Abstract:
|
We studied the changes in crime during the Covid-19 pandemic in Bogota, Colombia, focusing on the periods in which lockdowns were implemented. We used the quasi-experimental implementation of lockdowns in the city using a difference-in-differences estimator of intertemporal treatment effects. Since COVID-19 infection rates determined which zones of the city would be under lockdown, we exploited this exogenous variation to estimate the effect of lockdowns on crime rates. We used the intertemporal treatment effects estimator because it allowed us to unbiasedly estimate the treatment under heterogenous effects while having units treated and untreated at different moments in the period studied. Our results suggest that the national lockdown made all types of crime drop dramatically, and when lockdowns were eased sequentially from area to area in the city, some types of crime, but not all, started rising back up even though mobility was still low. Thus, we have reason to believe that as lockdowns are fully removed and mobility increases, crime will rise even higher. Furthermore, we find that lockdown restrictions have different effects on crime depending on Covid-19 case numbers.
|