a social contract that works.

If you think of what happened in the Asian crisis, it was actually not the poorest households who suffered the largest absolute or relative losses, but what I would call the "urban strivers"—the emerging, potential, incipient middle class in urban areas. And in Latin America, if you actually look at what is happening there, you see that there are so many households so close to the poverty line that we have made a mistake, I think, in distinguishing between the poor and the rest.

What we should be distinguishing between in the case of Latin America especially is the 10 percent of households that are middle class, at the top of the distribution, where household heads have post-secondary education, and all the rest. And if you look at data, for example, on median household income in a country like Brazil, you see that it is only one-third of average household income. And it is so close to the poverty line at $2 a day that it is not surprising that when there are external shocks and when there are positive changes, as when Brazil did the real program, you see big shifts in your poverty head-count up and down.

It may surprise you to know that the infant mortality rate for the middle quintile of households in Peru is higher than the average infant mortality rate in Ghana.

MR. ANINAT: Nancy, you are about to—

MS. BIRDSALL: Seven minutes?

MR. ANINAT: No; you are at 11. But please—

MS. BIRDSALL: You were supposed to tell me at seven minutes.


MS. BIRDSALL: Okay, I am almost finished. So I won't go on with these amazing facts.

I would just emphasize that the financial sector puts middle-income households at great risk.

I wanted to say a word about the "Lula effect" because I think it is a misreading to think that Lula was elected for populist reasons. I think he was elected because middle-income—not middle-class households—but middle-income, working class households are looking for a new kind of social contract.

One minute on what does this have to do with IMF conditionality. Well, I would say in defense of conditionality that we must think of conditionality as ownership, not as a substitute for conditionality, but—to use economic terms—a complement—wait, I've got it wrong. We must think of conditionality not as a substitute for ownership—of course, because there does have to be ownership—but as a complement to ownership.

That is one point that has been made clearly. I think the issue for the IMF is to focus not only on limiting the number and the domain of conditions, but to focus on having conditionality that reflects more clearly and more visibly the need for societies to own their social contract.

What that means to me is using conditionality with ownership to encourage and support economic policies and institutions that help build that social contract in open societies. And to me, that means emphasizing not just poverty as an issue and a problem—although that should be continued to be emphasized—but broader concepts, including fairness and equity.

So let me close by hoping that I have at least begun to get you thinking in the IMF about the concept of an open economy social contract.

Thank you very much.

MR. ANINAT: Thank you very much, Nancy, for many challenging thoughts. You will have another opportunity at the end to come back to them.

Let me, because time is running, come now to Jeffrey Frankel. I understand Jeffrey wants to use the machine, so it is all yours.

Jeffrey Frankel, as you know, is Jim Harper Professor at the John F. Kennedy School of Government, Harvard University. He was a member of President Clinton's Council of Economic Advisers from 1997 to 2000. And before moving to Harvard, he was Professor of Economics at the University of California at Berkeley. He directs the Program in International Finance at the National Bureau of Economic Research.

Jeffrey, the floor is yours.

MR. FRANKEL: Thank you.

I need the lights out and the projector on.


MR. FRANKEL: A word like "institutions" cries out for definition, and I am not going to try it—it is really too general a word to do that. A narrow definition would talk just about legal bodies and agencies, narrowly defined, like a Securities and Exchange Commission or an Accounting Standards Board, or what we call "commitment devices"—currency boards and stability and growth pact, and all that.

A broad definition would include everything, everything that is not in simple theoretical models, from does there exist a six-month futures market in gold to culture.

The literature has come to talk about something in between those two extremes—a cluster of institutions, having to do with property rights and rule of law—and in a minute, I'm going to talk about some findings from the growth literature which nobody has really mentioned that are quite striking—and that is the kind of institutions they are talking about.

But before I get to that, I do want to flag that it is a very broad concept that could include a lot of things, and they are not necessarily all correlated. For example, we think that democracy is an important part of institutions, but these commitment devices, like giving a central bank independence, or a currency board, or a balanced budget amendment, they are all things that limit democracy, so they are not always correlated.



not very

a little




not relevant


capital punishment; drug policy

democratic elections; labor rights

human rights; environment

a little

social capital

IPR rules; executive compensation

Legal systems; competition policy; land tenure

poverty; education; military spending


how to close banks & dispose of NPLs; what are the best accounting rules; bankruptcy procedures

corporate governance; financial systems (e.g. capital adequacy; or relationship banking vs. securities markets)

property rights; trade policy

corruption; fiscal transparency


how to restore confidence in a crisis

exchange rate regime; capital controls; PSI

budget deficits

monetary policy

This table has two dimensions, and I mainly just want to get across the concept that both dimensions matter.

