Abstract
Identifying the impacts of conflicts and understanding the origins of gender gaps are both seemingly unrelated but crucial questions in the literature. Focusing on the gap at the intersection of these two branches of literature, this study explores whether and how longrun exposure to violent conflicts contributes to and shapes the child gender preferences of parents.
I use temporal and spatial variations in conflicts in Nigeria and combine the Uppsala Conflict Data Program and the Demographic and Health Surveys Program to perform the analysis. The results show that the effect of long-run exposure to violent conflicts on stated preferences (attitudes) for boys is not homogeneous.
While conflict events with low or no civilian death increase preferences for sons, violence targeted at civilians works in the opposite direction and decreases preferences for boys. I find no evidence of translating these preferences into behaviour via sex-selective abortions.
Instead, evidence shows that parents use the stopping rule to achieve the desired gender composition of children. Further, analysis also indicates that, in the districts affected by conflict, parents have a positive bias towards boys in terms of their postnatal health investment.