Una propuesta de medición de daño del delito y su impacto sobre los precios de Airbnb en la Ciudad de Buenos Aires

Using information from Airbnb accommodations in the City of Buenos Aires, this work analyses the relationship between criminal activities and rental prices. The work proposes a hedonic model that allows estimating the impact of the crime harm index, controlling for other factors (accommodation characteristics and contextual variables). The main contributions of the work can be summarized in three points. First, geographic information is used to construct an indicator of crime damage that weights the types of crimes with respect to the damage produced by them. Second, the study acknowledges the problem of crime endogeneity and provides robust evidence on the impact of crime on Airbnb prices. Third, the main model includes a correction for endogenous spatial dependence on prices and for spatial heteroskedasticity in the error term. Our results indicate that Airbnb prices are negatively affected by criminal activity in the neighbourhood.