The promise and perils of the sharing economy: The impact of Airbnb lettings on crime

Abstract

Private short-term letting via Airbnb has exploded in the last decade, yet little is known about how this affects neighborhood crime rates. We estimate the association between Airbnb short-term letting activity and six types of police-reported crime in London, as well as an intervening mechanism, collective efficacy. We estimate these associations with maximum likelihood dynamic panel models with fixed effects (ML-SEM) using data on Airbnb lettings in 4,835 London neighborhoods observed for 13 calendar quarters. We explore mechanisms for the observed effects using multiple lag specifications and by disaggregating lettings into entire properties and spare rooms. We find that Airbnb activity is positively related to robbery, burglary, theft, and violence. These associations are attributable to lettings for entire properties rather than for rooms. Furthermore, associations are contemporaneous, as is consistent with an opportunity mechanism, rather than delayed, as would be consistent with a social control mechanism. Similarly, we find that the association between Airbnb activity and crime is not mediated by collective efficacy. Overall, these results suggest short-term letting contributes to neighborhood crime and these effects are more likely to be attributable to changes in criminal opportunity than erosion of neighborhood social control.