Does Legal Change Induce Behavioral Change in Domestic Violence Against Women?

In this paper, we study the effectiveness of the Brazilian "Maria da Penha" Law-a law that imposed criminal sanctions on and promoted rehabilitation programs for known perpetrators of acts of domestic violence exclusively against women, issued restraining orders against perpetrators, required perpetrators to surrender firearms, and established police units uniquely trained to address violence against women. Using difference-in-difference analyses estimated with data on Brazilian monthly assault-related hospitalizations and with gender as treatment variable, we find that the law contributed to a 22% reduction in assault-related hospitalizations among women as compared with men. Our data is collected from the Brazilian public healthcare system named Sistema unico de Saude, which provides public information on detailed records of hospitalizations in Brazil. We provide robustness checks of our analyses and examine heterogeneous effects of the law. With respect to the latter, we deduce that the effectiveness of the law in reducing assault-related hospitalizations among women was especially pronounced in municipalities where women are most vulnerable to domestic violence. We conclude with a discussion of our results, with an eye to policy implications and areas for future research.