Share This
December 2005 Cover Art

Search Finance & Development

Advanced Search
About F&D


Back Issues

Write Us

Copyright Information

Free Email Notification

Receive emails when we post new items of interest to you.

Subscribe or Modify your profile

F and D logo
A quarterly magazine of the IMF
December 2005, Volume 42, Number 4

The Inequality Trap
Francisco H.G. Ferreira and Michael Walton

Why equity must be central to development policy

A girl born to a lower-caste family of nine in the slums of Dhaka has vastly different opportunities from a boy born to well-educated and affluent parents in the well-heeled neighborhoods. An AIDS orphan in rural Zimbabwe is almost certain to have fewer chances and choices in life than a compatriot born to healthy and well-educated parents in Harare. Those differences are even greater across borders: an average Swiss, American, or Japanese child born at the same instant as one in a poor, rural area of South Africa will have incomparably superior life chances.

Such staggering inequalities in opportunity are intrinsically objectionable, and almost every culture, religion, and philosophical tradition has developed arguments and beliefs that place great value on equity for its own sake. But that is not all. We would argue—as elaborated in the World Bank's World Development Report 2006: Equity and Development—that there is now considerable evidence that equity is also instrumental to the pursuit of long-term prosperity for society as a whole.

In developing this position, we do not focus on inequalities of outcomes (such as incomes), but on a conception of equity in terms of equality of opportunity. A person's life prospects should not be influenced by circumstances outside his or her control—such as country of birth, gender, race, and family origins. Outcomes, by contrast, may well differ, as a consequence of differences in effort, talent, and luck. This approach draws on central trends in philosophy of the past few decades, notably in the work of John Rawls, Ronald Dworkin, Amartya Sen, and John Roemer. But we also recognize that societies may decide to intervene to protect the livelihoods of its neediest members (living below some absolute threshold of need) even if the equal opportunity principle has been upheld. For that reason, we include avoidance of absolute deprivation as a basic principle of equity.

Immense inequities persist

The shaping of opportunities begins before individuals are born. Who one's parents are, what country they live in, and how rich they are, make a large difference to a person's opportunities in terms of life expectancy, education, access to services, and economic prospects. As Chart 1 shows, infant mortality rates vary markedly both within and across countries. In El Salvador, for instance, babies born to mothers with no schooling are four times more likely to die before their first birthday than babies whose mothers are better educated.

Chart 1. Staying alive depends on circumstances

Moreover, inequities persist during childhood. Consider the differences in test performance among Ecuadorian children ages three to five years, across population groups defined by parental education, region of residence, and wealth. As Chart 2 shows, those from the wealthiest quartile of households, or with mothers with more than 12 years of education, experience cognitive development in line with international norms (normalized to stay at 100 for all age groups). Children of poor or less-educated households develop skills much more slowly, restricting future opportunities. Such differences in potential are often magnified over life, through the education system and access to work and services.

Chart 2. Opportunities determined early

The massive inequities across countries are sustained by restrictions in international trade and migration that constitute by far the greatest distortions in global markets. The good news is that there has been substantial convergence in some social indicators, notably life expectancy, over the past four decades—with the major exception of the sharp decline in life expectancy in African countries hit by HIV/AIDS in the past decade or so. By contrast, inequality in economic prospects across individuals in the world—as proxied by incomes—experienced a large, long-term rise until the 1980s, driven essentially by differences in country growth rates (see Chart 3). Although there has been a modest decline in this measure since the 1980s, primarily due to the rapid growth of China and India, the gap continues to widen for the slow-growing poorest countries.

Chart 3. Big gap in global incomes

Inequity holds back development

Inequalities in opportunity are reproduced over time and across generations, through economic, sociocultural, and political mechanisms—leading to what we call inequality traps. Even in the United States, often called the "land of opportunity," a recent study finds an intergenerational earnings elasticity of 0.6 (Mazumder, 2005). This implies that a family currently earning half the national average income can expect to take five generations to reach the average.

The same processes that reproduce inequalities can also harm efficiency and overall development. Evidence of that comes in two main categories: microeconomic analyses of the ways in which market imperfections interact with inequalities, and documentation of interactions between political inequalities and institutional formation. Consider the following examples.