Across the top is how sure are we that we know the right answer, and across the different roles is how relevant is this particular issue to balance of payments problems and the purview of the IMF, narrowly defined and as traditionally defined.

In the bottom row, I have in the bottom right column "money policy"—we are pretty sure we know how that works. Monetary expansion—we know what the effects are on the balance of payments and the exchange rate.

The budget, I have one row over; we are not quite as sure we know everything about that. There is some disagreement about whether raising taxes when you have a budget problem can make things better, or not.

More disagreement about exchange rate regimes, private sector involvement, capital controls.

And then under "We haven't a clue," I have how to restore confidence in the midst of a crisis—that's not quite an institution.

And then, the other dimension is how relevant is this issue to balance of payments problems. The last row is clearly in the purview of the IMF. The top one is clearly not.

I have "religion" under "We are not sure what the right answer is." I don't know what the right religion is, if there is one.

We have a lot of disagreement about drug policy, capital punishment, and so on.

Go over to the right—"human rights" and "environment"—that's something we all believe in, and furthermore, I think we have a pretty good idea about the policies that would bring them about. Nevertheless, whatever the protesters say, I think it is not in the purview of the IMF.

The middle two rows are probably the ones that we are talking about in terms of institutions that have not previously been considered in the purview of the IMF or the World Bank, but may now are more.

But before I talk about the role of the IMF, I want to talk about where the growth literature is. There are some people who think we have run too many regressions on the level of growth in income and that that literature has outlived its usefulness. I don't believe that. I think we are still learning a lot from it.

There is a group of papers recently which I cite here at the bottom—Acemoglu et al., Easterly and Levin, Engerman and Sokoloff, Rodrik and co-authors—that have come up with some pretty striking conclusions about institutions lately.

Of course, the standard growth regression has found that some variables like investment, education, stability, openness seem to do a good job at explaining growth rates.

But the critique is that, well, those are endogenous. And it's not just that they are not the policies—because you could put the policies in—you could put tariffs, and you could put spending on education and all that—and we still worry that they are endogenous, that they may be the result of deeper societal factors.

So I'm going to describe this literature now the way Dani Rodrik does—that we are not looking at the three deep fundamental determinants, and that there are three competing theories or views.

One is what is sometimes called "geography," but I think it is important to say that we're talking about tropical geography; the second is openness, and the third is institutions.

They each have their measures. For tropical geography, we are talking about malaria and other diseases, how long is the growing season, some things that Jeff Sachs has talked about.

Under openness, we are talking about trade-to-GDP, tariffs, direct investment.

Under institutions, we are talking about various measures of property rights and rule of law.

In each case, there are some serious endogeneity problems. There are some countries—Panama, Singapore—that had malaria, and it has been suppressed once you get superior technology, societal organization. So that is endogenous.

Trade is certainly endogenous—some goods or luxury imports increase, or investment increases with trade. Tariffs—where countries get their revenues from—changes with the degree of development, so that is endogenous.

And on institutions, regulation and tax systems are endogenous; when countries become more developed, they do more redistribution, they do more regulation.

So in each case, the literature has come up with, I think, some pretty good instrumental variables given that it might seem like insoluble simultaneity problems.

In the case of tropical geography, that one is pretty obvious—latitude, what percentage of land is in the tropics, temperature, rainfall. Those are pretty good exogenous instruments.

In the case of trade, variables out of the gravity model—how well-situated geographically is the country for trade, remoteness, land-locked, linguistic and historical links with other neighbors.

And then, more recently, in the case of institutions, there is this interesting work, Acemoglu et all, on European settler mortality rates, the idea being that in those regions, everybody, among all the countries that were colonized, in some of them, the early settlers died off quickly and didn't settle in, and in others, they lasted longer, and so that's where they brought their institutions.

Another version, the Engerman and Sokoloff version, is related, that some countries, some regions, are well-situated for extracted industries—mining and big plantations, sugar and all that—and those are the ones that got the bad institutions. And [inaudible] worries that Spain left the bad institutions in South America, and we got the good ones in North America, in this version because of that initial difference.

