The wealth and gender distribution of rural services in Ethiopia

Abstract

Over the past several years, the Ethiopian government has committed a substantial portion of the public budget to expanding public services and infrastructure in rural areas. This paper assesses who exactly is benefiting from this public spending. To do so, this paper undertakes a public expenditure benefit incidence analysis across gender and wealth groups of three public services/programs in rural Ethiopia: (1) selected components of the Food Security Program (FSP), (2) drinking water supply, and (3) agricultural extension services. The analysis uses data at the individual, household, kebele (a subdistrict administrative unit), and district level in Ethiopia. The literature on the benefit incidence of services in developing countries exclusively focuses on the education and health sectors, whereas its application to agricultural and other rural services is nearly wholly absent—a gap that this paper seeks to begin to fill. For the selected components of the FSP, the paper finds the average incidence of participation to be pro-poor, both in concentration curve analysis and quantile-based public spending incidence. However, examination of the value of cash and in-kind receipts from the programs finds the cash/food-for-work program to be progressive, whereas the direct support (unconditional transfers to households) tends to be nonprogressive. The incidence of water services is assessed using different measures of access: physical proximity to drinking water sources and the use of improved drinking water sources. Access, as proxied by physical proximity, is poverty neutral, whereas the use of improved water facilities is pro-poor. With regard to agricultural extension, concentration curve analysis finds the service to be relatively progressive, whereas the benefits-to-population ratio demonstrates a somewhat more differentiated picture, with nonprogressive features at both ends of the wealth spectrum. From a gender perspective, the incidence of agricultural extension is pronouncedly skewed in favor of men. The public works component of the FSP favors male-headed households, and the direct support component favors female-headed households. In the case of drinking water services, the incidence of safe water use is actually higher for female-headed households, raising considerations of how male and female heads may differentially prioritize safe water for consumption. As a complement to the benefit incidence analysis, regression results identify demand- and supply-side factors that are correlated with access to the three different services.