State-Building: Governance and World Order in the 21st Century

• Introduction
• Remarks
• Questions and Answers

Introduction

JOANNE MYERS: Today we're extremely pleased to welcome back Francis Fukuyama, this time on the publication of his new book, State-Building: Governance and World Order in the 21st Century, which will be available for you to purchase at the end of the program today.

When George Bush came to power almost four years ago, there was one aspect of America's foreign policy he was determined not to pursue, and it was what he called nation building. But that was before September 11th, before the launch of the war on terrorism, and before rebuilding governments became a reality for the U.S. government.

Today America finds itself in the business of actively preventing failed states from spawning terrorism, while also attending to the problems that weak states or states lacking firm governmental control present. Whether we are discussing the spread of AIDS, the threat of terrorism, global poverty, or the refugee crisis, the dangers to us and to the world order often arise in these weak, collapsed, or failed states.

Learning how to fix these failing nations and providing them with what Professor Fukuyama describes as "an organizational tradition needed to create sustaining state institutions so that they can survive the withdrawal of outside intervention?will be a defining issue for America in the century ahead.

When thinking about ways in which we can address the challenges these troubled states present, I am confident that we would all benefit by listening to Frank's sagacious suggestions regarding this issue.

The announcement of a new book by Professor Fukuyama is always a very special event. This has been true since the publication of his groundbreaking work The End of History in 1992 and it is so once again with the release of his latest work, State-Building: Governance and World Order in the 21st Century, in which he explains the importance of state building with the keen analysis which he is known for.

Professor Fukuyama's reputation is one for always staying ahead of the curve, for always being able to describe the present with a lucidity and boldness that is shaped by his rich knowledge of the past. Yet, one could argue that it is his extraordinarily prescient view of the future for which he is mostly admired.

He is the author of the hugely influential and international best-sellers The End of History, Trust, The Great Disruption, and Our Posthuman Future.

Professor Fukuyama is currently Bernard Schwartz Professor of International Political Economy at the Paul Nitze School of Advanced International Studies at Johns Hopkins University.

At this time please join me in welcoming our guest, who came from Washington today to be with us. Thank you for joining us.

Remarks

FRANCIS FUKUYAMA: Thanks very much, Joanne. The Carnegie Council has allowed me to talk about every single one of my books, so I appreciate your hospitality over the years.

I will not talk much about Iraq today. In Washington, any conversation that begins on any topic ends up on Iraq within about 30 seconds, and I suspect that we will get to Iraq in the question-and-answer period.

But my book was written, and the lectures on which it was based were given, before the Iraq war started, and I began looking at the question of state-building and the transfer of institutions to developing countries more in a broad development context, where the focus was not so much real basket-case failed states as on weak states that needed stronger institutions in order to develop successfully. I will start with a very broad way of thinking about the state in economic development and then narrow the focus progressively to particular issues.

I don't have any answers and I won't tell you how to successfully build states, because one of the conclusions that I've drawn is that we know depressingly little about this extremely difficult problem. We need to spend much more time intellectually and on the ground working on this area of public policy where there is little institutional learning and relatively little institutional knowledge. I would like to set the framework for thinking about this.

What is a state and what does it do? The size of the state has been probably the central political issue in the 20th century. At the beginning of the 20th century, most developed countries had a state that collected under 10 percent of GDP in revenue. There was no income tax in either Britain or the U.S.; it came mostly through Customs revenue. The state was a minimal night watchman operation.

Most of the political controversies of the century have had to do with the size of the state. The modern European welfare state in the course of two world wars, depression, economic cycles, grew to the point that the average continental European welfare state takes in approximately 50 percent of GDP through the state sector. In social democratic Sweden it's up to 70 percent. Even in the U.S. we take in about 35-to-40 percent of national wealth and funnel it through the state. States have grown enormously.

The ultimate growth of the state was communism, where the state sought unsuccessfully to take over not just the economy but civil society, the family, and every other institution.