Well, these papers, to generalize, come up with quite a remarkable finding. They say that institutions dominate. One of them is titled, "Institutions Rule." Some of the papers claim that institutions knock out everything else. I think that's too strong, but for present purposes, it doesn't matter. I think the idea is correct that institutions are very important, and I suppose for present purposes, it doesn't matter if they really knock everything out.

There is this interesting disagreement between Rodrik and Acemoglu, Johnson, Robinson and Thacheren. Rodrik accuses them of being "pre-destinationists"; that because they have this instrument or variable of where the early European settlers did well and where they didn't, they are claiming it is all destiny, that you can't do anything about it. And there are some interesting case examples, Ghana, for example, where the claim is that for historical and social reasons, the elite are going to take resources away from the internal inland cocoa farmers no matter what—they can do it through an overvalued exchange rate, they can do it through a cocoa marketing board. If the IMF prevents them from having an overvalued exchange rate, they will just do it in a different way. That says that it is the deeper institutions that matter, to the point where the advice is basically that you can't impose macroeconomic or maybe even structural conditions on these countries.

I think I agree with Dani that just because there has been a particular pattern in the past, and it accounts for a certain share of variation, that doesn't mean that you can't change institutions or policies in the future.

Let me turn to the role of the IMF. I couldn't read that, fortunately.

When we turn to the role of the IMF, it comes up in four ways, I think. First, the technical assistance, national statistical standards, and the IMF Institute, and all that. Second is the FSAPs and ROSCs—if I've got that right—Financial Sector Assessment Program and Reports on Observance of Standards and Codes, which I understand are increasingly linked to Article IV missions and in some cases even to country programs.

So number three is the most important and where most of the controversy is, and that is the structural conditionality in the country program. And the accusation is that there has been mission creep—back to my first slide—that the IMF used to stick to its knitting, which is the bottom slide, and that lately, they have been imperious and marched their way up this table. And certainly, financial sector institutions are now considered a major part of the Fund's business.

My view is that there has been some expansion in that direction and that it is appropriate. We have had, obviously, increasing degree of integration across countries; that creates a certain tension between sovereignty and regulation. I don't think anyone thinks that the right solution is therefore for the IMF to start dictating policy on the environment—but some movement at the margin like a roll-up seems to me to make sense, especially on the right margin where yo are more sure what the answers are. And of course, for the World Bank, it is their role to be probably another row farther up from that.

Let me say a bit about the issue of if you can force policies on countries. It used to be that an IMF staffer would say one of our roles is to take the heat politically for changes that a country, a finance minister, know they have to make anyway but cannot make. And there would be some pride in that.

Now, it is much less of that. It is much more that the local country has to take ownership of reform. Of course, it is much better to have ownership of reform, but that begs the question what do you do if they don't seem to want to own reform or if it hasn't seemed to work in the past.

The fourth place where the role of the IMF comes in on institutions, and the last one—it seems to me perhaps on this problem of prolonged users it is a desirable goal to increase differentiation. It shouldn't be that good performers and bad performers all basically keep continuing; I think there should be more differentiation. How do you cut a country off, particularly a country that has been a repeat borrower and has missed their targets over and over and over again?

I guess I am going to suggest that one should look not just at performance, but should also look at that point—and it is permissible to look—at deeper institutional causes. If a lack of democracy and a lack of stability is giving rise to repeated failures, then it may be appropriate to take that into account when deciding to cut a country off.

One more sentence. It is not the case that history is entirely destiny. There are important cases where countries have changed institutions fundamentally—Japan and Germany after World War II; the colonial empires when they broke up; the former Soviet countries; and recently, East Timor and the breakup of Yugoslavia. These are cases where institutions are being made today, and one can have an effect notwithstanding past history.

MR. ANINAT: Thank you, Jeffrey.

I think that after Frankel, and by popular demand, this session will extended by a further 10 minutes; so instead of 12:30, we will be aiming to finalize all of the work by 12:40 so we will give room for interaction by the floor as needed.

Our last panelist—last, but certainly not least—is our friend Jeffrey Sachs, who is Director of The Earth Institute and a professor at Columbia University.

He was formerly Director of the Center for International Development and Harvard Institute for International Development, and a professor at Harvard University.

