An Agenda for Development
for The Twenty-First Century

This is an exciting time for those committed to advancing economic growth, reducing poverty and sustaining policy reform in developing countries and countries that are making the transition to a market economy. The success, not just of one but of several countries in breaking out of the poverty in which they had been mired for centuries, shows that development is possible. In Latin America, the debt crisis and the growth stagnation to which it gave rise seem to be behind us, and the latest data show that developing countries are growing faster than industrial countries. In fact, between 1991 and 1995 the growth rate of high-income countries was 2.5 percent, while that of low- and middle-income countries was 4.5 percent. Although the financial crises in East Asia have attracted much attention lately, they should not obscure the amazing achievements of the East Asian countries. Per capita income in the Republic of Korea increased tenfold in just over three decades. There is almost no one in Korea, Malaysia or Thailand living on less than US$1 a day, and Indonesia is within reach of that goal. Even Africa, where many countries experienced negative growth in the 1970s and 1980s, has at last started to experience growth, and countries such as Uganda that have sustained reforms over several years are showing consistent growth averaging 6 percent — still not in the league of China, but far better than was the case a few years ago.

It is now clear that countries that pursue appropriate policies have a better chance of economic success than those that do not. And there is mounting evidence that economic assistance, when combined with good policies, promotes economic growth, especially among the poorest countries. That is, of course, good news not only for the countries involved, but also for those who offer advice and dispense aid: They can make a difference. The challenge is to understand which policies are appropriate and how to target assistance to promote growth and reduce poverty most effectively. There is clearly no magic formula: If there were, the number of successes would be far higher than it is. And the fact that the messages that have been emphasized have changed over time — and that many of the countries that were successful did not take the particular medicine that was then being dispensed by the development community — should, at the very least, induce a modicum of humility as future directions are considered.

Changed Perspective

It is an exciting time too because of the fast pace of change in the world and the concomitant changes in the world of ideas — changes that inevitably would have necessitated modifying the development strategy, even if there was one that had worked unfailingly in the past. Over the past decade, three changes in particular have influenced thinking about effective development strategies.

Collapse of the Socialist Economies

This event provided an immediate lesson. For almost a hundred years, two theories had competed for the hearts and minds of people struggling to break free of poverty — one focusing on markets, and the other on government. The failure of the socialist economies appeared to demonstrate that the second model was not viable. But another conclusion sometimes drawn, that markets by themselves would provide the answer, also has not been justified by economic theory or by historical experience. As I argue in my book, Whither Socialism, had the assumptions under which markets have been shown to yield (Pareto) efficient outcomes been satisfied, market socialism would have been a success — or at least not the great failure that it proved to be. Neither has the historical experience been kind to the view that by themselves markets would have generated development: Almost all the major successes — the economies of East Asia, the United States, many countries in Europe — involved heavy doses of government involvement. There is a remarkable similarity between the government activities undertaken by the East Asian tigers and those undertaken by the United States in a comparable period of its development. There are few examples on the other side, of success without government involvement.

One of the striking lessons that emerged from the post–Cold War era is how difficult it is for markets to get established. Those in the more advanced industrial economies take for granted a rich institutional infrastructure, much of which requires government action to establish and maintain.

Success of the East Asian Economies

This experience has had an intellectual impact, by teaching that development is possible and that successful development requires (or at least is enhanced by) governments undertaking appropriate policies that go well beyond simply getting out of the way of the market. While there remains an active debate about the precise lessons to be learned and the extent to which the experiences of East Asia are replicable elsewhere, there remains little doubt that government played a critical, catalytic role.

But East Asia’s success has had another consequence: These economies have been an engine of growth for much of the rest of the world. US merchandise exports to East Asia, for instance, more than doubled between 1990 and 1996, and exports to developing countries were responsible for roughly one-sixth of US GDP growth in this period. China, while still a developing country in terms of GDP per capita, could soon become the second largest economy in the world. Developing countries have become an economic force to be reckoned with.

