Two decades ago economists at the World Bank, under the leadership of the late Hollis Chenery, then the Bank's chief economist, spearheaded a near-revolutionary drive to reorient development policy from a single-minded emphasis on growth to an effort to see that the benefits of growth are distributed more equitably throughout a society (box 1). Concerned by persistent poverty even in rapidly developing countries, they proposed strategies for redistributing income through programs to improve the lives of the urban and rural poor. The Bank did not abandon the drive for worldwide economic growth. Its economists strongly emphasized that no efforts to reduce poverty could have a lasting impact if they were not consistent with a program of sound national growth. But they did set out to convince the development community that poverty must be attacked directly---that the gains from growth are not automatically distributed to all.
The concept---let alone the need for parallel and interlocked efforts aimed at promoting growth and redistributing income---was not immediately accepted. But over the years a broad consensus has been reached in the development community on most of the central points. At the same time the discussion has become deeper and more sophisticated. In the early 1970s, for example, the usual method of measuring poverty was a simple head count measure that used an arbitrary poverty line based on expenditure or income. Recent measures, such as the poverty gap index, are more sensitive to the depth of poverty and to inequality among the poor. Development economists now recognize the importance of using multifaceted measures of poverty that assess the quality of life. Such approaches have limits and can become arbitrary judgments of what constitutes the "good life." But the attempt to measure something broader than income has focused the attention of policymakers on the need for programs to improve longevity, literacy, basic health, and sanitation (box 2).
The past two decades have also brought to light the problems of special groups, particularly women, for whom escape from poverty is especially difficult. Formidable legal and cultural barriers can hinder the entry of women into the labor market. After entry they are often segregated into casual or dead-end, low-paying jobs. At home the psychological burden of culturally induced low self-esteem, added to the physical burden of domestic chores, inhibit their mobility and block opportunities for gaining better work. These conditions have important implications for antipoverty policy. Contrary to the implications of the economist's standard unitary household model, unequal bargaining power within the household can result in underinvestment in human capital for women. Public interventions targeting poor households can therefore be inadequate. Gender-targeted policies might be far more effective.
Gender equity and increased efficiency are likely to go together: a better education for women is often directly associated with positive gains in child education, nutrition, and health and thus with long-term benefits for the economy. At the same time better opportunities for outside work for young women can raise the average age of marriage. As a consequence, women bear fewer children and raise them in better conditions.
Targeting women as part of an effort to reduce poverty in a growing economy inevitably raises the controversial issue of fertility control. Those who oppose fertility control policies note that a large family may be a voluntary choice, a way of getting extra help or providing for old age. Others point out, however, that the family that decides to have the extra child imposes growth-inhibiting costs and burdens on society as a whole: more pressure on common property resources, more congestion of such services as education, health, and sanitation, more crowding of the labor market. Here again, the tie between growth and antipoverty programs is now more obvious than it was 20 years ago.
There is also much more awareness today of the two-way relationship that exists between poverty and environmental degradation. The daily livelihood of the poor, particularly in rural areas, depends vitally on the local environmental resources---forests, irrigation water, and fisheries. The local commons also provide elementary insurance for poor peasants---a fallback source of food and fodder in bad years. With the erosion of grazing lands, the decimation of forests, and the silting and increasing toxicity of rivers, the life of the rural poor, particularly in many parts of Africa and South Asia, has become even more insecure and impoverished. As the environment degrades and poverty intensifies, desperate people accelerate the process with even more intensive use, straining the already fragile and limited environmental base beyond its capacity for repair and renewal.
While recognizing the frequent need for poverty programs, recent experience also demonstrates that policies fostering economic growth facilitate poverty reduction. In many countries, particularly in Asia, growth-promoting policies have brought about substantial reductions in poverty. Economic growth helps, for example, by expanding opportunities for productive and remunerative employment, even if it is no more than self-employment on farms and in artisan shops. Policies that contribute to growth by reducing distortions in relative prices, exchange rates, and trade policies can indirectly benefit all levels of a society. The effects of growth on income distribution depend, of course, on a wide variety of factors, ranging from the initial distribution of assets to the governmentÕs policies toward international competition.