In January 2002, Professor Sachs was appointed—and this is relevant for the IFIs here, because we are involved in that work, as he know—by Secretary General Kofi Annan as his special advisor on the Millennium Development goals after the Monterrey Summit.

Jeffrey, you have the floor.

MR. SACHS: Thank you very much. It is a pleasure to be here.

I think I have one message, which is to avoid trivializing this complicated issue. I happen to think all of the papers that Jeffrey mentioned do exactly that, but I am not going to dwell on those papers.

I am going to stress why it is so easy to get this issue wrong, of institutional reform. It is tougher than it looks, and we are not always on the right track.

I want to start with a general methodological observation, which is that for along time, certainly since Valraux [phonetic], economists aimed to create a statistical mechanics of economics. We tried to fashion economics in the image of classical physics, and that was a great mission of general equilibrium analysis, and it had a lot of merit.

But I think we are finding more and more that real economic analysis has to be like ecology—that is, the operation of complex and interdependent systems—and that is a very different kind of analysis, and it is one in which I view the institutions as being embedded.

The analogy that I would use is, among many bad analogies, that maybe we can think of the institutions in a society like the organelles of a cell—the ribosomes, the mitochondria, and the other functions of a cell that need to be performed for the help of that complex entity—and that it is the interconnectedness of those functions which is critical.

One of the features of a complex interconnected system like a living cell is that when any one of those organelles is misfiring, the whole cell can die. If you are very naively looking at your favorite institution of choice and see that it is not flourishing, you may easily believe that the problem lies there, even though what you are finding is a failure of interconnections where the real problems are lying someplace else.

This is a particular problem for the economics profession, because we run to try to isolate the problem in the place we always know best, and whether that is the budget or the tax policy or the exchange rate, we have endless debates about manipulating macroeconomic variables, when the deeper problems in the system—the organelles that are misfiring—may be someplace completely different in the society.

There is another point about living systems that is particularly important for a lot of the countries that I am dealing with on behalf of the Secretary-General, and that is that living complex systems are open energy systems, as Schroedinger [phonetic] told us 50 years ago. They need to be fed and cared for. They eat. That means the energy has to come from the outside. And when you don't have enough energy input, enough basal metabolism, the whole society or the cell collapses, and the same thing is true with some of the poorest societies in the world. They cannot function at the level of—they cannot survive physically right now at the level of productivity of the internal system, and there is not enough energy coming from the outside. So we are literally dealing with mass death in a lot of parts of the world that we talk about as macroeconomic crises. But Southern Africa is a region of mass death right now. It is not merely a macroeconomic crisis, and we have to understand that if we are going to get this right.

A second—just one more point—actually, two more points on this perhaps dreadful analogy. Another point about cells is that if you compare organisms, there are homologous functions, so-called. You don't always find exactly the same DNA doing the same thing. You have to satisfy a way to metabolize, a way to get out waste, a way to protect against foreign invasion and so forth. But it is not the same proteins or enzymes that are working. We cannot expect one DNA template, of course, for each of the functions, each of the institutions that we are talking about. I think this is obvious, but I am going to point out a real place where we make some serious mistakes.

The last medical or public health or biological analogy—when you have an organism that is not functioning or a cell that is misfiring, you have to do a differential diagnosis, because it is a complex system. It is not good enough to say that the heart stopped beating, and therefore, the person had a cardiological problem. Maybe the heart stopped beating because it stopped being profused with oxygen, and that was a pulmonary problem; or maybe there was a cerebral stroke or something else. And that is what a doctor does; a doctor makes a differential diagnosis and doesn't go for the immediate effect, whereas we tend to. We translate in the end everything into a balance of payments problem or a macroeconomic problem, when sometimes the roots of the crisis are much deeper.

What is the range of interesting institutions to deal with? They are not just the so-called rule of law and budget and tax policy; they are also the judiciary, the water system, the power, the sanitation system. These are institutions when you look at them in any viable society—they are not self-organizing—anywhere in the world where they function.

The education system, the health system, the science and technology institutions—instead of textbooks, look at any operating systems in the world. These are institutions that I am talking about and not self-organizing. There is a tremendous role of cooperative or forced mechanisms that are at play—environmental management, ecosystem function management and the like.

Now, why do I go on perhaps at this trivial point? Because we are sometimes really barking up the wrong tree when we are trying to understand what is going on in a country. And we go straight for the jugular, if you will, of the tax policy or the exchange rate, and we think we are doing something, when the problems are really much deeper.