Globalization of the World Economy

This development is partly attributable to the tremendous drop in transport and communication costs in recent decades. Globalization has increased trade and capital flows, and developing countries that have opened their economies have especially benefited. Since 1990, private capital flows to developing countries have increased more than sixfold. And East Asia’s success was driven by exports: Production was not limited by the size of domestic economies, competition in exporting raised standards and increased efficiency, and exports helped transfer advanced technology and management practices. Increasing integration of the world economy has meant that a factory in Indonesia can be closely linked with markets in the United States and Europe, with changes in preferences and demands there being quickly translated into changes in what is produced. Computer programmers in India can be linked with programmers in California’s Silicon Valley. Distances have been shrunk, and geographical isolation has been reduced. These trends are just beginning, and they open up a host of new opportunities and challenges for all countries — and for developing countries, the hope of narrowing the knowledge and resource gap that has separated them from more developed countries, but also the challenge that the target they are chasing may be moving that much faster as well.

These changes in the global economy must guide development strategies in the coming decades.

Changed Objectives

Another set of changes was equally important in affecting thinking about development strategies — a change in objectives. It used to be that development was seen as simply increasing GDP. Today, there is a broader set of objectives, including democratic development, egalitarian development, sustainable development, and higher living standards.

It is recognized that there is more to living standards than is typically captured in GDP accounting. Improvements in education or health are not just means to an end of increased output, but are ends in themselves. Growth by itself does not ensure that the fruits will be equitably shared. It is recognized that there are costs, both to individuals and to society, of economic insecurity. Finally, the environment can no longer be taken for granted — in the struggle to increase GDP, the air in many developing country cities became so polluted as to make them almost unlivable. Moreover, it is now clear that cutting down an irreplaceable hardwood forest provides an increase in measured GDP that is probably not sustainable.

Fifty years ago, the common wisdom was that there was a tradeoff between rapid growth and democracy — Russia could grow faster, but at the cost of basic liberties. In retrospect, it is clear that people living under totalitarian regimes gave up their freedoms and sacrificed economic growth. In a regime without a free press, millions of people can starve, even starve to death, without public outrage mobilizing efforts to save them. And within developing countries, there is a growing recognition of the virtues of democratic development as an end in itself.

This broadened set of objectives leads to quite different development strategies: For instance, democratic development leads to increased emphasis on participation and the development of political institutions and education.

As always, perspective must be maintained: While increasing GDP is not an end in itself, or not the only end, increasing GDP is essential to achieving the other objectives. Raising education levels and improving the health status in a country require resources, and resources are scarce. The fact that some countries have not used their increased resources in ways that comport with these broader development objectives should not be confused with the fact that increased resources expand a country’s opportunities.

Similarly, while greater participation is essential to democratic development, the actions to achieve it must be constantly scrutinized: How representative, for instance, are those whose voices are being expressed? This is a concern even with respect to democratically elected leadership when voter turnout is small: extreme groups may be more likely to express their views. And in some countries the structure of electoral processes in primaries has resulted in centrifugal forces that appear to have more than offset the centripetal forces that Hotelling emphasized. There is concern, too, about the influence of money in electoral processes. But electoral processes in which there is a simple rule of "one person, one vote" still provide the most systematic way of ensuring representativeness.

In some cases, participation can improve other outcomes (for instance, the amount of learning that occurs). But in other cases, the effect on outcomes may be less certain. Some studies have found no correlation between rankings of perceived health impacts of various environmental hazards by scientists and rankings by nonscientists. Those who want to ensure that funds are spent on improving the environment to reduce overall health risks would surely want to rely more on the informed judgments of scientists than on a broader set of participatory views — while at the same time recognizing the importance of educating people more broadly about the scientific evidence on actual health risks associated with various hazards.

Reflections on the Past

Given the changes in the world, in our ideas, and in our views of development objectives, it is not surprising that development strategies have changed markedly over the past half century. When I was a graduate student, development planning was all the rage, and development economics was a mechanistic exercise. A country had a certain amount of resources that had to be allocated efficiently. Development strategies consisted of increasing those resources — in particular, the amount of physical capital — and ensuring, through development planning, that those resources were optimally used and that the various investments were coordinated. Government was required because markets were not sufficiently developed to provide the signals for resource allocation or to perform the coordination role that they were supposed to perform. Even when behavior was introduced into these models, it was done in a mechanistic manner. Savings rates were fixed, and they were lower among workers than among capitalists. Increasing the savings rate thus required shifting the distribution of income toward capitalists and away from workers.

