Population sex ratios and violence against women: the long-run effects of sex selection in India

This paper investigates the consequences of sex imbalance in India's population for violence against women. We match district level administrative crime data by category to age- specific sex ratios in census data across four decades and, to analyse mechanisms, we also use administrative data on marriage rates and household survey data on attitudes to violence against women and marriage quality. We estimate that the elasticity of violence against women with respect to the surplus of men age 20-24 is unity, and that this explains about 35% of the rise in gender-based violence since 1995. Although less robust, there is some evidence that the youth sex ratio also raises non-gendered forms of violence, but we find no discernible impact upon property and economic crime. In probing mechanisms we argue that men are more prone to crime than women, that the share of unmarried men is increasing in the youth sex ratio, that attitudes to violence against women are evolving as a function of the sex ratio at birth and marriage quality measures, including self-reported domestic violence, are negatively related to sex ratios.