Let me turn first to the poorest countries. The IMF has been in the business of the poorest countries, really, for about 20 years intensively and since the first structural adjustment fund, since the late 1980s. The record is dismal by the objective standard—not the counterfactual, but just the objective standard of what has happened in the poorest countries in the last 15 years. It is generally a continuation of decline, disease, pandemics out of control, environmental degradation in almost all parts of Sub-Saharan Africa and in many other unstable parts of the world. I would argue that it is because we don't have a realistic diagnosis of what is happening in those countries, and therefore, the instruments that we are turning just are not the right instruments.

For the countries that I am dealing with in Sub-Saharan Africa, you have a basal metabolism problem. These are literally—not just figuratively—these are literally dying societies where there is not enough food intake, there is not enough food productivity, there is too much disease, there is not enough control of basic instruments. The macroeconomics, the way that it is practiced, of saying that a budget balance and an exchange rate that is convertible, will not be sufficient to address these problems. They need an infusion of help, of real help, of a much larger amount than is on play right now.

The Managing Director makes this point all the time these days, but it is not operationalized in any way in the IMF programs and what the Board reviews and how we talk about it in the Article IVs. No place is there a realistic assessment of what would be needed institutionally to make these societies function. That is very bad, and we continue to get crises absolutely out of control.

So that is the poorest countries. There are other countries—I would say the Andean countries have this quality—where again, the basic problems are not macroeconomic. Some of the basic problems are the same problems of the last 300 years, 12,000 feet above sea level. Those are really serious problems—the highest transport costs in the world.

Drug trafficking is a phenomenally important part of the real life of Bolivia, Peru, Colombia, and increasingly, Ecuador and other countries in the region. The level of corruption that it causes, the distortions of the macroeconomic environment, the inability to run other, normal functions is profound. Militarization of that crisis by the United States solves nothing. But we don't talk about the real things—we talk about the pretend things—and we have a whole region that remains in profound crisis.

For a lot of emerging markets, when we talk about Argentina, for example, we can talk forever about the exchange rates—a fascinating question; we can talk about provincial government budgets.

I actually think that in the end, that doesn't get to the core of what Argentina's problems were myself. Argentina, it strikes me, was a society that had another set of missing institutions—the science, the technology, the higher education that could have allowed Argentina to move away from dependence on the pompous and foreign money in hydrocarbons to something that could be a sustained, competitive, industrial and service sector for the economy.

Well, that is just my own hunch, and I don't have time to justify it, but the basic problem was the underlying economy, and the budget deficits were a symptom of uncompetitiveness. Then, there is the proximate cause, which is like bleeding out, when the exchange rate regime, despite promises of 10 years, was changed.

But I think, again, there is a huge missing element even for the emerging markets, which is that a major institution which we don't talk about but which is the core institution of U.S. competitiveness is the science and technology component of our society. It is the public money that puts $110 billion into research and development in the United States, quote, "free market system." It is the higher education system which is dependent on tens of billions of dollars of non-market funding in order to run the innovation part of the economy.

Okay. What I want to say is that when we think about institutions, we are just thinking about extremely narrow aspects of what are complex phenomena.

What should the IMF do in this context? I used to think the argument was "Get out of Africa—it is not your business." But the decision was taken that, "No—we want to be in Africa." But then, I would say that's okay, too—you have to take the responsibility of understanding, then, what's going on. And if you want to be in Africa, you have to understand that macroeconomics is not the essence of the issue in Africa. And if you try to force it into that peg, you will continue to fail, because that is not what is happening in Africa.

It is not a failure of macroeconomics, and this has to be understood.

A final point—can there be a Washington consensus about the right range of institutions? Well, not the kind that we have. The phoney, textbook, simplistic, simple-minded list doesn't stand up to any scrutiny of how any real societies function. But the last thing a United States official will tell you is: "We don't have a free market economy in science and technology. We spend $110 billion. And we have an industrial policy to get the internet going, and we have an industrial policy for nano-technology, and we have an industrial policy in many different areas."

They don't tell you that. That is not on the Washington consensus list. So what we call the good institutions were not even getting the accurate picture.

Thanks a lot.

MR. ANINAT: Thanks, Jeff. As expected, you were polemic, very polemic.