We have been in a period of the retreat of the state ever since the Reagan-Thatcher revolution some 25 to 30 years ago, accelerated by the collapse of the Soviet Union, in which the major thrust of world politics has been to reduce states, to make them smaller, to privatize, to deregulate.

If you transpose the problem to the developing world and to poorer countries, the central problem is in most cases not an excessively large state but an excessively weak one.

What does a state do? States have many functions, from those that are necessary to those that are counterproductive and bad.

The World Bank's World Development Report in 1997, the title of which was "The State in a Changing World," provides us a list of functions of a state. It was an interesting document because the Bank had suddenly woken up to the idea that the state and politics were critical for development.

Minimal functions include pure public goods, like defense, law and order, property rights. Any hard-core libertarian economist will agree that the state must provide these basic functions. Various intermediate functions, and then activist ones are listed, such as promoting industrial policy, allocating credit, and over-regulating.

If you ask what is the essence of a state, Max Weber's classic definition still applies. He said that the state is a "legitimate monopoly of violence over a defined territorial area," which means ultimately that the state's essence is its coercive capacity, its ability to send a guy with a uniform and a gun to make you obey the state's laws. We hope that most states do this not by brandishing that coercive power overtly; they do it in more subtle ways, but if a state cannot enforce the law, it is not a state.

I have taken the World Bank list and rotated it on its side to make an X axis. And so one way of thinking about the state is in terms of these different functions, which I would call the scope of state activities.

But there is a completely separate dimension to stateness which has to do with this Weberian ability to enforce the state's rules and laws, which I would label the strength of institutions, as opposed to the scope of state functions. And so any particular function that the state does can be done less well or better.

If you then break down the different things that states do, their actual capacity varies across these different functions. Unfortunately, all too many states in the developing world are good and effective at jailing journalists or intimidating political opponents; on the other hand, they can't issue visas and they can't register businesses and provide for security walking down the street in their capital city. There's a great inconsistency in where a state is with respect to any of the functions on the X axis along the bottom and how well they can exercise any one of those particular functions.

How well you perform the most ambitious function locates you as a state in terms of where you are within an overall matrix of stateness. I array the scope of state functions against the strength of state institutions. As you move outward on the X axis, you get into more and more ambitious state functions, away from the minimal state. As you move up the Y axis, you get to more and more effectiveness in doing any particular function.

From the standpoint of economic efficiency or economic growth, which is the best quadrant for a state to be in? If you had to choose between having an extensive or a limited scope and being relatively strong in your enforcement ability, which is the optimal quadrant of the four?

Quadrant 1 means that the state is doing the basic things that states should do—it's providing for property rights, rule of law, domestic security, national defense, and doing them well.

The problem in many developing countries, unfortunately, is that they are not in quadrant 1, they are in quadrants 3 and particularly 4. In many respects, quadrant 4 is the worst place to be because it means that you are doing all sorts of things that you shouldn't be doing—running agricultural marketing boards, steel mills or cement factories, and you're doing it incompetently, you can't manage macroeconomic policy, you can't run peristatals efficiently.

Quadrant 3 is also quite problematic—you're not trying to do very much and you're not doing that well. That describes the situation of Liberia, Sierra Leone.

I have tried to put a variety of different countries in different points in the matrix. Let's begin with the United States. We have a relatively limited state compared to other industrialized democracies. We don't tax as much, we don't regulate, we don't have national health insurance like many countries in Europe do.

Japan and France go further to the right in my matrix because they run industrial policies, they regulate more than we do, they have more in terms of income transfers and protections for poor people. And they are probably better at public administration than we are. These are two countries that have very powerful elite bureaucracies that set the gold standard for public administration.

The former Soviet Union is way out at the right side. We thought it had an extremely strong state in terms of the ability to enforce, but we now realize in retrospect that they had a huge black market and lots of corruption. Their state didn't perform as effectively as we believed a totalitarian state did.

