Abstract
AbstractUsing data from 60 countries, we measure how much couples agree on sex preferences for children and whether differences in sex preferences are associated with gender gaps in children's education. Results show extensive disagreement in sex preferences for children, with husbands far more likely to want more sons but their wives more likely to prefer having equal numbers of boys and girls, wanting more daughters, or having no preference. India has the highest share of agreement on sex preferences (59 percent), and Niger has the lowest (32 percent). The association between couples’ sex preferences and gender gaps in education differs considerably by country. In some countries, girls have worse outcomes when their parents agree on son preference and better ones when parents agree on daughter/no preference. But there are numerous counter‐examples as well. Gender gaps in education appear more often when wives hold son preference but not their husbands than the reverse combination. Agreement on daughter/no preference is the only category that is systematically associated with better outcomes for girls relative to boys (although even here there are caveats). Balanced preference (wanting as many boys as girls) is an ambiguous category with heterogenous patterns in terms of educational gender gaps.