I am not a panelist, but I'm sure we would have a lot to say about his setting of the record about our experience in Africa and Latin America, but let me turn to the public here.

I will only request that you identify your last name and make your question very precise to one or all or combinations of the panelists, so that we give them time at the end to close.

Is there a walking microphone somewhere? If not, simply shout.

The gentleman at the back, Mr. Boughton, please.

QUESTION: Thanks very much.

This whole panel discussion has been absolutely fascinating. I think the analysis of why institutions and social compacts and environmental and health policy are at the root of problems is extremely persuasive.

What I find less persuasive, though, is the idea that this implies a different role for the IMF. And in fact, we are at a point where the IMF is moving in exactly the opposite direction, and I think we need to have a serious discussion about what the right direction is.

In a way, you could argue that we are in a world in which we have a lot of people who are having heart attacks—to pick up Jeff's analogy—a lot of people are having heart attacks for a lot of different reasons. What we have is a brilliant cardiologist coming in and treating everybody's heart symptoms—but the question is should we be turning the cardiologist into a holistic medicine person or institution, or should we be focusing on how to get the other parts of the profession coming in to deal with all these other different things and getting a better solution that way.

MR. ANINAT: Thank you for the question to all four panelists.

The gentleman at the back.

QUESTION: Thank you. Mario Riete [phonetic] from Transparency Honduras, who is [inaudible] by Cardinal Rodriguez.

I really agree that it is necessary to move from the Washington consensus to Monterrey and Johannesburg consensus. A very important thing is how to put research into action.

My feeling is that it is necessary first to have transparency. Transparency is the most important part of everything. As Guillermo Ortiz said, this is something that we have to push in our countries.

In the case of Central America—Professor Jeffrey Sachs knows very well the Central American case—we have been signing letters of intent for many years with the IMF without results. And if we are putting more effort now—there is a mission of the IMF in my own country, Honduras, signing another letter of intent, and what is going to happen?

I agree with you that we have to change from the macroeconomic to the diagnosis of the real problem. it is a structural problem, and the structural problem—what we are going to do, for instance, on the agricultural sector with the subsidies—it is easy to produce—I mean, here in the States, we can import any type of agricultural thing—

MR. ANINAT: Mario, what is your question?

QUESTION: What I want to know is how to move to put research into action, and how can we move from the theoretical to the practical things for structural change in our countries.

MR. ANINAT: Thank you very much.

QUESTION: My name is Marty Weiss, and I'm with the Library of Congress.

I have a quick question for Nancy. I am curious as to how you think the IMF should be working with the World Bank and the multilateral development banks; how you see the cooperation between the institutions going forward to promote better institutions.

MR. ANINAT: We have two more questions.

The gentleman with the beard; yes?

QUESTION: Andrew Wilson, Center for International Private Enterprise.

Following up on this medical analogy, my question is who does the diagnosis, and how do we develop the criteria to make the diagnosis.

MR. ANINAT: In the back; yes?

QUESTION: [Inaudible; no microphone.]

MR. ANINAT: Thank you for two very interesting questions.

The lady here; yes?

QUESTION: Andrea Ati [phonetic].

I want to have a definition of what you call "national institution," because the each country has a different size, so when you try to put an analogy on a SAL in this case, the society meaning household—each country may have a lot of different institutions, because their size is totally different.

China and Singapore in this case are totally different in the multi-complexity of the institution.

I believe that there should be an international institution that compares one-to-one with the national institutions so they can work together to improve the quality of the national institutions.

Thank you.

MR. ANINAT: Thank you.

One more question, please.

QUESTION: Yes. Jeff Frankel gave a taxonomy along two dimensions—one, what we should do to save balance of payments; and the other is what we know it helps.

Nancy Birdsall seems to have gone to the wrong case. Building a social contract is something that we don't know how to do, and it is not clear how it helps the balance of payments. So I would like a clarification.

MR. ANINAT: I think we have a whole array of questions that go from methodological, conceptual, clarifications needed certainly in this complex debate, to very precise questions about positions held and regions in particular.

Let me start perhaps by mixing the order a little. This will be your last participation given the time, and certainly I will be generous and concede about three to three-and-a-half minutes to each of the four if you wish the floor.

Let me start with Nancy Birdsall for your 3.5 minutes, and this has to be arithmetically correct.

Nancy, please.