Throughout this period, however, another strand of thought argued that government was part of the problem, not the solution; that more private initiative and entrepreneurship was required; and that government was inhibiting that private initiative. In the academic literature, this perspective was reinforced by data showing the responsiveness of peasants to price signals: Peasants in developing countries were just as rational as their urban cousins in the developed world. From here it was an easy step to suggest that the government should simply get out of the way, liberalize trade, and get the prices right. Development would follow.

This simplistic advice ignored the background from which the development planning literature had emerged: Many developing countries lacked markets for many commodities, and there was little reason to believe that markets would develop on their own. Market imperfections — imperfections that imply that the resource allocation that the market would provide on its own will not be (Pareto) efficient — while common in industrial countries, are rife in developing countries. Not only were more markets absent, but competition was often more limited and information more imperfect.

There was a curious incongruity: Economic practitioners were preaching the free market gospel just as economic theorists began to realize its lack of robustness. One of the central theoretical results of the 1950s was to establish rigorously the conditions under which Adam Smith’s insight about the invisible hand guiding the efficiency of markets is valid. Yet one of the central theoretical results of the 1980s was to show that whenever information is imperfect and markets are incomplete — which is essentially always — markets are not even constrained Pareto optimal. In other words, taking into account the costs of acquiring information and establishing markets, there are interventions that in principle could make some individuals better off without making anyone else worse off.

To be sure, another strand of thought emphasized not the perfections of markets but the imperfections of government. These were taken to be inevitable — or at least a judgment was made that it was better to ignore market imperfections than either to attempt to improve the performance of government or to use imperfect governments to correct the market imperfections.

The debt crisis of the 1980s shifted the focus to macroeconomics: Countries could not grow if governments did not provide a stable macroeconomic environment. Governments needed to hold their expenditures to their revenues and to limit the expansion of the money supply.

Many governments began following what came to be called the Washington consensus: They liberalized trade, achieved macroeconomic stability, and got the prices right — yet growth did not follow as quickly or as strongly as envisaged. By contrast, the governments of East Asia took a less dogmatic approach: While they achieved macroeconomic stability, they intervened extensively in the market. They helped create and regulate markets, and used them to achieve their development objectives. They also experienced the most rapid growth. Since this paper was presented at the Mediterranean Development Forum in May 1997, many East Asian countries have experienced serious financial turmoil. In my view, far from a refutation of the East Asian miracle, this turmoil may, in part, be the result of departing from the strategies that have served these countries so well, including well-regulated financial markets. In part, too, it may be the result of the failure to adapt quickly enough to changing circumstances, especially in the international financial system.

The New Perspective

The new agenda is informed by these experiences and is responsive to the changes in the economic environment and the broadened set of objectives noted above. It sees government and markets as complements rather than substitutes. It takes as dogma neither that markets by themselves will ensure desirable outcomes nor that the absence of a market, or some related market failure, requires government to assume responsibility for the activity. It often does not even ask whether a particular activity should be in the public or the private sector. Rather, in some circumstances the new agenda sees government as helping to create markets — as many of the East Asian governments did in key components of the financial market. In other areas (such as education), it sees the government and the private sector working together as partners, each with its own responsibilities. And in still others (such as banking), it sees government as providing the essential regulation without which markets cannot function.

And behind all of this lies a special responsibility for government: to create the institutional infrastructure that markets require in order to work effectively. At a minimum, this institutional infrastructure includes effective laws and the legal institutions to implement them. If markets are to work effectively, there must be well-established and clearly defined property rights; there must be effective competition, which requires antitrust enforcement; and there must be confidence in the markets, which means that contracts must be enforced and that antifraud laws must be effective, reflecting widely accepted codes of behavior. Laws that ensure a level playing field (for example, that restrict insider trading in securities markets) are not just matters of consumer protection: Without such laws investors will be reluctant to invest their funds in securities markets lest they be cheated.

