2021-06-222021-06-22http://hdl.handle.net/10784/29884We study persistence in criminal capital by looking at the effects of inmates’ releaseson crime around prisons in Colombia. Leveraging detailed geographic and temporal information on the universe of releases from all prisons and crime reports, we find that property crimes are 16% higher around prisons on days inmates are released. Inmates specialized in property crimes drive the impacts. Improvements in non-criminal human capital or longer incarceration spells do not mitigate these effects. These results suggest the specific deterrence or rehabilitation effects of incarceration are weak for individuals with higher initial levels of criminal capital. We also document two externalities resulting from incarcerating specialized criminals. First, we find evidence of adverse peer effects. Second, in a back-of-the-envelope estimation, we document that crime incidence due to prison location drops property values and property tax revenues. Our results raise concerns about the usefulness of incarceration in its most widely adopted form.engCriminal capital persistence: Evidence from 90,000 inmates’ releasesworkingPaperinfo:eu-repo/semantics/openAccesscrimeviolencerecidivismprisonsColombiaAcceso abierto2021-06-22D04H41J24K14K42Escobar Bernal, Maria AntoniaTobón Zapata, SantiagoVanegas Arias, Martín