MS. BIRDSALL: I won't ask you to tell me when I have taken up 2.5 minutes.

What is the difference in the role of the IMF, thinking about a social contract? First, do no harm. And here, I didn't include, because I was running out of time, but I think the push for the opening of capital markets is an example where, in some societies, that had negative effects on the capacity of those societies to strengthen their own social contract in the short run. We know that from empirical work that, particularly in Latin America, the opening of capital markets has increased wage gaps. Okay.

Second, intervene with transparency when—or, at least, maybe not intervene and tell countries that they have to do something in the sense of imposition—but we have Article IV consultations, we have make transparent what the issues are.

I raised the case of Argentina. It wasn't just about the aggregate. It wasn't just about having a small fiscal deficit of one or two percent overall in the late nineties. Where was the IMF—telling the Argentines, telling the world, clarifying what was going on.

So, first, do no harm.

Second, be clear and transparent, and link IMF issues to effects in terms of the social contract in a transparent and clear and visible way that takes into account concepts like fairness. Let's not be so timid that economics is only about stability and growth. It also can be about justice, social justice.

And third, obviously, understand the impacts, just as now the IMF is eager to understand the impact of everything it suggesting and recommending on poverty. Think more institutionally. Think about the impact of these macro policies on the institution that is so central to open, democratic societies, namely, the social contract.

MR. ANINAT: Thank you very much.

MS. BIRDSALL: I can't say anything more?


MS. BIRDSALL: I think the issue of how to put research into action and link it to all those letters of intent and all those repeated adjustment loans in Central America and Sub-Saharan Africa has to do with the enforcement issue and the exit option which Jeffrey Frankel raised, and I think he raised it in a brilliant way.

If a country is not fulfilling the conditions, you have got to get out, or you have got to explain why you are not getting out and you are changing the diagnosis, which goes to what Jeff was saying.

And finally, if I can say one other thing on the cooperation going forward, I was on a Commission of the Overseas Development Council several years ago which thought very deeply and recommended that facilities like the PRGF, which are essentially for the poorest countries that Jeff was talking about that have basal cell metabolism problems, that that facility, with the money that finances it, be moved to the World Bank and that the World Bank consult with the IMF on the piece of a strategy in a poor, disease-burdened country—let there be consultation with the IMF on the macro part. It is completely wacky to have it the other way around, because what Jeff was saying and Guillermo and Ken and myself is that institutions are not really, in the end, the domain of the IMF, but the IMF does sometimes dangerously affect them.

MR. ANINAT: Thank you, Nancy.

Let me go now to Jeffrey Frankel, who was just alluded to; so, Mr. Frankel, you have the floor.

MR. FRANKEL: I'm going to use my three minutes to ask Jeff Sachs a question. It is a question and answer he alluded to at the end and that Jim Boughton said before, which is even if it is very important, is it the IMF's job.

Jeff, if I heard you right, you said that not long ago, your advice to the IMF was "Get out of Africa; it's not your job; you have no business there," and that now your answer is, "Well, if you're going to be there, okay, but then you should take responsibility for absolutely everything—for health and environment" and I think you had drug policy, and some things that I had at the very top of my list, you mentioned.

If you want the audience to be well-rounded people and to understand that the world is a complicated place, and history and culture, I agree with you—those are important things to understand. But are you actually saying that IMF missions should have to get involved in all these things? To my mind, the right answer is not to be at one extreme or another but to be somewhere in the middle, which means a couple of things. It does seem to me appropriate that the IMF has expanded one step up in the table to look at financial sector institutions. It seems to me that they should be in the African countries—not all of them all the time, but when there is a balance of payments problem which is in their purview, they should work on that and look around at the broader context a little more than in the past—and among other things, that means cutting countries off if it really looks hopeless and giving up on them for a while.

But that is a kind of intermediate solution. It isn't that we should either not be in Africa at all, or we should be responsible for everything, even while recognizing that everything is important.

MR. ANINAT: Thank you, Jeffrey Frankel, for making my life easier.

We will follow the rational order, then, and go to Jeffrey Sachs—and again, time is running.


MR. SACHS: The issue is—what I meant by being "in" is the current process in which there are about 30 countries that are long-timers and are going to continue to be long-timers on the way the system works. That is because these are not viable countries right now against a standard of decent health, education, and so forth, so they become dependent on a donor world and the way the system is organized; therefore, within the context of an IMF purview.

