Click here for search results

Kenya: Poverty Assessment


Kenya FY95 PA

Read the Full Text (14Mb PDF)

Poverty Profile

About 47 percent of the rural population is estimated to have been below the poverty line in both 1981-82 and 1992, meaning that there was no significant improvement in the incidence of poverty between those two years. About half of the population was, therefore, unable to consume a minimum requirement of food and essential non-food commodities. Further, the shortfall in consumption (the difference between the minimum required consumption and the actual consumption) increased and inequality worsened; the Gini coefficient increased from 0.4 to 0.49. In urban areas, the incidence of poverty was 30 percent in 1992. The urban unemployment rate doubled to reach 22 percent in 1992.

The majority of the poor in rural areas are food and subsistence farmers and those who derive the bulk of their income from the informal sector; the urban poor are mainly in the informal sector. About a third of rural households are female-headed, and two-thirds of them have no male support, in other words, are headed by widowed, divorced, or separated women with children. The incidence of severe poverty is significantly higher among such households—44 percent compared to 20 percent for male-headed households. A major cause of poverty among divorced and separated women is the loss of access to land. Such women often flee to urban areas where they remain poor.

Kenya has distinctly more favorable social indicators than most countries in the region. However, the distribution of the benefits from better health and education is uneven and is related to income. The bottom expenditure decile has a net primary school enrollment rate of only 62 percent, compared to 82 percent for the top decile. The inequality is worse for secondary education; the bottom decile has an enrollment rate of only 2 percent compared to 20 percent for the top decile. Child status indicators vary by the education of the mother, a partial proxy for income. For example, the rate of stunting is 21 percent among children with mothers who have the secondary education in contrast to 37 percent for the children of mothers who have no education. Infant and child mortality rates are about 50 percent higher for children born to uneducated mothers.

There is also a strong urban bias. Rural secondary school enrollment rates are half those of urban areas; the rate of stunting is 60 percent higher and infant mortality is 43 percent higher. There is no evidence of gross discrimination against females in human capital investment. Nevertheless, the school completion rate is lower for females, and, in times of economic stress, female students are more likely to drop out of primary school. Girls from poor families in rural areas typically do not attend secondary school at all.

Incentive and Regulatory Framework

The price incentives for private sector activities are reasonably undistorted. In agriculture, Kenya is one of the few countries that has not taxed its farmers heavily at any time, and producer prices reflect world prices. As part of its structural adjustment program, Kenya has reformed its exchange rate management and trade. The exchange rate is now determined by auction and the real effective exchange rate has been reduced. Real wages in both the private and public sectors are flexible and have fallen, reflecting market conditions.

The distortions in the Kenyan economy arise instead from the over-extended public sector. Inefficient parastatals, civil service wages, and interest payments make large demands on the budget, resulting in the neglect of operational and maintenance expenditures. Consequently, many essential public services of particular importance to the poor such as basic health and education services and rural infrastructure suffer from a lack of adequate recurrent funds. Another set of distortions arise from the protection given to the large farm sector in the form of lower tariffs and taxes on farm machinery and subsidized credit. These preserve a highly capital-intensive and inefficient mode of production that uses very little labor.

Incidence of Public Expenditures in Social Sectors

Real per capita public expenditure on primary schooling was held constant during the 1980s and early 1990s and succeeded in reaching poor households who received a greater share of public subsidies than the nonpoor.

Private spending on primary schooling amounts to one-third of total expenditure, well above the share in many low-income countries. Thus, by a combination of public and private spending, nearly universal primary school enrollment was maintained. However, the poor spend less than half the amount the nonpoor do on primary education; since the government's contribution is mainly in the form of teachers, this inequality in private spending means that there are fewer non-teacher inputs in schools attended by the poor and, therefore, a lower quality of education. For the bottom 10 percent of the population, the private cost is altogether too high, resulting in significantly lower enrollment. Parents have to spend nearly 10 times as much on secondary education as they do on primary education, and, therefore, the enrollment of poor children is negligible.

Public subsidies go almost entirely to the nonpoor, especially in the urban areas. The most inequitable aspect of public expenditures on education is the high share devoted to university education. Although public universities account for only 0.6 percent of Kenya's student population, they absorb 19 percent of the government's recurrent spending on education, and the per student public expenditure is 30 times greater at the university level than at the primary level.

Although per capita public health expenditure declined over the 1980s and the early 1990s, the share of expenditures on preventive health doubled to 20 percent. The continued improvements in life expectancy and in infant and child mortality despite little growth in per capita income suggests that the relative protection given to preventive health services has been helpful.

