Does a Federal Penitentiary impact crime? Evidence from Catanduvas, Brazil

Abstract

The investment in prison infrastructure is a policy to fight crime. On the other hand, if it influences in the spatial displacement of crime, the construction of a penitentiary may generate negative localized effects. Dispite this, there are few studies evaluating the potential localized effects arising from the establishment of new maximum-security prisons. We use a 24-year longitudinal data on sinthetic control models to estimate the impacts of the construction of the Catanduvas Maximum-Security Penitentiary on homicide rates in its microregion. Furthermore, we combine the observation of spatial spillover effects with the synthetic control methodology to achieve better fit between the pretreatment control and the synthetic unit. Our findings suggest a transient and isolated increase in local homicidies, which support that the prison did not act as a catalyst for long-term local violence.