Turkey and Brazil belong in quadrant 4. The Turks do too much. The Brazilians used to have state industries, many of which they have now privatized, and so they have been moving back down the X axis, but unfortunately they do not have strong public administration in many areas.And then there is poor Sierra Leone, close to the origin of the graph.

There are different reform paths that states have taken, because a state moves over time. Let's take the U.S.S.R. In 1980 it had a very extensive state; it was a communist regime, with no private property, and with central planning. After 1989, it became a market-oriented, feckless democracy. Unfortunately, at the same time, it moved downwards in terms of the effectiveness of its institutions. The Russian state has had a chronic problem in the 1990s: it can't collect income tax, as it doesn't have the administrative ability to even collect taxes; it can't control mafias; it can't provide law and order on the streets of Moscow. And so it has moved in a southwesterly direction.

Even a state like Japan has moved along that general vector. They have liberalized and privatized to some degree, but the fabled Ministry of Finance has made a huge number of mistakes over the last ten to 15 years, and so they have lost a certain degree of institutional capacity.

The famous Washington Consensus that dominated the policy advice that we were giving developing countries in the late 1980s and early 1990s was all about moving to the left on the X axis, sloughing off state functions, deregulating, privatizing, reducing the scope of the state.

Unfortunately, the way that actual states implemented this was by simultaneously cutting institutional capacity, so that not only did they stop doing things that they shouldn't have been doing, but they stopped performing necessary functions or ended up doing them less well.

Another vector is to move in a northwest direction. For example, New Zealand in the early 1980s began a series of Thatcherite economic reforms. They were facing a currency crisis, they had much too much protected agricultural and other markets, and so they deregulated, and did all of the things that the economists advise you to do.

In the late 1980s, they also paid attention to the question of public administration and public management, and began a series of reforms that sought to improve their ability to perform better the remaining functions that were left after they had gone through their equivalent of the Thatcher revolution. Theirs was called Rogernomics, after Roger Douglas, their Labour Finance Minister. As a result, they improved their public management. They put senior public officials on management contracts, they had strict performance criteria, they had clear benchmarking of public services. It's one of the few countries that has moved along the optimal reform vector.

A change in thinking has occurred in the development policy community in Washington, D.C., where the IMF, the World Bank, the Inter-American Development Bank, and other such institutions are headquartered.

The New Zealand path is likely to improve your economic performance. The Russian path, in response to policy advice from Washington, is less than successful. If you could have a pure choice between path 3, where you simply reduce the scope of the state, and path 4, where you kept the same large state but improved your administrative ability, which one of those would you choose?

The consensus among people involved in economic development has changed over the past ten years. If you wind the clock back to 1992-1993, people would have said path 3 is preferable to path 4. Today, people would probably say path 4 is the optimal path.

I have this on the authority of no less a figure than Milton Friedman, who gave a fascinating interview a couple of years ago where he said, "Ten years ago my advice to any transitional country coming out of socialism would have been privatize, privatize, privatize, because you have to get stuff out of the state sector as quickly as possible. Today I wouldn't give that advice any more. I would say a rule of law is more important before privatization."

This is a very important change in priorities and is one of the reasons that much of the development community, particularly since the publication of that 1997 World Development Report, has been talking about the importance of institutions and politics as the source of economic growth. States that are incapable of performing the reduced set of functions of state will not grow.

A slew of economists have been doing empirical studies to see whether this was true. If you look at all of the factors that go into successful economic growth—resource endowments, geographical location, policy factors, whether you have a big state or a small state in terms of scope, and institutions—it turns out that institutions by far trump all of those other factors as correlating with positive growth outcomes.

If you want a single reason why East Asia did so remarkably well in the 40 to 50 years following World War II compared to Latin America, and compared especially to sub-Saharan Africa, it is that that part of the world had a tradition of strong, competent state bureaucracy that other parts of the world did not have.

