Michael Lipton and Alexander Shakow
The World Bank has always emphasized economic growth and efficiency. In the 1970s, however, as part of a growing worldwide recognition that growth was not reaching the poor majority in many developing societies, the Bank in addition placed special emphasis on direct measures to alleviate poverty, in both its lending operations and its economic research and analysis. An internal review of the Bank’s experience with this approach has recently been completed (see box). This article summarizes the findings on three main questions: What changes have occurred in Bank activities as a result of the effort to reduce poverty? Have projects supported by the Bank been cost-effective in benefiting large numbers of poor people, in ways consistent with efficient growth? How should policies on poverty alleviation respond to changing pressures in the 1980s?
Changes in activities
During the 1970s, it was World Bank policy to use its funds to raise the productivity and living standards of the poor. The International Bank for Reconstruction and Development (World Bank) and the International Development Association (IDA) commitments to the countries where most poor people live—those with per capita annual incomes below US$680 in 1979—increased from 37 per cent of the total program through 1968 to 58 per cent in the 1979–83 program. IDA now lends only to this group (see Table 1).
|Income group||Through 1968||1969–73||1974–78||1979–83|
|Per capita income up to US$390 in 1979:|
|Per capita income of US$391–680 in 1979:|
|Total of both income groups:|
The Bank has also increased its lending for sectors and subsectors widely considered to offer the most direct benefits to the poor—notably rural development, primary education, small-scale industry, water supply and waste management, and population, health, and nutrition. These sectors’ share of Bank/IDA lending increased from 5 per cent in 1968–70 to 30 per cent by 1979–81 (see Table 2). Within sectors there were also changes: rural development loans (over half of whose benefits reach the rural poor) increased from 21 per cent of all agricultural lending in 1969–73 to 53 per cent in 1978–81. Bank research and analysis, by the latter part of the 1970s, was increasingly concentrated on poverty issues. Meanwhile, methods to measure the extent of poverty and the impact of development on the poor, while still crude and in need of improvement, became an important tool in helping to focus the lending program on direct poverty alleviation.
|Directly poverty-oriented projects|
|Education—primary and nonformal||0.2||0.8||1.4||1.6|
|Population, health, and nutrition||0.0||0.7||0.6||0.8|
|Energy—oil, gas, and coal||0.3||0.9||0.9||3.5|
|Industry (including development finance companies)||15.7||16.4||16.6||13.8|
|Total lending, in millions||US$1,641.3||$2,959.7||$7,369.9||$11,261.1|
A child finishes her meal in a small Peruvian village.
The Bank’s program, however, finances less than 2 per cent of total investment in the developing countries, and individual projects normally affect only a minority of a country’s poor people. Government policies, on the other hand, have an impact on vast numbers, and on the success of projects. This makes the Bank’s economic dialogue about poverty with borrower governments especially significant. Experience suggests that the Bank has been more successful at incorporating an emphasis on poverty in projects than in policy dialogue.
Impact of new policies
The evidence indicates that projects with particular emphasis on poverty have benefited large numbers of poor people and have had good economic rates of return. There are, of course, unsuccessful projects; but the proportion is small and no higher for poverty-focused projects than for others.
The Bank’s independent Operations Evaluation Department (OED) also reports that poverty projects have similar appraisal times, implementation delays and problems, and cost overruns. They also absorb similar amounts of Bank supervision time per project or per beneficiary—although they have taken more time per dollar loaned, as early new-style projects were often relatively small, pilot efforts. In general, the Bank has benefited from extensive “learning by doing”—from both successes and failures—which should help efforts to reduce poverty and stimulate growth in the 1980s.
In September 1981 the World Bank established a task force of senior staff to review the Bank’s approach to poverty alleviation. The group included Leif Christoffersen, Anthony Churchill, Aklilu Habte, Mahbub ul Haq, Basil Kavalsky, Herman van der Tak, Bevan Waide, and Hollis Chenery, Chairman. The task force was assisted by Norman Hicks, Michael Lipton, and Alexander Shakow.
