Effects of Schooling on Female Genital Cutting: The Case of Kenya

Abstract

There is not much conclusive evidence on conditions for female genital cutting to decline. Relying on data from the Kenya Demographic and Health Surveys, I study whether increased educational attainment for a sample of women born between 1950 and 1980 has had any effect on the probability that the genitals of their eldest daughter have been cut. In order to mitigate the problem of omitted variable bias, I utilize an exogenous change in years of schooling as a result of the 1985 restructuring of the education system. According to the two-stages least-squares results, receiving an additional year of schooling decreases the probability that the eldest daughter of respondents complying with the reform was cut. This change accounts for 11 % of the sample mean. Because educational attainment is measured with some noise, I focus on the intention-to-treat effect by estimating a reduced-form model. On average, the reform led to a decrease in the probability that the eldest daughter of respondents was cut. This compares to 16 % of the sample mean. I demonstrate that the results might be consistent with an intra-household bargaining narrative. This finding differs from that of UNICEF where the role of family dynamics is downplayed. Because the intention-to-treat effect is heterogeneous across ethnic groups, the intra-household bargaining narrative might be consistent with convention theories since marriage markets are overlapping. My findings challenge the current practice by the policy community.