Ethnic Favoritism: Micro Evidence from Guinea

Abstract

In an ethnically diverse country, does it matter to the welfare of ordinary citizens which ethnic group is in power? This paper exploits a plausibly exogenous change in the ethnicity of the president of Guinea in 1984 to identify the effect of living under the rule of a national leader of the same ethnicity on infant mortality. By using the retrospective fertility survey of women in a nationally representative sample of households in 1992-93, mother fixed effects estimation results show that, after 1984, infants born to mothers of the new president's ethnicity are not substantially less likely to die within the first year of life than those born to mothers of other ethnicity. Since the government had the capability to reduce infant mortality during the period of study (the yearly average infant mortality rate declined rapidly under the new president's rule), the finding suggests that the new president did not favor his own ethnic group in his effort to improve child health to a large extent as far as ordinary citizens were concerned.