Research Challenges for Development Economists

International Monetary Fund. External Relations Dept.
Published Date:
January 1991
  • ShareShare
Show Summary Details

Research Challenges for Development Economists

Much of what is said about the propensity of economists to disagree is vastly exaggerated, as evidenced by the striking degree of consensus in the advice that the broad mainstream of economists has given on the transition in Eastern Europe. But certainly, there is much we do not know. Despite rapid technological progress that has created all sorts of new economic opportunities, living standards over the last decade have actually declined on two continents—Africa and Latin America. It seems that we, as economists, have more ability to identify and reject quack prescriptions that violate necessary conditions for development and growth, than we have to provide policies that ensure sufficient conditions.

How can we help lift billions out of the poverty trap and set them on the path to sustainable development? This article argues that more than ever before, the central priority for the World Bank, and indeed for the entire aid community over the next decade, is to create and help implement improved strategies for economic development. These strategies must rely, to a greater extent than before, on the transfer and transformation of knowledge, so as to compensate for the expected paucity of development assistance in the waning years of this century. To put it bluntly, since there will not be much development money over the next decade, there had better be a lot of good ideas.

Money vs. ideas

Consider the historical volume of transfers to the developing countries and the projected volume of aid for the next decade. For the developing world as a whole, aggregate net transfers in 1991 dollars totaled $50 billion in 1980. Over the last five years, however, these transfers were significantly negative on average. A restoration of the earlier levels of flows is not in sight. While the cessation of net positive transfers to the developing world during the 1980s reflected, in large part, the commercial banks’ response to the debt crisis, the prospects for large net transfers from institutions like the Bank have declined as well. Over the 1975-80 period, the World Bank Group made net transfers to the developing countries of $9 dollars per person in the developing world (in 1991 dollars). Over the next five years, these net transfers are projected to decline to just over $2 per person in the developing world. For a large number of countries that do not qualify for concessional IDA flows, there will actually be a negative flow of money to the Bank (see table).

Resource transfers will be limited. Yet this potential financial constraint need not diminish the effectiveness of our role. While aid flows are crucial, the reality is that improving the efficiency of development assistance to the developing world will often make a greater contribution to growth. Consider, for example, the following comparison: a two tenths of one percent increase in total factor productivity in developing countries would contribute more to their GDP growth than an additional $100 billion of capital invested at historical rates of return.

This suggests that, in the end, what will make the greatest difference for development are well-implemented, good strategies, and not simply on the transfer of resources. Consider the most widely cited of all foreign aid successes—the Marshall Plan. During the Marshall Plan years, the United States granted the recipient countries less than 2.5 percent of their incomes per year, and less than one third of the resources transferred were used directly for investment. Yet what the Marshall Plan did, and on a major scale, was to help provide economic stability and allow governments to provide an enabling environment in which the private sector could flourish.

Public inaction vs. action

With money scarce, ideas are needed. But what type of ideas? The great lesson of research and experience in the 1970s and the 1980s was that governments that tried to occupy the commanding heights of their economies found themselves looking a long way down, to stagnating growth and deteriorating performance. The policy lesson is clear enough: Governments must get out of producing and allocating goods and services—activities that are best performed through competitive markets. That message has been heard around the world and to great effect. One need only consider the booming economy of Mexico, the vast increases in agricultural productivity obtained in China, or the bold reforms being undertaken in Eastern Europe, to see its power.

The market message was new and true. But it is only part of the story. Getting the government out of economic activities is not the correct answer every time. Sometimes, governments will be called on to do more of what they must do, and to do it better—making the tangible and intangible infrastructure investments that underpin a healthy private sector and rapid growth, and ensuring social and economic justice.

Undertaking such activities effectively is much more difficult than divesting them, especially in a resource-constrained environment. We have concentrated attention in recent years on governments’ errors of commission. It is time for more work on governments’ errors of omission. The Bank’s research agenda for the next few years, therefore, will address the pivotal issue of helping governments become more effective in identifying and performing the central functions that only they can do.

What do we need to know?

An important dimension of any research program should be helping to set the priorities for tomorrow, as well as responding to the needs of today. To this end, the Bank will continue working on its current areas of operational emphasis—the environment, reform of socialist economies, debt, adjustment, and so forth. Against this background, the Bank will need to re-examine and revivify its work in a number of traditional areas.

