Job Displacement, Unemployment Benefits and Domestic Violence

Abstract:

We provide the first causal estimates of how individual job loss of men and women influences the risk of domestic violence (DV), and whether unemployment benefits mitigate these impacts. Estimating this confluence of three parameters on a given sample places us in a strong position to illuminate the underlying mechanisms. Using data on about 2 million domestic violence cases brought to criminal courts in Brazil during 2009-2017, we identify the defendant and the plaintiff in longitudinal employment registers. Leveraging mass layoffs for identification, we find that both male and female job loss, independently, lead to a large and persistent increase in domestic violence. These results satisfy a number of checks on selection into treatment, identification of dynamic heterogeneous effects, and the concern that they are driven by endogenous changes in reporting. The same pattern of results emerges using data on couples in the sub-sample of individuals captured in the social welfare register. It also holds when we measure domestic violence using information on women’s use of public shelters rather than as prosecution or protective measures against men. Exploiting a discontinuity in unemployment insurance eligibility, we find UI does not have an ameliorating influence while benefits are being paid, but that eligible men (who have longer unemployment durations) are more likely to commit DV than ineligible men once benefits expire. Our findings indicate a role for income loss and exposure as mechanisms. The reason that UI does not mitigate is that it increases exposure while relaxing income constraints.

If you are interested in attending this seminar, please contact host Rubén Poblete-Cazenave.