Parental gender preferences over three centuries: Evidence from Argentina

In this paper I examine the evolution of parental gender preferences in Argentina (i.e., parents who prefer a certain gender composition in their children). To do this, I use census microdata that spans the 19th, 20th, and 21st centuries. The estimation strategy exploits the plausibly random assignment in the gender of children. The results show a persistent preference for a mixed gender composition (i.e., having at least one boy and one girl) instead of children of the same gender. This translates into an increase in the probability of having a third child, conditional on already having two children of between 9%-23% for those couples who have children of the same gender -in relation to couples with children of opposite genders-. These preferences are heterogeneous over time and have important implications in terms of fertility (i.e., the reduction of these mixed gender preferences -in favor of greater gender-neutrality- could contribute to reducing the number of children per couple). In addition, the findings of this work support the empirical literature that uses the gender composition of the first two children as an instrumental variable to study the impact of fertility on labor participation.