Sub-Saharan Africa has a number of states that have had decreasing per capita GDP over the last 20 to 30 years, and in some cases you have less competent state administration in those countries than you did at the time of independence from colonialism. This is not for lack of the outside world trying to help them—approximately 10 percent of the GDP of the entire region outside of South Africa is contributed by the international community—but because of the lack of strong, non-corrupt, transparent government institutions.

What do we mean by strong states or institutional capacity? This has become something of a cliche now in Washington within the development policy community. Everybody says very matter-of-factly that you need administrative institutional capacity. But the moment something becomes a cliche your ears should perk up and you should start wondering whether there isn't something wrong with that. If you disentangle the different components of what people mean when they say "institutions matter," the most straightforward answer has to do with organizational design and management. This is what you call management science if it's done in the private sector in a business school, or public administration if you do it at the Kennedy School. You have certain real knowledge that you can transfer to other societies about how to organize a tax agency, or how to create a central bank.

We know less about this than we think we know, and even in this case where the knowledge that we have about state-building is fairly technical, there are more degrees of freedom in how to go about building a modern state institution than many experts in public administration would have us believe.

The second component is something that political scientists or lawyers engage in all the time: institutional design, constitution writing. The former Soviet Union, all the states in Eastern Europe, and many of the countries in Latin America have rewritten their constitutions and restructured their institutions.

More than a decade ago Juan Linz suggested that one of the big problems in Latin America is that they have all modeled themselves on the U.S., with presidential systems combined with proportional representation legislatures. This is a deadly combination that produces political gridlock and lack of representation, when they would be much better off with European-style parliamentary systems.

The problem is that although we know something about structuring institutions, one gets to restructure institutions on very rare occasions. If you have a presidential system, you can't go out tomorrow and replace it with a parliamentary system because a political scientist tells you to. It's done in times of emergency—after a war, revolution, economic crisis—but it's not something that is very easy to effect.

The third area, where we've made some progress over the years, has to do with the basis of legitimation of a regime. If you wind the clock back 30 years ago, Samuel Huntington wrote the very famous book, called Political Order in Changing Societies, in which he was arguing a similar line, that politics matters, that a strong state is necessary before you have democracy.

We would say now that he was only partly right, that democracy is quite important, because if people do not believe in the legitimacy of the state, it will not be a strong state. In the former Soviet Union, we saw the collapse of the entire political system because people in the end did not believe in its legitimacy. This is something of great concern in Iraq as we work on reconstruction.

If you zero in on just the question of public administration, what can we teach poorer countries or developing countries about how to govern themselves better?

First of all, we have to presume that they want to do it themselves. If there's not a domestic demand for institutional reform, no outsider will ever be able to help.

Where can you help them the most? Transaction volume is important—that is to say, how many decisions does a state agency make? Alan Greenspan sitting in the Fed makes one important decision every six weeks, whether to raise or lower interest rates, whereas a public school system makes hundreds of thousands of individual decisions on how to teach and what curriculum to pick and spread out over every city and province and town in the entire country, and so they have a very high transaction volume.

Specificity has to do with how well you can monitor the impact of what a public agency does. Some are very easy to monitor and others are not.

Central banking is a classic case of low transaction volume and high specificity. With an incompetent central banker, like some of the great central bankers that Russia produced in the early 1990s, you get hyper-inflation and disastrous macroeconomic policy.

Low transaction volume is the simplest problem to fix. Ten Ph.D. economists can be air-dropped into almost any developing country and set up a competent central banking system. If you look at Latin America, that is what has happened in the last generation. Bolivia, Argentina, and Mexico now all have MIT-trained central bankers that know what hyper-inflation is and how to manage monetary aggregates. That aspect of public administration has been fixed.

