African leaders are widely presumed to favor members of their own ethnic groups
with patronage resources. We assess the empirical validity of this claim by studying
ethnic favoritism in the education sector in Kenya. We use data on the educational
attainment of more than fifty thousand Kenyans dating back to the colonial era, as well
as information about the ethnic identities of Kenyan presidents, cabinet members, and
high-level education bureaucrats since independence. We find that having a coethnic
as president during one’s primary school-age years is associated with about a onequarter
of a year increase in years of primary schooling with and substantial increases
in the probability of attending and completing both primary and secondary school.
We also find that ethnic favoritism extends beyond the president: coethnics of the
minister of education also acquire more schooling than children from other ethnic
groups. In contrast to the findings of some recent studies, we find that multiparty
political competition has no impact on the degree of ethnic favoritism by presidents and
ministers of education. Nor has ethnic favoritism in the education sector varied across
Kenya’s presidents. Having established the role of ethnic favoritism in educational
outcomes, we identify and provide suggestive evidence about the possible mechanisms
explaining these patterns. We also highlight the substantive importance of ethnic
favoritism by comparing our estimated effects to those estimated in studies of policy
interventions designed to increase educational achievement in the developing world and
by estimating the long-term impact of ethnic favoritism through education on a number
of long-term socio-economic outcomes.