The partnership between the government and the private sector has other dimensions.

Financial regulations that ensure the safety and soundness of banking institutions not only help mobilize capital, by giving depositors more confidence in the banking system, they also help ensure the efficient allocation of investment. (The dangers from excessive risk taking or looting that occur when banks are undercapitalized have been widely discussed in the aftermath of the banking crisis of the 1980s and 1990s.)

Government support for education helps ensure a supply of well-trained workers.

Government either helps provide infrastructure or provides a regulatory structure that ensures the private provision of infrastructure at reasonable prices.

Government often plays a vital role in developing and transmitting technology, such as through agricultural extension services.

Government can help promote equality and alleviate poverty, policies that in East Asia contributed to overall growth.

In each of these areas, the rationale for government action can be found in the theory of market failure. For instance, knowledge (especially its production) is a public good, and like other public goods it will be undersupplied. But the rationale for government action in these areas can also be found in the historical record: Economies in which government performed these roles well also performed better.

The exact role of government will change over time. It used to be thought that telecommunications and electricity as natural monopolies in which competition was not viable. Government would either have to regulate these industries closely or take charge of production. Today, partly because of changes in technology and partly because of changes in thinking, it is known that competition is viable in large parts of these sectors if an appropriate regulatory structure is in place. What is required is not deregulation, the naive stripping away of regulations, but regulatory redesign, that is, the changing of the regulatory structure in ways that promote competition where it is viable and that ensure that monopoly power is not too badly exploited where it is not.

Improving Government Performance

Once the vital role that government plays in a development strategy is recognized, the next step is to determine how the performance of government can be improved. There is a great deal that is known about markets and market failures, but far less understanding about governments and government failures. Still, a few things are known.

Governments can become more effective if they use market and market-like mechanisms. Auctions (both for procurement and for the disposition of government assets) can increase transparency (and thus confidence in government), improve the allocation of resources, and bolster the government’s budgetary position. Rather than addressing specific actions or transactions, government regulations may be designed to establish incentives for private firms to act appropriately — for example, by ensuring that banks have adequate net worth. Similarly, government may require commercial firms to have fire insurance, leaving it to the insurance company to ensure that appropriate precautions are taken (such as installing sprinklers), knowing that the insurance company has appropriate incentives.

Governments that are less frequently subjected to temptation are less likely to give in to it. Discretionary actions, such as the allocation of quota rights, provide opportunities for corruption. Corruption has adverse effects on economic growth (partly by raising the costs of doing business and thereby discouraging investment) and undermines confidence in government Thus using transparent, market-like mechanisms has a double set of advantages.

Governments should constantly reexamine the rationale for the regulations they have imposed. Historical processes often lead to regulations that are not appropriate to today’s circumstances: The objectives of the regulations could be achieved at far less cost. For instance, before there were devices that could measure the pollution being emitted by a smokestack, it might have been appropriate to require technologies that met environmental standards. But once reliable monitoring devices are available, regulations should focus on outputs (here, the level of emissions) rather than inputs (technology standards.)

The quality of a government’s bureaucracy depends on the composition and quality of its workforce. Those in turn depend in part on its internal systems for hiring, training, and promotion and, as with any enterprise, in part on wages. Low wages and a mismatch between rewards and results discourage output and result in low performance.

Government can help stimulate and can use competition: It can create competing public agencies, and it can allow private agents to compete with public agencies. The success of market economies is based on deep institutions that support incentives built around private property and competition. While there is ongoing debate about the relative importance of private property and competition, a plausible case can be made in a variety of settings that competition is what is essential. (Some view the success of China as bearing witness to this proposition. More narrowly, studies of government-run enterprises that are subjected to competition suggest that they can perform as effectively as private firms. In large organizations, whether public or private, principal-agent problems arise; the scope for putting in place individual incentives may differ little between public and private organizations.) One of the central themes of the World Bank’s report on the East Asian miracle is that competition played a crucial role (especially competition for exports).