Evaluating infrastructure investments. Perhaps the most traditional role of government is investing in public goods— broadly defined—and getting as much return from those investments as possible. Helping governments make necessary infrastructure investments was, after all, an important rationale, perhaps the central rationale, for the creation of the Bank. Economic logic suggests that the Bank’s capital should be allocated to activities that produce the highest returns. But that is not the way the Bank and other development institutions operate. Our portfolio is highly diversified, with investment in many sectors in most countries. Why? I fear that we practice a kind of diversification out of ignorance. Not knowing what types of projects have the highest returns, we invest in many different things, as do our member governments.

World Bank net transfers to the developing world(In millions ot 1991 US dollars)
Net transfers
Total net transfers28.3688.696
(In US dollars)
Net transfers per
Source: World Debt Tables, and other World Bank sources.
Source: World Debt Tables, and other World Bank sources.

Yet, there is every reason to think that there are large differences in the social rate of return to different types of project lending. The Bank’s own operational evaluations suggest a large variation in the rate of return to projects in different sectors. This raises, as a vital question for research, the challenge of making tough judgments about the social return to different types of investment. Of course, comparing bridges to power plants is like comparing apples to oranges. But seeking to compare their returns analytically must surely make more sense than investing in both to ensure that the better one is included in the portfolio.

Making comparisons of investment returns across different types of investments will demand techniques of analysis that go beyond traditional cost benefit analysis, to cover a variety of externalities that different types of projects produce. For example, recent trade theory has shown that the gains from free trade go beyond reduced “deadweight loss triangles” (i.e., reducing the additional costs incurred by consumers as a result of trade restrictions), to include increased efficiency, reduced rent-seeking, and the gains associated with the exploitation of economies of scale and increased product variety. Yet these gains are hard to quantify with standard methods. Similarly, improved transportation no doubt produces many economic benefits beyond reducing travel time, yet such considerations have not been fully or even partially accommodated in our analyses of the attractiveness of this type of lending. Our goal must be to devise ways of making more systematic and more complete evaluations of different types of projects that recognize the full range of their impact.

Pushing market development. A different kind of infrastructure that government provides is the development of markets that otherwise would not exist. Advocates of laissez faire would do well to remember that there was no commercial mortgage market in the United States immediately after World War II. It took the Federal Housing Administration to demonstrate that loans for houses were feasible and that lenders could collect those loans and make a profit, while allowing people who could not otherwise afford homes to purchase them. That experience, and its message, is relevant for institutions and places where capital markets are much less developed than they were in the United States even after World War II.

There is also a major and obvious need to help countries and companies guard against the enormous risks they face from highly volatile international financial and commodity markets. Countries are also exposed, in complex ways, to changes in interest rates, and in their own and their competitors’ exchange rates. The challenge is to find ways, and to help governments find ways, of using more effectively the vast world capital markets to spread risk and permit it to be borne more efficiently. Often this will involve enabling speculators in developed countries to take on risks that developing countries can ill afford. In some cases—for example, in the case of developing country oil importers and oil exporters—this can be done within the developing world.

Attracting capital. Capital market development also involves attracting capital. As mentioned at the outset, recent trends suggest that the flow of capital to the developing world will, even in fairly optimistic scenarios, be much smaller in the near future than it has been in the recent past. Commercial bank flows to many developing countries will not be restored any time soon. The downside risk associated with lending to the developing world became obvious during the last decade. As a result, in the coming years, investors will demand a large potential upside to their investments in developing countries, so as to balance the downsides they faced in the past. In order to replace commercial bank lending, the developing world will need equity flows and direct foreign investment on a large scale. The Bank and the International Finance Corporation (particularly the latter) have done important work in catalyzing those flows. But equity flows are still very small on the whole. We need to think about ways of leveraging them more effectively, because it is equity not debt flows that will effect large transfers or large foreign-financed investment in developing countries during the next decade.