But if you go to something like public health or education, where you have miserable results and very poor performance, in many developing countries, it's different. There is not a public education system in the world that ten Ph.D. experts can fix. We can't fix our public education system in the U.S., and we're the most data-rich, richly endowed-with-resources country in the whole world.

You can't fix many of those problems because you can't measure them. To this day we still can't measure the outputs of public school systems. Therefore, all the accountability and transferability that we look for in public institutions are not possible.

In short, if you look at the different functions of government, some of them are much more readily fixable, through restructuring the incentives and institutions to provide better transparency. "Transparency" is a nice word that people use now for monitoring. You want make sure that you see when your public officials are screwing up. "Accountability" is a nice word for punishment—when they screw up, you fire them or your dock their salaries or do something so that they don't screw up again in the future.

We have calls for transparency and accountability, but you can do that more easily in the case of aircraft maintenance, or running a state telecom company, or central banking, than you can when you get into something like running a big public health system.

Even in the more technical fields, there is no such thing as best practice. One of the big problems with Westerners in developing countries is that we think we know how to fix these problems and we are subject to the best practice mentality.

This is endemic at the World Bank. They get an agricultural extension program that seems to work in Peru and immediately say, "Aha, this is a best practice, let's transfer it to Botswana and to Kazakhstan and other places." Unfortunately, it simply does not work, because there are many different ways to handle the same administrative public management problem, many of which will involve taking account of local conditions and practices.

One good example is the HIV/AIDS crisis in Africa. NGOs and agencies that are trying to administer retrovirals in places like Uganda and South Africa are faced with the problem of traditional healers. All of these countries have traditional healers—it may be voodoo or some other form of traditional medical practice. It's not scientific. Do you try to use these people in administering a modern health care system to distribute retrovirals? You may say, "Well, of course not. These people are not trained doctors or health practitioners." But on the other hand, if you don't get them on-board, they will go around and tell the villagers that these outsiders are distributing poisons to our children and that villagers shouldn't take them, and they can completely undermine the effectiveness of the health care system. And so do you incorporate them, do you try to bring them into your modern health care system; or do you try to bypass them?

These are decisions of enormous complexity, and there's no simple answer. Certainly not one that can be devised in a city like Washington that will have any applicability on the ground in most developing country situations.

We need to very much get away from this best practice mentality. When we train people to do development, it requires a very complex, contextual judgment about what is a universally applicable model. There are some things that are like central banking, where your ten Ph.D. economists can do a lot of good, but most other public administration problems are not the same.

In the past decade, foundations, USAID, the World Bank, and a lot of aid organizations have taken a theory about the importance of civil society to heart and promoted civil society around the world. A lot of money has gone into this.

I believe that civil society is an important function. It's important for the healthy functioning of society. But what that has done in a way is dispersed all of our efforts, because there isn't anything that isn't civil society.

In many cases, when we tried to promote civil society in Russia, what we did was create a very thin layer of people that were good at writing foundation applications to Ford, Rockefeller or the World Bank to get money to promote their women's organization or their labor or human rights monitoring organization that disappeared the moment the outside source of funding disappeared.

We need to refocus policy and concentrate on things that we are able to do, and much of that has to do with building the administrative capacity of states.

Applications: What does it mean that there's no such thing as a best practice when you talk about something like state-building?

Take a couple of important decisions that have been made about reconstruction efforts in the past. Let's begin with Japan. General MacArthur was very intent on keeping the Japanese emperor. To this day there is a lot of historical controversy over whether this was an appropriate decision. In Germany we got rid of the old regime completely. In many respects, Germany is healthier than Japan is today because we did that, because there is not a right wing that is still loyal to the old regime. The new democratic regime can blame all of the sins of the old regime on something that is now dead and gone. The Japanese can't do that. On the other hand, many people would say that the reason we didn't have a guerrilla war in Japan during the American occupation was because MacArthur made the right call in that particular circumstance.