Sequencing of Reforms

The simple lesson that emerges from this discussion is that incentives matter: that they matter in both the public and the private sector, that government should make more extensive use of incentives to guide its behavior, and that government should take actions to improve incentives in the private sector. (The variety of restrictions within the banking sector are examples of actions taken to improve the incentives of banks.)

By the way it sequences reforms, government can affect not only the performance of the economy in the short run but also the momentum for the continuation of reforms. Consider, for instance, the consequences of privatization in a large, closed economy before trade is liberalized and before an effective competition law is in place. Under these conditions, privatization would convert a government monopoly into a private monopoly. Consumers (or firms who use the privatized firm’s output as input) probably would not see any benefits (real prices might actually increase), overall economic performance probably would fail to improve (increased efficiency within the privatized firm is offset by loss of systemic efficiency from higher prices), and a vested interest might be created that would have the incentive and the resources to work against efforts to implement an effective competition law. A regulatory structure might evolve, but it would arrive already compromised, captured by those whom it should regulate.

Or consider the consequences of the privatization of a utility before an effective regulatory law is put in place. Uncertainty about the future regulatory structure could cause the government to receive less from the sale of the assets than it should. There is also the possibility that users, both consumers and firms, will face higher prices and that a vested interest will be created that has the incentive and the resources to resist effective regulatory reform. Moreover, in the absence of an effective regulatory structure, other public objectives that had been pursued by the industry before privatization — such as universal service — may be abandoned. High profits, low sales price, and decreased service will all undermine public support for the continuation of reform.

A regulatory structure can be created to ensure that some of the efficiency gains from privatization are shared by consumers and other users and that other social objectives, such as universal service, are enhanced. But the proposition that privatization can, in principle, increase the efficiency of the economy and achieve other social objectives should never be confused with the proposition that, in the absence of effective regulatory structures, privatization may do neither in practice.

Neither economic theory nor historical experience gives clear guidance in these complicated matters. (Even in the case of privatization the principal theoretical proposition, the fundamental privatization theorem, shows how restrictive the conditions are under which privatization can guarantee a welfare improvement.) Theory may predicate that under certain idealized circumstances market economies are efficient. But theory provides less clear prescriptions for the inevitable second best situations in which many of the idealized circumstances underlying the pure theory are not satisfied. And theory gives even less clear guidance on sequencing: how to get from here to there. For these decisions, experience and judgment — as well as perhaps the guidance that comes from a careful consideration of cross-country experiences — must be relied on.

The Role of High-Income Countries

Most of this discussion has focused on what developing countries must do to succeed in enhancing their economic growth. But there is much that high-income countries can do as well.

The enormous increase in private capital flows to low- and middle-income countries has been a boon to countries that have created a welcoming investment environment. But the magnitude of these flows should not conceal the fact that they have been very concentrated: Ten countries, representing half of the developing world’s GDP, have received 75 percent of the funds, and only a handful of low-income countries have received any significant amounts relative to the size of their economies. Moreover, these private flows have not gone to all sectors of the economy — health and education, for instance, remain largely within the public sector.

That is the reason why the decline in official development assistance is so disturbing. In 1996, such aid, measured as a percentage of high-income countries’ GDP, reached an almost fifty-year low. It is ironic that such cutbacks are occurring just as evidence of the effectiveness of aid is mounting.

Development assistance promotes economic growth, especially among low-income countries, if they have put in place good economic policies. True, aid cannot buy reform, and it works best for countries that can help themselves. But as a complement to local initiatives, aid can enlarge and precipitate the positive effects of reforms, and by helping to build political momentum it can make the reforms sustainable. More important, aid may be an effective instrument for transferring knowledge and, as pointed out below, the diffusion of appropriate and well deployed knowledge can greatly enhance development prospects even in the poorest countries.

Some of the high-income countries that have been least forthcoming with aid have emphasized the importance of trade. But here too there is a gap between rhetoric and reality. For instance, while preaching the virtues of a market economy, Western producers accused Russian firms that were selling their excess aluminum at international prices of engaging in dumping, and encouraged their governments to impose trade restrictions. As an alternative, they proposed establishing what in effect would be an international cartel to restrict worldwide production and raise prices — which is what was done.