Market failures, inequalities, and investment inefficiency. In a world in which markets worked perfectly, investment decisions would have little to do with the income, wealth, or social status of the decision maker. However, for various reasons—mainly economic, but also political—markets are not perfect. In Ghana, for example, land is typically allocated by custom, and security of property rights is thus often linked to the local power structure. Individuals are less likely to leave their land fallow (an investment in long-run productivity of the land) if they do not hold a position of power within either the hierarchy of the village or the hierarchy of the lineage, for fear of having their land reallocated while it is lying fallow (Goldstein and Udry, 2002). Because women rarely hold those positions of power, women's land is not left fallow often or long enough and is much less productive than men's. The resulting decline in land productivity is a pure loss for society.

Another example pertains to the forest-savannah region of Southern Ghana, where many farmers cultivate a cassava-maize intercrop. Recently, pineapple cultivation for export to Europe offered a new opportunity. Survey results in the late 1990s revealed that the profitability of pineapple production dominated that of the traditional intercrop, with average returns associated with switching to pineapple in excess of 1,200 percent! Yet only 190 out of 1,070 surveyed plots were used for pineapple. When farmers were asked why they were not farming pineapple, the virtually unanimous response was: "I don't have the money." The authors conclude that the fixed costs involved in switching crops—and the absence of a well-functioning credit market—prevent a large number of farmers from making a very profitable investment. Output and income levels in these areas are correspondingly below potential (Goldstein and Udry, 1999).

In rural North India, a recent experimental study shows that one's caste can affect individual performance even at a fairly basic level (see Chart 4). In the first experiment, groups composed of low-caste and high-caste junior high school students were asked to solve mazes and were paid based on the number of mazes they solved. In one game, no personal information about the participants was announced. In a second, caste was announced with each participant's name and village. In a third, participants were segregated by caste and then each participant's name, village, and caste were announced in the six-person group.

Chart 4. The costs of caste

When caste was not announced, there was no caste gap in performance. But increasing the salience of caste led to a significant decline in the average performance of the low-caste children, regardless of whether the payment scheme was piece rate (with 1 rupee per maze solved) or tournament (the winner who solved the most mazes received 6 rupees per maze solved; others received nothing). When caste was announced, the low-caste children solved 25 percent fewer mazes on average in the piece-rate treatments, compared with the performance of subjects when caste was not announced. While we do not know what the children were thinking, some combination of loss of self-confidence and expectation of prejudicial treatment likely explains the result. If similar declines in productivity occur in real work situations, the private and social losses would be no less important. This could lead to further underinvestment in the education of lower-caste groups, if it leads to the expectation of lower returns.

Political inequalities and the formation of institutions. This category of evidence concerns the relationship between political inequality and the development of governance institutions. One point of departure is the correlation between measures of the quality of institutions and the level of incomes. While debate on the econometrics continues, an important strand of recent literature argues that there is a significant causal relationship between historically formed institutions and contemporary levels of income.

What lies behind this? Historical analysis supports the view that inequality influences institutions. In particular, extreme political inequalities can lead to the design of institutions that are good at extracting and concentrating rents for elites, but less effective in protecting property rights for all; providing broad-based education, risk-management, and economic infrastructure; or fostering broad, competitive financial and industrial structures.

Take, for example, a study of the early institutions and the long-term development paths of European colonies in North and South America (Engerman and Sokoloff, 2002). The authors found that the abundance of unskilled labor prevalent in the South American colonies—where either native Americans or imported African slaves were available in large numbers—combined with the technology of mining and large plantation agriculture to provide the economic base for hierarchical and extractive societies, in which land ownership and political power were highly concentrated. By contrast, in North America, English colonists tried to impose oligarchic structures and extractive economic institutions but failed to do so; there was neither the natural resource base nor ready supplies of subordinate labor.

Leveling the playing fields

What can policymakers do to provide greater equity? In the domestic arena, they need to invest in people; expand access to justice, land, and infrastructure; and promote fairness in markets. In the international arena, they need to focus on the functioning of global markets and the rules that govern them. Governments must seek to reverse the usual order of things, starting with the poorest and most excluded. They must also recognize that personal freedoms are ultimate goals of the development process, and individual incentives are the chief engines of growth and prosperity. Equity must be pursued through well-functioning, competitive markets, not against them.

