糖心TV

Abstract: Africa remains an outlier in global fertility, yet beneath this continental trend lies striking and persistent variation across ethnic groups. I argue that these differences reflect deep-rooted cultural predispositions toward the quantity–quality trade-off in childrearing. I construct a novel measure of precolonial ethnic human capital investment—capturing the historical presence and intensity of traditional schooling—by combining ethnographic records with machine-learning. Linking this measure to individual-level data from the Demographic and Health Surveys across 23 African countries, and exploiting variation among individuals of different ethnicities who reside outside their ancestral homelands but share the same local environment—thus exposed to the same disease ecology, institutions, and economic conditions—I show that ethnicities with higher precolonial schooling have significantly lower fertility today. Consistent with the proposed mechanism, these groups exhibit greater investment in children鈥檚 human capital and lower desired family size. The findings highlight the long-run persistence of norms associated with human capital investment and introduce a new method for expanding ethnographic records.