How can the advanced economies preach the gospel of competition and free markets, yet turn to managed trade and restricted markets when their own interests are in jeopardy? There are many, many more examples, but the point is clear: The advanced economies should be the role models. They are far more able to absorb the shocks that inevitably occur as a result of changing trade patterns. Why is it that they seemingly find it so acceptable to refer to "political pressures" forcing their deviations from policies of openness while finding such excuses so hard to take from developing countries?

The World Bank’s Role in the Development Agenda

Some analysts claim that the huge flows of private capital to developing countries have made the World Bank unnecessary. I obviously disagree. The rationale for the World Bank and for development banks more generally can be summarized as follows.

Private capital flows are targeted. Most low-income countries receive very little of this capital, and little of the money goes to vital sectors, such as health and education, that are complementary to the private flows.

The World Bank and the other international financial institutions have an important role in helping developing countries establish the institutional infrastructure (regulations and laws) that is required to attract capital. This infrastructure is a public good; thus the private sector cannot be expected to provide adequate assistance in its establishment. Moreover, in a world where competing private interests are always looking for the establishment of rules and regulations that favor themselves, the World Bank can serve as an honest broker.

While private capital flows are much more varied in form than they were in the past — there are now substantial equity flows — there are still important gaps.

The World Bank has a distinct advantage in gathering information and producing knowledge about successful development practices and policies. Knowledge is an international public good that will be undersupplied if left to the market.

In some cases, the World Bank and other multilateral development banks may signal that a developing country has embraced sound policies and hence boost its credibility, providing an additional incentive for maintaining these policies in the future.

World Bank staff have a special responsibility not only to produce knowledge that will enable developing countries to grow more effectively, but also to ensure that that knowledge translates itself into appropriate action — including in its lending practices. A reference was made above to research showing the effectiveness of aid when good economic policies are in place. The flip side of that research suggests that when good economic policies are not in place, aid may have no effect on economic growth. This finding poses a moral dilemma: If a loan does not increase an economy’s resources, unless it is explicitly aimed at achieving other social objectives, it leaves future generations more impoverished, since they inherit the indebtedness. And even if it does achieve some other social objective, the tradeoffs need to be considered.

The issue of the fungibility of funds adds further complications: To the extent that funds are fungible, the issue is not the quality of the project being funded, but what happens as a result of the additional flow of funds. Evidence on the extent of fungibility (or the circumstances under which funds are more fungible) remains somewhat ambiguous. Yet recent studies indicate that, with the exception of sectors such as transport and communications, aid funds are largely fungible and foreign finance does significantly substitute for (rather than complement) domestic spending. But people living in the poorest countries that also do not have good economic policies cannot be ignored. They need help in putting in place good policies by providing advice and technical assistance. In addition, investments should be made in areas such as human capital that will be ready to support growth once good policies are adopted. And effective delivery mechanisms must be found when governments have demonstrated their lack of capability.

Conclusion

The development agenda for the twenty-first century outlined in this discussion includes a wider set of objectives than those of the past. It includes a changing role for the state — with a partnership between government and the private sector — that involves a catalytic function for government in helping to create markets. In some areas, it includes a more enduring role for government in regulating markets. And it requires governments to improve their own performance, partly by making more extensive use of market-like mechanisms through using and helping to create competition wherever it can.

For both developing and industrial countries, the new agenda requires an openness that both groups have often resisted. The mutual benefits are clear: Developing countries will continue to be an engine of growth for industrial countries, providing a broader variety of products at lower prices to their consumers and offering higher returns to their investors. For developing countries, the transfer of resources and knowledge will enable a continuation of growth.

As mentioned above, there is an important role in this agenda for the World Bank and other international financial institutions. But what that role is and how it can best be performed must undergo a continuing process of reexamination.

There is no room in this agenda for dogma or for doctrinaire approaches. The general consensus on basic economic reforms — keep inflation to a moderate level, limit the size of the fiscal deficit, avoid introducing large distortions in the economy, open the economy to foreign competition — addresses issues of fundamental importance and has made substantial contributions to stabilization in several countries. There is a danger that this consensus has become dogma and, as dogma, it may sometimes be applied inappropriately.