What kind of knowledge would be necessary to enable you to make the right call in that kind of decision? There is absolutely nothing theoretical that you could possibly learn in any political science department, in any public administration program, or anywhere that would help you with that. What you need is all of the contextual historical/cultural knowledge about Japanese society and what was happening on the ground in Japan.

Fast forward to the future, the decision by Walt Slocum and Jerry Bremer to disband the Iraqi army. Many people are casually saying now, "Obviously that was a bad decision." But on the other hand, one of the reasons they did it was that the old Iraqi army had a Shiia enlisted corps, and an officer corps that was almost entirely Sunni-dominated. One of the important political reasons to disband that army entirely was to convince the Shiia majority, on which the success of any democratization of Iraq would have to rest, that that old Baathist army was gone forever, and that they were making a new start.

And again, this is a contextual judgment that is extraordinarily complex. In the fullness of time we will probably know whether it was the right or the wrong decision.

We have to confront this problem quite honestly, that in many cases our attempts to improve the institutional capacity of developing countries has led to the opposite result. Many countries in Africa are in this situation.

It's not because we have bad intentions; it's because we have contradictory motives. We say that we want to strengthen institutions in developing countries, make them stronger, but we also want to provide the services that those institutions are supposed to provide. When push comes to shove, we're more interested in providing the services than we are in the institutional strengthening.

You have a public health problem where you want to deal with the AIDS epidemic, and you can do it through strengthening the indigenous public health infrastructure in the local developing country that is full of corruption, nepotism, patron-client relations, missing administrative capacity, so that it will be ten years before the first retroviral is out in the field. So what do you do? You come in with a richly endowed public health program, you bypass the local institutions, you undermine their ability because you are paying salaries.

Michael Ignatieff said that when he was out in Afghanistan last year as a New York Times reporter, the salary of his chauffeur was ten times the average salary of a minister in the new Afghan government. And so who the hell wants to work for the government when you can go to work for an NGO or for a development agency?

Bosnia is another clear example. The Office of the High Representative and the whole infrastructure of the international community runs that country in a very dictatorial fashion. They overturn the results of elections, they routinely replace officials for reasons of corruption. Were the international community to withdraw tomorrow, you would not have anything by way of self-sustaining domestic institutions that would be capable of running the country without an outside crutch.

I hate to leave things on this somewhat depressing note, but, quite honestly, this is where we stand. We know something about what is now called post-conflict reconstruction. The Bush Administration unfortunately didn't. The knowledge does exist in the international community about how to organize in the wake of a severe civil war or interstate conflict—how to organize humanitarian relief, get electricity and water connected, get the banking system running again.

The part that we have the hardest time with is the second phase. Once you have stabilized things, how do you make the transition to local institutions that are strong enough to withstand the removal of that external support? Even as a conceptual matter, we are so much in the dark as to how to do this that we are in quite a lot of trouble, because after all, that's your exit strategy. The chaos in the failed states gets you in—that's the easy part. The hard part is the getting-out. But unless you resolve the issue of self-sustaining institutions, you're not going to have an exit strategy.

JOANNE MYERS: I'd like to open the floor to questions.

Questions and Answers

QUESTION: We do a lot of damage oftentimes as consultants when we go abroad because we do things out of context. I'm about to embark on a project in the Balkans, the idea of which is that we would go in and begin to get the country to do what they can industrially-farming, timber, fishing.

How many cases have you seen in your own work abroad where consultants have been able to come in from developed nations and jump-start new industries?

FRANCIS FUKUYAMA: No good examples of this come to mind and plenty of bad examples exist. In the 1950s and 1960s, under the influence of the Harrod-Domar growth model, we thought that every country needed a steel mill, and so we would try to jump-start an industrial structure that was based on the existing industrial structure of Belgium in what was then called Zaire or in Ghana.

There's more flexibility in the way that we approach these things, and so we now no longer think it's terrible if poor countries concentrate on agriculture and try to export agricultural commodities instead of trying to jump-start industrialization. The success depends on just the contextual awareness of what the real capacities of a country are.