Bank management and the Executive Directors at a Board seminar have generally endorsed the conclusions of the task force report. In a recent speech in Lagos, Nigeria, Bank President A. W. Clausen emphasized that “a key and central aim for the Bank is the alleviation of poverty. Our objective in any developing country—anywhere in the world—is precisely the same: to assist the country both to accelerate its economic growth and to reduce its level of domestic poverty by enhancing the productivity of its poor, and thus making possible a better standard of living for all its people.”
Agriculture and rural development are central to poverty alleviation. In developing countries, some 70 per cent of the total number of poor depend on farm incomes, and a similar percentage of poor people’s income is spent on food; yet typically rural productivity—and incomes-per-head—are only 30–40 per cent of urban levels. Lending for rural projects increased dramatically in the 1970s—from $2.6 billion in 1969–73 to over $13 billion in 1978–81. A rising share of this total supported the poorest countries and/or crops most likely to be grown or eaten by the poor.
Projects approved since 1974 were thought, at the time of appraisal, to be likely to benefit about 125 million poor people. Recent OED audits of completed projects suggest this estimate may be exceeded. As for the cost, rural development projects audited in 1979 benefited 660 small farmers for every $1 million loaned (at a cost of $1,500 per farmer), compared with 47 farmers per $1 million (at $21,000 per farmer) in other agricultural projects.
Emphasis on small farmers has seldom conflicted with economic and financial efficiency; audits show that average rates of return on poverty-oriented rural development projects were at least as good as those on other agricultural projects. Small farmers also appear able to obtain higher returns on investment and to default on loans less often. This confirms other evidence that small farms usually achieve greater annual employment and output per acre than larger farms.
Some difficult problems remain. First, low-risk technical “packages” appropriate for poor farmers in semi-arid rainfed areas, particularly in Africa, are not readily available. In irrigated areas, inadequate or corrupt water delivery practices often deny water to poor and downstream farmers.
Second, the Bank’s rural development strategy seeks mainly to raise the production of small farmers. Direct benefits have been small for other rural poor—the landless and the marginal farmer who finds it hard to borrow, acquire inputs, and take risks; some projects may have inadvertently made the landless worse off by introducing machines that replaced labor. The poorest 20 per cent have seldom gained much from rural development projects. This suggests the need for more emphasis on various aspects of land or tenancy reform, rural works projects, and improvement of labor markets and worker mobility.
Third, domestic pricing and postharvest policies often undermine the success of projects aimed at the rural poor. Inadequate price-incentives have jeopardized several small farm projects. Large-scale storage projects, including some with very low returns, too often shift grain from efficient small farm stores to less efficient centralized systems. These are complex issues because consumers gain when food prices are low, but price policy analysis and marketing are weaknesses in some of the Bank’s rural development projects.
Fourth, success in rural development often rests on sociological and cultural factors. These are difficult areas that deserve more attention.
Urban poverty projects. It is harder to affect poverty by increasing the productivity of the poor in towns, where investments in industry, power, and transport are often efficient only on a large scale, and not especially focused on either employment or poverty. Recognizing that the urban poor often cannot afford housing and services, the Bank has, since 1972, strongly endorsed an innovative approach to provide “sites and services” instead of structures. This has evolved to include upgrading squatter settlements, which reached far more poor people per dollar, and avoided ineffective and unpopular slum clearance by bulldozers.
These urban programs involved secure land tenure, affordable design, consumer payment for services to enhance replicability, and self-help labor. Bank urban programs have also come to include low-cost transport and city-wide investment programs. They have strengthened local institutions and gained acceptance for new approaches for national programs to increase the efficiency of cities. Government policies toward the provision of urban services have improved in some 35 countries as a result of Bank assistance, which has demonstrated that it is possible to provide affordable services to the poor.
Since 1972, 52 Bank projects centered on urban shelter have involved $1.6 billion of Bank/IDA funding and $2 billion of borrower government and other financing. Appraisals anticipated some 11 million beneficiaries—with over 55 per cent of project costs benefiting the poor—and real economic returns averaging 20-25 per cent. Evaluations of completed projects are few, but they too suggest high returns, good cost recovery (66–95 per cent), and success in reaching the poor. Bank projects initially concentrated on owner-built housing, but now increasingly encourage owners to hire construction labor and rent out rooms, helping to spread benefits to those too poor to own even low-cost housing.