Achieving economic justice. The World Bank is committed to poverty reduction. But poverty alleviation is a means, not an end; it is a way of allowing people to live healthier, longer, more comfortable, and fulfilling lives. We know a great deal about the distribution of income, and a fair amount about how to change it, but, as evident in the work done for the World Development Report 1991, there is little hard quantitative evidence on the effect of social sector interventions on health, literacy, nutrition, and so forth. Any effort to quantify the financial effects of lives saved and relate them to the value of original social sector investments is problematic to say the least. But it is not unreasonable to ask for some amount of information bearing on what can be accomplished through social sector programs.

Economic growth and social progress are related, and rising tides do lift all boats. But there are vast differences in social sector performance among countries at similar income levels. Among low-income countries, Indonesia, Sri Lanka, and the Republic of Yemen have almost exactly the same per capita income. Yet, life expectancy is 51 years in Yemen, 60 in Indonesia, and 70 in Sri Lanka. Among upper middle-income countries, Brazil and Uruguay have essentially the same income level, but life expectancies differ by six years. To put these figures in perspective, consider that eliminating all forms of cancer would add only three years to the average life expectancy in the United States, and much less in poorer countries.

What are the roots of the differences? What can be done to reduce them? The very different experiences among countries must demonstrate the vast potential of strategically designed, relatively inexpensive government interventions. But how? The answer must go well beyond targeting programs to reach specific groups. Differences in life expectancy or illiteracy of the kind described above cannot be ascribed simply to the failure of some countries to concentrate on the lower social strata of their population. They must reflect differences in all social strata. The problem of delivering those services seems to me an important one for the Bank’s future research.

Beyond the comparisons across countries, and of potentially equal or greater importance, is a micro-dimension. We do not have well crafted and tested systems of comparing smaller “control” groups of people within different parts of a society and tracking their socio-economic development over time. Medicine has agonizingly come to the conclusion over the last century that only double blind evaluations of drugs or carefully controlled statistical evaluations of surgical procedures have any validity at all. Without such comparisons, mistakes are incessant. Obviously, the scope for controlled evaluations is greater in some settings than in others. In the real world, not every ceteris can be paribus. But, I would suggest that too often we, as development economists, operate in a kind of statistical Stone Age. We would learn much from maintaining a standard policy of setting up baseline groups to use as controls in evaluating social sector projects or, to the extent that matched enterprises can be found, even projects directed at restructuring enterprises.

Social services are important. So is raising the income of those who work, and that means causing their skills to be more highly valued in the labor market. We must rechannel some of the energy we have directed at understanding the capital market to understanding the labor market. That will involve fathoming more accurately the impact of labor market regulations on output and participation rates, so their removal would become a more significant part of our adjustment programs. But it also means finding ways of harnessing labor markets to provide employment not for a small fraction of workers at very high wages, but for a much larger number of workers in labor-intensive industries. Human capital is a principal asset of the poor. Raising its return has to be a central investment, and, therefore, a research priority.

Population is another area, nearly unique among areas of development policy, where there is a disjunction between widespread seemingly compelling beliefs and the current state of the scientific literature. Perhaps this is not surprising given how much emphasis in recent years has been devoted to the determinants of population growth and how little to its consequences.

It is widely believed that population growth is a major barrier to economic development and a perpetuator of poverty in developing countries. Robert McNamara represented the views of many when he equated population growth and nuclear war. Yet a relatively recent National Academy of Science report provided a dramatic shift in viewpoint, reaching a rather agnostic view about the consequences of population growth. It seems likely that improving our understanding of the population question will require recognizing that reducing population growth has differing consequences in different circumstances. Changes in population growth are likely to have different outcomes when they are brought about by declines in death rates of older adults, declines in child death rates, responses of families to changes in social security arrangements, and specific efforts at promoting family planning. The impact of changes in population is also likely to vary across places. Where a large fraction of the population lives off the land, population growth is likely to be more pernicious than where the land has diminished in importance as a factor of production.

Beyond the traditional question of the impact of population on growth, there are a host of other questions that need to be answered, so as to guide a rational population policy. How strong are the links between population growth and environmental degradation? Is it really population pressure that is pushing migration into Rondonia in western Brazil? How much extra pressure does population, as distinct from other problems, put on African soils? What are the links between population growth and inequality in the wage structure? Does population growth affect the wage structure? What are its fiscal consequences? And will investments in population policies pay off for governments?