However, the large differences between rural and urban areas in these indicators suggests that the poor in rural areas have not received the same attention. Curative expenditures form the bulk of the public budget, and these in turn are heavily biased towards hospital-based care in those hospitals that are less frequently used by the poor and away from health centers that are used by the poor three times more often than the nonpoor. As a result, the poor receive a less than proportionate share of government subsidies in curative health. The poor have much less access to safe water; in urban slums, the poor buy water at two to three times the price that better-off households pay through individual connections.

Food security interventions are ineffective and inadequate. The open market price of the staple grain was highly unstable during the period examined, and drought management was inadequate. At the same time, the grain parastatal received large subsidies. Frequent use of import controls had the effect of unduly raising the price of food for consumers, more than a third of whom were unable to satisfy the minimum caloric requirement. Public spending on child nutrition was extremely low, about a third of the deficit of the grain parastatal. The bulk of it was spent on the costly and poorly targeted school milk program, to the neglect of other programs such as nutritional education and targeted interventions for infants and pre-schoolers which are known to significantly improve child health and performance.

Targeted Programs

Three targeted programs were examined:

  1. The rural roads maintenance program;
  2. The micro-credit programs; and
  3. The arid lands development program.

The labor-intensive minor roads maintenance program was found to reach the poor including landless laborers and female-headed households. Yet the wage paid was higher than the reservation wage of the poor and, as a result, the scheme attracted many nonpoor workers.

The micro-credit programs are run by NGOs and follow group lending practices adapted from those pioneered by the Grameen Bank. Partly because they are new and not as large as the Grameen Bank, their costs are higher, and, in addition, the loan size is larger relative to GNP. They reach fewer women than the Grameen Bank despite the fact that there are 23,000 women's groups, many of which are engaged in income-generating activities.

The arid lands of Kenya contain a large nomadic population, most of whom are poor and isolated. Bilateral and NGO projects, although small in scale, have had a favorable effect on meeting basic needs and improving the health of their livestock but less impact on increasing income and on integrating themselves with the rest of the economy.

Poverty Strategy

The lack of sustained per capita income growth is the primary cause of continued poverty in Kenya. Achieving sustained growth in per capita income should, therefore, be the government's top priority.

The fastest growth is likely to come in the manufacturing sector, based on a reorientation towards exports. The government should continue import and exchange rate liberalization and export promotion policies, and support these with macroeconomic stability. However, manufacturing accounts for only a fraction of the economy; broad rural development is, therefore, of overwhelming importance for food security and poverty alleviation.

Raising smallholder yields, particularly in less well-endowed areas, improving and extending research and extension, improving and liberalizing the dairy and seed industries, and maintaining the road infrastructure (through the Roads 2000 strategy) are all critical for broad-based growth and diversification in rural areas. In addition, all subsidies to the large farm sector, including the price and marketing support provided by the grain parastatal, should be removed so that the sector becomes more efficient (and therefore labor absorbing) or the land is sold to the many small peasants in the active land market that exists in Kenya.

In the health sector, the share of preventive health in total health expenditures should be increased with an expanded program of child immunization targeted at the poorer districts and poor families. Within the curative budget, spending must shift towards health centers and away from hospitals. Although cost recovery is providing a significant share of expenditures in health centers, this must go hand in hand with a reallocation of public expenditures towards those facilities most used by the poor.

In education, a targeted mechanism to reduce the private costs of primary education in the poor areas and for poor students (for example, a system of bursaries for the poor, particularly female students, administered by community and local authorities) is recommended. The government should conduct a study to examine the reasons for the high costs of secondary education and the determinants of low enrollment with a view to finding solutions to the problem. In the meantime, the system of giving secondary school bursaries to poor and female students in rural areas should be continued. In the area of food security and nutrition, the main recommendation is to remove the present high-cost and ineffective interventions and subsidies—the large subsidy to the grain parastatal and the school milk scheme—and concentrate resources on nutrition and child development programs that mainly target infants and pre-school children.

A number of special initiatives and targeted programs are recommended: a pilot Rural Water Fund to finance communities and NGOs directly to construct small water schemes, to provide more water kiosks in urban slums, and to support community-based initiatives in waste-removal and sanitation; an expansion of the labor-intensive minor roads program; and an expansion of credit schemes with greater targeting to women and women's groups. In arid areas, road infrastructure should be selectively improved and community-based pilot income-earning schemes should be tried out.




Permanent URL for this page: http://go.worldbank.org/21IW196EZ0