It is evident that better urban shelter is needed—and that the poor are frequently willing to pay for it. A central problem remains: how, once decently housed, can the urban poor use these programs to expand income from work? Greater emphasis on small-scale, labor-intensive enterprises may well be required. The smallest, in particular, can be an effective and economical way to increase productive employment. With low cost per job as an important selection criterion, Bank/IDA loans to small firms have grown from almost none before 1974 to 2 per cent of lending by 1980–81. Based on problems experienced in early projects, the Bank has shifted toward more ample, unsubsidized credit, smaller enterprises, and recommending the elimination of policies favoring large, excessively capital-intensive enterprises.
Another area important to poverty alleviation is water supply and waste management, which now receive over 6 per cent of Bank lending. Due to the use of simpler, innovative technologies and pricing procedures, these projects reach very large numbers of mostly poor beneficiaries—although, by value, most benefits go to the urban nonpoor, who create the vigorous pressure that attracts public funding. In this sector, as with small enterprises, very few Bank projects have thus far addressed rural needs.
Social sectors. Research and experience provide increasingly strong evidence that gains in productivity from investments in primary education, family planning, basic health care, and nutrition make them worthwhile on economic, as well as humanitarian, grounds.
About 5 per cent of Bank lending is for education, some 40 per cent of it financing primary education (compared with near zero in 1969). Primary education generally shows higher economic returns than other forms of education. Since the better-off are usually already in primary school, such expansion benefits mainly the poor, but the rapid pace may sometimes sacrifice quality; recent projects, notably in textbook distribution, attack this problem. The timing, location, and cost of schooling greatly affect poor families’ demand for it; prospects for some cost recovery, along with general improvements in revenue, influence governments’ readiness to supply it. The Bank’s work on project design in these matters has helped increase school attendance among poor families. Recurrent cost control will be especially important as budgetary constraints become even tighter for developing countries in the 1980s.
Despite the importance of population, health, and nutrition to broadly based poverty alleviation, these areas still absorb less than 1 per cent of the Bank’s total lending program. The alleviation of poverty is closely linked to reductions in population growth rates. Several countries have made encouraging progress through a combination of more effective economic development programs and increased access to family planning services, but in most only a small percentage of fertile couples practice family planning. In Africa, few countries have even rudimentary programs; moreover, the growth of food production has fallen behind increases, in population. Only $400 million in population loans were made to 13 countries in the 1970s, reflecting the fact that the Bank has not found this an easy area in which to work. The Bank’s investments in agriculture, education, and other sectors can also strengthen the role of women in development and help reduce fertility, but these indirect approaches will not fund the increased access to services that is also necessary. Given the large size of most poor families, returns to investment in preventing births are very high, with major gains for the poorest.
In the 1970s, the Bank’s lending for health was a component of projects in other sectors; only recently have separate health projects been started. These projects and associated policy and institution-building activities will be an important part of future programs, given the demonstrated contribution of improved health to labor productivity and the well-being of the poor, particularly with the Bank’s emphasis on primary health care and preventive measures.
Multisectoral nutrition projects have been hard to administer. However, persistent Bank project and policy work has led to some replication and has raised the priority of nutrition in the national policies of several countries.
Impact on growth. The Bank’s strategy in the 1970s supported many projects that benefited the poor, both directly and through some local replication. Poverty-alleviating agricultural and urban projects showed good returns; much investment in human capital has paid off as well.
The macroeconomic evidence also suggests large areas of complementarity between growth and poverty alleviation. However, the wide variety of country experience provides little guidance on the ideal balance to achieve both objectives. At the extremes there is surely a trade-off. Massive social expenditures can undercut investment, and thus jeopardize sustainable economic growth; unlimited emphasis on growth can leave too few resources to improve poor people’s living standards. In general, countries that have placed special emphasis on poverty alleviation have also been able to achieve average or somewhat above average growth rates.
Emphasis for the 1980s
In some countries the incidence of poverty fell sharply in the 1970s, and poverty alleviation policies can take some of the credit. But massive poverty persists. Further improvements based on the lessons of the past are needed, especially because the difficult circumstances of the 1980s present the poor with greater problems and increased risks.