Even the attempt to promote agriculture in many places has resulted in problems. A country doesn't have roads, such as in Uganda and you try to grow coffee. You should have a comparative advantage, except that it costs ten times as much to truck your beans to a port where you can sell it to somebody than it does to produce the coffee.

QUESTION: I'd like to take a look at your paths to reform and then apply them to a couple of countries, particularly China, India, and the countries of Eastern Europe.

What is the role of ideology here in the path of China, where rule of law is in place but usually through ideology control, as opposed to the real rule of law that we understand in Western democracies? China has grown dramatically with infusion of Western money, and harnessing of its energies, but there is a problem.

India, on the other hand, is about to undergo a big change. What is the problem here in terms of the acceptance of the new government when India was undergoing a lot of reform in the post-Soviet model system?

How do we apply what is going on in Eastern Europe with paths 3 or 1?

FRANCIS FUKUKYAMA: China and India don't fit my model terribly well because they both had the advantage of having relatively strong state administrative structures at the time that they began their reform program. Their problem was more like Britain or New Zealand, having a state that was relatively strong but way too much scope. The Chinese began in 1978 to then move towards the origin of the X axis by privatizing agriculture and then allowing the market sector.

India is probably the only real case of true state-building or true nation-building brought about by the British. When the British first came to India, it wasn't a single political entity; there was no common language, and a lot of princely states, the moguls up in the north. The British created the idea of a modern India, a unified state, and gave them a central administration.

Their problem was never this failed/weak state problem, but that they were just regulated too much, and so beginning in the early 1990s they have joined the same path as the Chinese. Fortunately, when they moved back down the X axis, they are not simultaneously cutting the need of their administrative capacity. So both of those countries have been rewarded by healthy economies.

In Eastern Europe, they were fortunate in being able to move, having had relatively strong, competent governments, except in the Balkans. They were fortunate in being able to walk back down the X axis very quickly, and so they got much more flexible labor markets, lower wage rates, and they're eating the lunch of the Germans because Germans didn't introduce that similar kind of flexibility.

They will become re-ossified to some extent now that they're back in the EU, but the question is whether the EU will change them before they change the EU.

QUESTION: You suggested that it's probably better to leave weak states in their basic agriculture and fishing, but how can they exist when the major nations, like America and France, have such high tariffs against the type of products that they'd be producing?

FRANCIS FUKUYAMA: That's a separate policy issue. We spend approximately $3 billion a year to subsidize cotton, which then imposes a cost on India of about $25 billion in lost cotton revenues, or we give Mali something on the order of $35 million in foreign assistance and we take away from them another $40?0 million of lost cotton revenues as a result of these subsidies. The tariffs are a bad policy decision on our part. The American consumers, and certainly foreign producers are the big losers in all of this. My great hope for the future is that now that you've admitted a big agricultural producer like Poland into the EU, that the two-tier system of subsidies will survive in the end, and provide the basis ultimately for the political undermining of agricultural subsidies.

QUESTION: Would you discuss the relationship between what you have defined as the state and the people? The Preamble of the UN starts out "we the peoples," but then when you read the fine print it is the governments. Now, as governments change, some social political entity continues as a member. If the Republicans get voted out, the U.S. is still a member of the UN. In terms of what you have defined as state functions, how do these institutionalized functions functionally, operationally, relate to so-called civil society?

For example, UNICEF and the World Health Organization have worked together very closely over the years. We tried to reorient public health policy to incorporate the people as active participants. How does that work, not just for health but generally, as an illustration of a general issue?

FRANCIS FUKUYAMA: There is an impeccable theory behind the role of civil society in a modern democracy, which is that the only way that you can hold governments accountable is to have an organized, active civil society that acts as a watchdog. It's very hard to imagine public education in this country working even as well as it does without PTAs and the active involvement of parents who can complain like hell when their local school systems are failing.