Gauging more accurately the impact of population growth on development performance is only the beginning of what we need to know about demographic change and its consequences. Over any but the longest horizon, there is very little that can be done about the rate of population growth. We need more effective policies to deal with the increasingly aged population of the developed world—and we must think through the pressures for migration that are rapidly building. If current projections hold, over the next two decades the developing world will account for 95 percent of labor force growth, but for less than 15 percent of capital investment. This calls for either transferring capital to places like Africa where the labor force is growing more rapidly, accepting large flows of migratory labor, or recognizing that an important force will be operating to increase inequality over the next 20 years.

Governance. Beyond the questions of what governments must or can do, is an issue that may well be the most important, and perhaps the most difficult to grapple with: the question of governance. We have learned over the last three decades that the heavy hand of government retards growth, and we have seen that, contrary to economists’ beloved tradeoffs, very often even as governments retard growth, they are also insensitive to objectives that are thought to conflict with growth. Concerns about the environment or inequality, which go beyond the bottom line, are not met more effectively by the heavy hand of government.

That much we know. We are much less clear on the right line between the invisible hand and government’s helping hand. It may be fair to assert that the rhetoric surrounding the use of selective credit policies in Japan and East Asia is not very different from the rhetoric that surrounds development banks and other tools of selective credit policy in other parts of the world. But the results have been dramatically different. In a sense, the policies must have been different also. Failed policies in Asia have been reversed much more quickly than failed policies in Latin America or Africa, and have not, therefore, mushroomed or bred the same kind of corruption. But in a deeper sense, the underlying policy intention has been the same in both places.

Why the difference in outcomes? It must go back in some way to politics—broadly defined. Perhaps differing degrees of urbanization or differing amounts of underlying inequality shape political forces differently, or so, at least, it has been argued. Just how susceptible these things are to change, one cannot say with certainty. The view that governments should simply swear off selective policies if they cannot do them well may be a bit naive. The same political forces that make governments unable to implement those policies well may make governments unable to resist engaging them.

Understanding more fully the political roots of economic success and failure is an intellectual task of great importance. By itself, it can contribute to judgments about where assistance is and is not likely to be effective. Beyond this question, there is the issue of crafting appropriate policy advice that reflects political realities, which insulates results from inherently political pressures, as in the case of countries that rely on foreign companies to collect custom duties; or that create dynamics that push the politicians in the appropriate direction, as in the case of making them substitute shorter prohibited for seemingly endless permitted lists of goods that can be imported.

There is a very sensitive and closely related question—that of the links between democracy, civil liberties, and economic performance. We have learned an important lesson over the last decade: authoritarian regimes are not necessary to bring about growth. Growth, on average, proceeds at about the same rate in democratic and nondemocratic regimes. We know two other things. Since World War II, there has been no war involving two democratic nations, and in the last 50 years, as Amartaya Sen pointed out at the Annual World Bank Conference on Development Economics in 1990, there has never been a famine in a country with a free press. These facts raise the question of whether democracy, while not essential to growth, may in a whole set of ways improve development performance on other dimensions. It is time that we look in much more detail than we have in the past at the relationship between democracy and development. We need, as well, to begin to think about what we can do, beyond economic policies that promote growth, to improve the performance of governments.

Recent events press yet another question related to governance. What is the future of nation states as independent actors? On the one hand, pressure for integration and cooperation mounts, as in the European Community. On the other hand, pressure for devolution increases in a number of countries. This raises questions of managing the transition to more autonomous constituent states within a nation-state to preserve the stability that is essential to growth.


In a world that is short of money, it is important to have ideas. Research spawns ideas. Focusing the development research effort over the next decade in the areas discussed above can be given the Pascal defense. To paraphrase Pascal, if there is no God and I go to church on Sundays, I will have wasted some time. If there is a God and I fail to go, the costs will be transcendently greater. So, too, with the search for new ideas. The payoffs to success dwarf the consequences of failure.

This essay is based on the keynote speech delivered by the author at the World Bank Annual Conference on Development Economics in Washington, DC, in April 1991. The full text of the speech will be appearing in a special supplement of the World Bank Economic Review and the World Bank Research Observer in 1992.

Other Resources Citing This Publication