The “oil price shocks” of the 1970s (real oil prices, despite the recent decline, remain six to seven times higher than in 1972), slower world growth, and the increasing costs of financing development compound the difficulties of alleviating world poverty. Developing countries need to improve their utilization of energy and to achieve the structural adjustment of their economies in an environment where protectionist trade measures and scarcer and/ or costlier inflows of capital threaten the resources and concern previously devoted to poverty questions.
These circumstances heighten pressures to turn away from balanced development programs where growth and poverty alleviation are pursued together. Resource constraints create a crisis atmosphere where budgetary choices tend to stress the reduction of current-account and public-sector deficits at the cost of longer-term investments in, say, agriculture and education. In these circumstances, the real trade-off may be between short-term balance and both poverty alleviation and growth, with poor people in the poorest countries suffering lasting damage. Increased flows of external assistance for these countries can help cushion the worst effects of a painful short-term trade-off, but prospects for substantial additional concessional resources in the near term are not encouraging.
There are many ways in which the alleviation of poverty can be addressed and individual decisions on projects and programs will need to be made in the context of each government’s circumstances and policies. The Bank, however, should seek to minimize the damage to poverty programs, and should encourage the efficient use of resources to protect and enhance the productivity and living standards of the poor.
The Bank can work with individual governments to develop specific programs and policies in which long-term growth and poverty alleviation are complementary. More attention should be paid to the impact of country and sectoral policy options on these dual objectives. Structural adjustment programs should consider, as far as possible, how the burdens of adjustment are shared among income groups, and how adverse effects on the poor—for example through increased unemployment, higher prices of basic goods, or higher taxes—can be mitigated. Research should give high priority to reducing the gaps in our knowledge of the effects on the poor of macropolicies, including alternative routes toward structural adjustment.
Declining IDA resources may unavoidably cut total lending to the poorest countries. However, as experience shows that projects in rural development, urban shelter, primary education, and health produce high returns, a strong argument can be made that scarcer IDA resources should, wherever possible, be even more firmly directed toward projects meeting both efficiency and anti-poverty objectives. Wherever feasible, poverty and employment aspects should also be considered in the design of projects in the traditional sectors, for example, power, transport, and industry. It is misleading to suggest that only a handful of sectors can benefit the poor.
It has proven extremely difficult to benefit people without productive assets—for example, the rural landless, the urban jobless, adult illiterates, or female-headed households. In addition to continued support for education, population, and health programs, greater efforts should be made to understand how the productivity and employment of such people can be increased, including further experiments in designing projects that benefit the poorest families.
Better analysis of the social environment of a project, assessment of its likely social impact, and increased technical assistance are important to lasting and successful development activities. Projects are more likely to benefit the poor to the extent they improve administration and management, and build a strong institutional base that will endure after the project is completed.
A balanced strategy of growth combined with poverty alleviation provides the best general framework for development in the 1980s. There is no panacea in this approach, nor will it lead to precise answers for each country’s specific needs. No development strategy does. What it does is to provide a viable and sensible starting point for making critical decisions. The Bank’s task is to build on the experience of the 1970s and to strengthen the approach that now underlies its program.
RobertAlbert Berry and WilliamR. ClineAgrarian Structure and Productivity in Developing Countries (The Johns Hopkins University PressBaltimore, MD1979).
LeifChristoffersen “The Bank and Rural Poverty,” Edward Jaycox, “The Bank and Urban Poverty,” and Abdun Noor, “Education Lending for the Poor” in RobertS. McNamara and others. The World Bank and the World’s Poorest (World BankWashington, DC1980).
NormanHicks “Growth vs. Basic Needs: Is There a Trade-off?” World Development (1979).
MichaelLipton “Agricultural Finance and Rural Credit in Poor Countries” in PaulStreeten and RichardJollyeditorsRecent Issues in World Development (Pergamon Press, Inc.Elmsford, NY1981).
Operations Evaluation Department World BankSeventh Annual Review of Project Performance Audit Results Report No. 3640 (1981).
GeorgePsacharopoulos “Returns to Education” inTimothyKingeditorEducation and Income World Bank Staff Working Paper No. 402 (1980).
Urban Development Department World BankRetrospective Review of Urban Operations (forthcoming).