Civil society is part of good governance, in that it provides the transparency. Those are the people who do the monitoring and see when the government is screwing up.

When I said that we have probably put too much emphasis on civil society, I do not want to be interpreted as saying that civil society isn't important or that civil society doesn't play this very strategic function. I am on the Board of the National Endowment for Democracy, and so we face these kinds of decisions all the time as to whom to support in promoting institutions around the world. We spend a lot of money supporting civil society organizations.

However, the 1990s emphasis on civil society opened the door to a broad scope. Look at the importance, for example, of political parties. The single thing that Iraq lacks now is political parties. The Shiia have some parties that grew up in Iran and other places outside the country. The Kurds have very well institutionalized parties. But one of the reasons that the Sunnis are waging a guerrilla war is that they don't see any peaceful way of participating in a future democratic Iraq. There is a desperate need to create political vehicles.

That's why, as an NED Board member, I have been asking them to pump as much of their money as possible into political party-building, as opposed to setting up yet another human rights organization. All of these things are good and important, but there has to be a certain sense of priorities when resources are scarce.

QUESTION: I appreciated the basis of your presentation, which is that we are facing a crisis in this area and that over the years it has become increasingly evident that we don't know how to undertake capacity-building, institution-building, and state-building.

One of the things that we are looking at in the International Peace Academy [IPA] state building program is the search for a more realistic solution. The ideas that come up are to what extent do we need states, and what is the element of stateness that is required? We're thinking about the post-conflict/failed state model.

Have you thought about, say, Somalia as an area of land without a state but where activities are taking place and services being provided on the scale of where we need to define the element of stateness which is essential?

FRANCIS FUKUYAMA: There are two ways to go about doing it. If you look at the World Bank list of state functions, the further you are towards the left side of the chart, the more critical those functions will be.

But the other way of looking at it is in terms of what we can realistically provide, and that goes to the chart on the transferability and the transaction volume versus the specificity, because it may be that there are certain critical functions, like public education, where we just don't have much to offer in terms of institutional reform, whereas there may be others where we do have a technical solution to certain problems.

The approach must be to figure out what is at the intersection of those two that are inherently necessary, but that are also things that we know something about doing and can allocate resources to first. A research agenda is to figure out, of all the public administrative functions, which ones can we realistically transfer knowledge about.

QUESTION: If you look at your reform paths, I would caution against thinking that path 1 or path 4 is necessarily superior in general, the reason being that if you look at the scope of state functions as you have identified them, often the learning curve that particular states go through, much like organizations or enterprises, can be nonlinear. Namely, the capacity to deal with industrial policy or to provide social insurance or regulate monopoly may exist, and working on that significantly may provide you with competencies to help with equity and other more minimal functions.

In South Africa, there are a number of the intermediate and activist functions performed rather effectively, but the capacity to enforce the rule of law with respect to crime or public health demands is not very well met.

Therefore, the question is: (a) are you maintaining that it's path 4 rather than the other path that's superior; (b) are you insisting that quadrant 1 is the superior quadrant; and (c) can this be demonstrated on a linear axis in this way, given that the functions are nonlinear in reality?

FRANCIS FUKUYAMA: It can be empirically demonstrated that it is much better to be in quadrants 1 or 2 than in 3 or 4, according to the econometric data on institutions.

What it doesn't tell you is whether it's better to be in 1 or 2, because the Europeans do surprisingly well given how extensive the scope of their state is. Sweden should be a basket case if you think that the scope of the state is as critical as a lot of economists would have you believe.

As to the nonlinearity, it's probably the case that you can learn to ride a bicycle or a motorcycle before you learn to walk, although one would think that there is a certain natural progression. I would also wonder if your capacity to perform some of the more complex tasks necessarily rubs off on your ability to accomplish the more basic things. But I can be persuaded that in fact that has happened.