Expanding opportunities. In the case of human potential, greater opportunities should be reflected in an array of measures in the social sectors—quality basic education for all, preventive health care, and risk management to deal with shocks associated with the weather, health, and labor incomes. As for markets, the focus should include rural roads for market access, tenurial security for peasants and slum dwellers, and microcredit. In the political arena, it should include the "empowerment" agenda—access to justice, accountability of basic service providers, and local democracy.

For example, there is a considerable body of evidence on successful early childhood development interventions in a number of countries. A controlled experiment in Jamaica compared children that were stunted (a measure of long-term undernutrition) with children of normal height. Stunted children who received nutritional and behavioral interventions were able to effectively eradicate a substantial developmental deficit after 18 months in the program. Undernourished kids outside the program never closed the gap (see Chart 5).

Chart 5. Closing the developmental gap

Reducing privileges. This matters not only because privileges are often the political obverse of restricting opportunities for poor and middle groups but also because of the association between privilege and growth-sapping structures of protection and rents. Privilege is manifest in various arenas of public action. In the social sectors, many countries sustain regressive subsidies for university schooling, pensions, and health services. In the economic arena, protected industrial sectors, concentrated financial systems, and (often) privatizations captured by dominant economic groups are examples—underlining the need to "save capitalism from the capitalists" in the phrase of Raghuram Rajan and Luigi Zingales (2003). And in national political systems, local government, and judicial systems, the influence of the powerful and rich is all too often both unjust and inefficient in its consequences.

A recent study on Latin America found important linkages between inequalities of influence and wealth (De Ferranti and others, 2004). Connected lending can be a source of rapid credit growth and poor asset quality, increasing the vulnerability of financial systems to crisis—and workouts can be highly inequitable. Those with significant financial assets often get their money out before the crash, even experiencing capital gains when exchange rates and asset prices tumble (as in Argentina). And bailouts are biased to insiders in the financial system, especially to larger firms and individuals. The bill is picked up either through higher taxes (typically proportional) or forgone spending (that is often progressive at the margin, implying that lost spending hurts poorer groups most). In terms of policy, this implies building more competitive, open, and accountable financial systems, and institutionalized insurance systems that are less vulnerable to discretionary pressure by those with influence. Such systems would include deposit insurance policies explicitly limited to smaller depositors and employment guarantee schemes, backed by contracyclical fiscal policies.

* * * * *

The bottom line is that a concern with equity is of immense importance for development, both on intrinsic grounds and as the result of a hardheaded analysis of the development process. In fact, equity—defined as equality of opportunity and avoidance of absolute deprivation—is an essential part of the development strategies needed to achieve the Millennium Development Goals by 2015. Policymakers must now find a way to bring equity into the center of the development discourse and policy design.


De Ferranti, David, Guillermo E. Perry, Francisco H.G. Ferreira, and Michael Walton, 2004, Inequality in Latin America: Breaking with History? (Washington: World Bank).

Engerman, Stanley, and Kenneth Sokoloff, 2002, "Factor Endowments, Inequality, and Paths of Development Among New World Economies," Economia: Journal of Latin American and Caribbean Economic Association, Volume 3 (Fall), pp. 41–109.

Goldstein, Markus, and Christopher Udry, 2002, Gender, Land Rights and Agriculture in Ghana (New Haven, Connecticut: Yale University).

——— 1999, Agricultural Innovation and Resource Management in Ghana (New Haven, Connecticut: Yale University).

Mazumder, Bhashkar, 2005, "The Apple Falls Even Closer to the Tree Than We Thought," Chapter 2 in S. Bowles, H. Gintis, and M. Groves, eds., Unequal Chances: Family Background and Economic Success (Princeton: Princeton University Press).

Rajan, Ragurham G., and Luigi Zingales, 2003, Saving Capitalism from the Capitalists: Unleashing the Power of Financial Markets to Create Wealth and Spread Opportunity (New York: Crown Business).

Francisco Ferreira and Michael Walton (currently at Harvard's Kennedy School of Government) codirected the World Development Report 2006: Equity and Development. The team included Abhijit Banerjee, Peter Lanjouw, Tamar Manuelyan-Atinc, Marta Menéndez, Berk Özler, Giovanna Prennushi, Vijayendra Rao, Jim Robinson, and Michael Woolcock.