What I wasn't saying was that the IMF should therefore take responsibility for everything; it should take responsibility for understanding what it is doing.

Where this comes to the nub of the question is the framework of the macroeconomic policy, which of course is the core of the IMF's business. When the IMF says that the financing gap for country "X" is so-and-so, that has got to be against a standard. The standard that it has been against up until now, roughly speaking, has been: Here is what the donors are going to do, and you live within your means. And since we are against inflation, open credit creation, and instability, live within your means.

What I am saying is that the international system can't stop there. We have to put the perspective a different way, which is what does the country need in order to be able to achieve some internationally-accepted goals, for example, of controlling disease, staying alive, reducing hunger, and so forth—and if that's not available financially within a macroeconomic framework as currently funded, then it is the IMF's responsibility, actually, to go back to the donors and say that's the financing gap. The financing gap is not the gap to fill some notional balance of payments target that leaves million of people dying this year. The financing gap is the gap that is needed to close the gap with the development goals that the international system has achieved. That requires a deeper knowledge of what the underlying problem is.

If you think the underlying problem is that your property rights institutions are wrong, then it is free—there is no financing gap. If you think, like I do, that malaria and AIDS and drought and soil nutrient depletion and transport costs that are enormous and other real things are at the nub of the crisis, then there is a financing gap from a development goal point of view.

And what I would like to see the IMF do is to understand those factors enough to say not that you have a balance of payments gap, but you have a development gap, and the donor world which has signed on in Monterrey and in Johannesburg and in the Millennium Summit and in a dozen other places, to be partners in addressing that gap—that is what is needed to get to the core of the issue. That is very operational.

And a final point—sorry; it is important, I think—when I say "operational," I'm not saying the IMF should become experts in malaria. I am saying that you need to look at those goals and then ask the WHO for advice; ask UNICEF for advice; ask UNAIDS for advice, and build up a financing gap which is realistic to the underlying realities of the countries—not to make a Malaria Department at the IMF, but to understand that that is part of your responsibility, and we have a big, complex international system that you can draw upon to get the financial assessment to put into your macroeconomic framework so that we are finally addressing real variables that can make a real difference in achieving real results.

MR. ANINAT: Thank you, Jeff Sachs.

Our last closing panelist will be the policy banker, Mr. Guillermo Ortiz.


MR. ORTIZ: This has been, I think, a very difficult discussion on this last part, because it goes to the heart of how far can the IMF go. And I think that Jeff just made a very effective and emotional appeal to the understanding of the underlying causes of the problems in that subset of African countries.

Just to point out the complexity, let me turn to another of Jeff's examples—the Andean countries. You say that the problem there is drug trafficking.

So what is the IMF to do? We can start a big discussion here, but my own view, at least, is that as long as drug consumption in the United States is what it is, and the double standard that we all know applies, then the problem will not be solved, and the structure of these societies will be dissolving gradually and slowly.

Then, given that framework, what is the IMF going to do? Obviously, these countries all have balance of payments problems and so on and so forth.

I think that the IMF can continue to play a very useful role and traditional role. And by analogy to what you just said, it has to take a stand on these kinds of issues. This is a new role for the IMF and a very difficult one.

So I will just conclude by pointing out some of the complexities of this thing, but my own inclination is that the IMF should actually narrow its scope on this cardiologist approach and be very careful about trying to quote-unquote "change" institutions, but do things that are already proven right—for example, transparency, information, these kinds of things that do work.

MS. BIRDSALL: And the cardiologist could be listening to the other specialists, too.

MR. ANINAT: We are listening, Nancy, we are listening—very much.


MR. ANINAT: Let me just in closing refer here in the last half-minute to the issue of, as we know it, the subject of institutions is going to be complex, fascinating, and challenging for all of us—for the four panelists, for the academic world, for the policymakers' work, and for us, the international institutions.

I think this has really been a very polemic, thoughtful, but also challenging panel, and in closing, I would like the floor to join me in giving them recognition by your applause.

Thank you very much.


MR. ANINAT: The meeting will resume at 2 o'clock in this room.

[Whereupon, at 12:35 p.m., the proceedings were recessed, to reconvene at 2 o'clock p.m.]


Public Affairs    Media Relations
E-mail: E-mail:
Fax: 202-623-6278 Phone: 202-623-7100