One of the principal advantages of the this consensus is its simplicity. But the policy agenda of the coming decade will not be as amenable to such a cookie-cutter approach. Regulatory regimes and legal and other institutional structures should be better adapted to country circumstances. And in many cases, reforms are far more technically complex than the consensus suggests. For instance, designing regulations that promote competition in telecommunications and electricity is extremely difficult, as the experience of the United States and other countries have shown.

There is a need to learn from theory and history, from best practices, and from what has worked. But care must be taken in extracting the appropriate lessons.

• The world today is different from the world thirty, twenty, or even ten years ago. Private capital flows are more important today than they were then; this opens up new opportunities — and new challenges. The fact that some East Asian economies made relatively little use of foreign capital may have little bearing on whether a developing country today should make more extensive use of these flows.

• Some of the most successful economies did not follow all the key prescriptions that are commonly given today. Much of the growth in GDP among all low-income countries can be accounted for by the growth in China. China focused on creating new enterprises and engendering competition rather than on privatizing state enterprises. Would these successful enterprises have grown even faster had China followed the alternative strategy? Most observers are doubtful. The Republic of Korea has been widely criticized during its 1997 financial crisis for its failed economic system — yet that system somehow increased per capita incomes tenfold in three decades, a record unmatched by any large country following the prescriptions now commonly proffered.

• What works in one set of circumstances may not work in others. Analysis is required to identify the factors that determine success and to establish the counterfactuals (what would have happened in the absence of a particular project or policy). Examples of best practices may provide useful and persuasive anecdotes, but they are no substitute for analysis.

• Care should be taken not to confuse means with ends. Macroeconomic stability, deficit reduction, and even enterprise reform are not ends in themselves, but means to the broader development objectives described above. If a government reduces its fiscal deficit by cutting back vital investments in infrastructure or in human capital, growth may actually suffer. If a government reduces its fiscal deficit by cutting back on food subsidies and that leads to rioting that undermines the country’s political stability, is that likely to make the country more or less attractive to foreign investors?

• Ends also must not be confused with means: Improved education and health are essential means of increasing GDP, but they are also ends in themselves.

• The broadened set of objectives is an essential aspect of the new development agenda. But while it is recognized that these is a richer set of objectives, the constraints provided by limited resources are no less binding. In the words of one of my colleagues, these issues must be approached with soft hearts and hard heads. There is a difficult tradeoff: It might be possible to reduce poverty more today, but only at the expense of fewer resources — and therefore more poverty — in the future.

• One of the lessons of East Asia’s experience is that there are important instances of policies that increase economic growth, promote equality, and ameliorate poverty. The search for such policies — and their implementation — must be at the center of the development agenda of the future.

• If one truly believes in democratic development, the limited role of technical advisers must be recognized. They can offer judgments about the consequences of alternative policies. But an essential part of the new development strategies involves the creation of institutions and the changing of cultures — the movement to a culture of change and science, where existing practices are questioned and alternatives are constantly explored. (These cultural changes were, of course, the subject of considerable discussion in the development literature a half century ago.) Deciding how best to effect such changes requires a great deal of local knowledge, and it is not obvious that development advisers have either that local knowledge or the corresponding technical expertise. Moreover, democratically elected governments must, in the end, make the judgments about both the tradeoffs and the political consequences. These are principles that apply to advisers in advanced as well as in developing economies. Having said that, it is important to emphasize that there are vast areas in which technical expertise is highly relevant. For example, certain fundamentals in establishing effective banking regulations apply to all countries.

Today, development is recognized as more than the accretion of physical capital and even more than the accretion of human capital. It includes closing the knowledge gap between rich and poor economies. And it includes other transformations, such as those that result in lower population growth rates and changes in economic organization.

I am confident that the coming decade will certainly see enormous growth in the developing world and a reduction of poverty. It will be a struggle. The challenges are great, but the opportunities are many.

