Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We study the causes of inequality in human capital accumulation across ethnic and religious groups. An overlapping generations model in which agents decide how much time to invest in human capital versus ethnic capital shows that the demand for human capital is affected positively by parental and group's older cohort human capital and negatively by group size. Two ex ante identical groups may diverge in human capital accumulation, with the divergence mostly occurring among their low-ability members. Furthermore, group and ethnic fragmentation increases the demand for human capital. We validate these predictions using household data from Nigeria, where ethnicity and religion are the primary identity cleavages. We document persistent ethnic and religious inequality in educational attainment. Members of ethnic groups that historically converted to Christianity outperform those whose ancestors converted to Islam. Consistent with theory, there is little difference between the high-ability members of these groups, but low-ability members of historically Muslim groups choose Koranic education as an alternative to formal education, even when formal education is free. Moreover, minorities and more religiously fragmented ethnic groups fare better, and local ethnic fragmentation increases the demand for formal education. Our analysis sheds light on the political context that underlies the recent violent opposition to "Western education" in this country.