But some types of growth clearly do not help the poor. For example, capital-intensive or skill-intensive industrialization and commercialization projects can delay the decline in poverty among the unskilled. The centripetal forces of growth can drain resources away from backward regions, increasing poverty for those left behind. And large development projects can so damage the environment that they uproot groups of the poor and cut off their access to common property resources. At the same time all societies contain the unemployable---the physically disabled, the elderly, the ill, and women overburdened by reproduction and childcare. Here the solutions are social safety nets (such as welfare payments and food stamps), basic health care, adequate facilities for education, and the provision of low-cost shelter.
While economic researchers and policymakers now generally agree that growth and redistribution must go hand in hand, some approaches and policies remain hotly contested. Land reform is probably the most controversial. Most major land reforms have taken place in the context of war or social revolution. Even then, as in Bolivia and Mexico, they have not always achieved much reduction in poverty. Moreover, even when nonconfiscatory land redistribution is politically feasible and is carried out in normal, nonturbulent times, the economic effects have been ambiguous.
In traditional agriculture, where the use of farm machinery is limited, economies of scale are not necessarily substantial. In that case the small family-operated farm created in a land reform program can have a labor cost advantage and be more productive than the large farm. But owners of large farms are likely to have better access to credit, insurance, mechanisms to diffuse risk, and information and marketing networks. If these services are not available to the small farmer, land redistribution might not always help. A redistribution from large to small farmers can also reduce demand for hired labor, depressing wages or driving into poverty those who lose their jobs and do not gain their own land. Elsewhere, land reform policies that restrict the use of tenant farmers have the paradoxical effect of closing off opportunities for agricultural laborers at the bottom of the heap. And, finally, in some densely populated poor countries such as Bangladesh, imposing a reasonable ceiling on farm size wouldn't free up enough land for redistribution to make much of a dent in rural poverty.
Another area in which opinion and attitude still diverge is in the basic approach to poverty reduction itself. One concept, popular in some international lending agencies and donor countries, is to rely primarily on market-based growth and then take care of those who fall through the cracks of the market with targeted public welfare programs. Countering this is an approach that uses massive public programs to improve the health, education, and nutrition of the poor. A third way moves away from both state paternalism and indiscriminate market forces, relying more on local self-governing institutions and community participation to improve the material conditions as well as the autonomy of the poor.
Shifting responsibility to the local community can be an effective way of improving the lives of the poor. Community-level institutions can play an important role in coordinating projects to conserve natural resources vital to the local economy. Their access to local information can help in identifying more efficient ways of providing basic services. And when decisionmaking responsibility lies with these institutions, accountability and performance in the delivery of local services may improve.
But local institutions may fail if the necessary conditions are lacking. Where social and economic inequality is high, for example, the local elite may find it easy to "capture" local institutions, while the poor remain too weak and fragmented to put up an organized fight for their rights. Local institutions may be unable or unwilling to cope with needs, such as infrastructure or flood control efforts, that span communities. And community-level organizations are likely to be constrained by inadequate administrative and revenue-raising capacity.
The limitations of the state---its lack of access to local information, its vulnerability to waste and corruption, and the lack of local accountability---need to be recognized. But in many countries the limitations at the local level mean that the state must continue to play an activist role in mobilizing local groups and then providing supra-local support in such forms as finance, training, coordination, and region-spanning infrastructure.
After years of research, discussion, and practice, issues remain to be debated. Questions are still asked about tradeoffs between distribution and growth---and are countered with examples of poverty programs that enhance the efficiency of a society's production. And in the discussion on reducing poverty, economists still differ in where they think the focus should be---whether on promoting faster economic growth, allocating more state resources to welfare programs, or building local organizations that are more sensitive to the needs of the poor. Yet there are many points of agreement. And there is a broad consensus that distributional objectives are consistent with growth---just as the Bank's economists envisaged some 20 years ago.
Drawn from Pranab Bardhan, "Research on Poverty and Development Twenty Years after Redistribution with Growth," and Albert Fishlow, "Inequality, Poverty, and Growth: Where Do We Stand?" papers presented at the seventh Annual Bank Conference on Development Economics.