Determinants of Poverty in Egypt, 1997

Poverty reduction is the ultimate goal of all development policy. In Egypt about 15.7 million people, or about 26.5 percent of the population, are considered to be poor. Of these, 5.1 million are deemed ultrapoor. To reduce poverty, policymakers first need to understand how it is distributed across geographic and socioeconomic groups, what are its characteristics, and what factors determine it. A series of studies, undertaken by IFPRI in conjunction with the Egyptian Ministry of Agriculture and Land Reclamation (MALR) and the Ministry of Trade and Supply (MOTS), provides a profile of poverty, examines the complex interaction among poverty-related variables to identify which policies are most likely to change the level of poverty, and investigates the relationship between food prices and wages.

As part of this series, this study focuses on the factors that determine poverty. Although poverty profiles provide important clues to the underlying determinants of poverty, they do not take into account the complex interaction among multiple poverty- related variables. Therefore, they do not provide a clear enough guide to the effects that policy inter- ventions will have on poverty. This study explicitly examines the factors that determine poverty. The analysis, using data from the 1997 Egypt Integrated Household Survey, a nationwide survey of 2,500 households carried out by IFPRI, MALR, and MOTS from March to May 1997, suggests which policies are most likely to change the level of poverty in Egypt.

This is the first such study of its kind for Egypt. No previous survey has allowed a household-level anal- ysis of living standards based on multiple variables. The multiple-variable analysis provides a clearer guide to the effects policies will have on living standards, measured here by consumption per person.

Study Findings
The study uses urban and rural models adjusted for socioeconomic and agroclimatic conditions in each governorate to identify the factors that determine poverty and predict the changes in poverty that will result from simulated policy changes. The model for the urban sector accounts for interaction effects between schooling characteristics, unemployment, and landownership. The rural model takes into account the interaction of these three variables, as well as community characteristics such as irrigation and distances to railroad. The policy simulations indicate that three areas are important for poverty reduction (as measured by improvements in per capita con- sumption): educational investments, investment in infrastructure, and reduction in unemployment.

Increasing average years of schooling per household and improving parental education hold considerable potential for raising living standards in Egypt.

A key conclusion of the study is that education plays a vital role in alleviating poverty throughout Egypt. Simulations show that a two-year increase in the average level of education attained by Egyptian households results in an 18 percent decline in the number of individuals living in poverty. The depth and severity of poverty decline by about one-fourth, with a larger reduction in urban areas. The results for a one-year increase in average schooling follow the same pattern, but at about half the level of the two- year increase. The beneficial effect of improved school attainment holds true regardless of where it takes place: in the rural or urban sector or in Upper or Lower Egypt. If par- ents complete primary schooling, per capita consumption in Egyp- tian households in- creases by 20 percent, attesting to the intergener- ational importance of human capital. The proportion of the population affected by this change is larger in rural areas.

Investment in infrastructure is another important area of antipoverty policy intervention. (Infrastructure data are only for rural areas.) Distance to railways and the supply of irrigation by themselves are insignificant determinants of poverty, but are significant in interaction with other variables. If years of schooling are taken into account, greater distance to the railway station has a negative effect on living standards, while better supply of irrigation has a positive effect. Clearly, better infrastructure and household education complement each other. Policy simulations show that an increase in the social capital index—which indicates the presence of a secondary school, a hospital, or a police station—to one-half in communities where it is less than half, lowers the ratio of people in poverty by 3 percent. Reducing the distance to the nearest railroad to 4 kilometers would also lower the incidence of poverty by 3 percent. But if considered just for those households more than 4 kilometers away from a railroad station, the change in distance would reduce the proportion of people in poverty by 8 percent. If communities have access to irrigation that they consider sufficient, the proportion of rural people living in poverty would decrease by 10 percent overall, or by 16 percent if only the affected communities are considered. The overall depth and severity of poverty would decrease by 12 and 13 percent, respectively.

Other results suggest that the overall level of unemployment matters as well, of course, though the negative effects of female unemployment are larger than those for men. Lowering unemployment by one person per household reduces the proportion of the population living in poverty by 2 percent in rural areas, and 3 percent in urban areas. Improvement in social capital decreases the incidence of poverty, but by a smaller amount than does reduction of unemployment. The proportion of the rural population living in poverty would decline by only 1.5 percent if cultivable landholdings were increased by 20 percent. The inci- dence of poverty would decline by only 1 percent if the value of livestock holdings were increased by the same amount. But when considering only the affected population in rural Egypt, increased landholdings would decrease the proportion of poverty by 7 percent.

Perhaps the most disappointing simulations are for the two that increase the number of bakeries and flour warehouses per capita. Such investments have virtually no effect on poverty.

Policy Implications
Estimates of the effects that policy changes can have on poverty show that education plays the largest role in diminishing poverty. Raising educational levels lowers poverty by almost twice as much as other interventions. Increasing average years of schooling per household and improving parental education hold considerable potential for raising living standards in Egypt. The results also clearly endorse the importance of investment in rural infrastructure.

In interpreting these results, one should not assume that the effects on poverty will be instantaneous. Educational investments, for instance, have inherently long gestation periods, but they can be powerful instruments for long-term poverty reduction.

To order a discussion paper, please fill out an online order form, email IFPRI-FCN@cgiar.org or send requests to Food Consumption and Nutrition Division, IFPRI, 2033 K Street, N.W., Washington, D.C. 20006, U.S.A.