The challenge for those who advise governments will be to strike hard-to-find balances between the roles of the state and the roles of the private sector, and between the doctrinaire positions that have often characterized the policy advice of the past and the agnosticism that gives little guidance to those struggling to make the hard choices that will affect millions of lives.

Further Readings

Arnott, Richard, Bruce Greenwald, and Joseph E. Stiglitz. 1994. "Information and Economic Efficiency." Information Economics and Policy 6(1): 77–88.

Burnside, Craig, and David Dollar. 1997. "Aid, Policies, and Growth." Policy Research Working Paper 1777. World Bank, Policy Research Department, Washington, D.C.

Daves, D.W., and L.R. Christensen. 1980. "The Relative Efficiency of Public and Private Firms in a Competitive Environment: The Case of Canadian Railroads." Journal of Political Economy 88 (4): 958–76.

Dr�ze, Jean, Amartya Sen, and Athar Hussain, eds. 1995. The Political Economy of Hunger: Selected Essays. New York: Oxford University Press.

Feyzioglu, Tarhan, Vinaya Swaroop, and Min Zhu. 1997. "Foreign Aid’s Impact on Public Spending." Policy Research Working Paper 1610. World Bank, Policy Research Department, Public Economics Division, Washington, D.C.

Greenwald, Bruce, and Joseph E. Stiglitz. 1986. "Externalities in Economies with Imperfect Information and Incomplete Markets." Quarterly Journal of Economics 101 (3): 229–64.

— — — . 1988. "Pareto Inefficiency of Market Economies: Search and Efficiency Wage Models." American Economic Association Papers and Proceedings 78: 351–55.

Hellman, Thomas, Kevin Murdock, and Joseph Stiglitz. 1997. "Financial Restraint: Towards a New Paradigm." Stanford Graduate School of Business Research Paper, Stanford, California.

Pack, H., and J.R. Pack. 1990. "Is Foreign Aid Fungible? The Case of Indonesia." Economic Journal 100 (March): 188–94.

— — — . 1993. "Foreign Aid and the Question of Fungibility." Review of Economics and Statistics 75 (May): 258–65.

Sappington, David, and Joseph E. Stiglitz. 1987. "Privatization, Information, and Incentives." Journal of Policy Analysis and Management 6: 567–82.

Slovic, Paul, Mark Layman, and James H. Flynn. 1993. Perceived Risk, Trust, and Nuclear Waste: Lessons from Yucca Mountain. Durham, N.C.: Duke University Press.

Stiglitz, Joseph E. 1990. "The Economic Role of the State: Efficiency and Effectiveness." In Thomas P. Hardiman and Michael Mulreany, eds., Efficiency and Effectiveness. Dublin: Institute of Public Administration.

— — — . 1994. Whither Socialism. Cambridge, Mass.: MIT Press.

— — — . 1996. "Some Lessons from the East Asian Miracle." The World Bank Research Observer 11 (2): 151–78.

Stiglitz, Joseph E., and Marilou Uy. 1996. "Financial Markets, Public Policy, and the East Asian Miracle." The World Bank Research Observer 11 (2): 249–76.

Stiglitz, Joseph E. 1997. "An Agenda for Development for the Twenty-First Century." In Pleskovic, Boris, and Joseph E. Stiglitz, eds. Annual Bank Conference on Development Economics 1997. The World Bank, Washington, D.C.

Taylor, Lance. 1979. Macro Models for Developing Countries. New York: McGraw-Hill.

USEPA (US Environmental Protection Agency). 1987. Unfinished Business: A Comparative Assessment of Environmental Problems. Washington, D.C.

World Bank. 1993. The East Asian Miracle: Economic Growth and Public Policy. A Policy Research Report. New York: Oxford University Press.

— — — . 1994. Adjustment in Africa: Reforms, Results, and the Road Ahead. A Policy Research Report. New York: Oxford University Press.

— — — . 1997a. Global Development Finance 1997. Washington, D.C.

— — — . 1997b. Global Economic Prospects and the Developing Countries. Washington, D.C.

— — — . 1997c. World Development Report 1997: The State in a Changing World. New York: Oxford University Press.

---