Association for Public Policy Analysis & Management

Using proprietary commercial real estate data from the CoStar Group, this paper is the first to estimate the highly localized effect of a recreational marijuana dispensary opening on nearby retail rents. In order to account for the potentially endogenous relationship between marijuana dispensary location and retail rents, the analysis focuses on Washington State, where recreational marijuana dispensary licenses were distributed via a lottery in 2014. Hedonic difference-in-differences modeling is employed, in which identification comes from comparing highly localized changes in retail rents, in areas where a recreational marijuana dispensary opens, before and after its opening, to counterfactual areas where lottery losers would have potentially opened. The results indicate that the opening of a recreational marijuana dispensary has little to no impact on nearby retail rents, as the estimated effects across a variety of specifications and for various radii around actual and counterfactual dispensary locations are all small. These findings differ from those of recent research, which study the impacts of recreational marijuana dispensaries on residential properties in Colorado and find large positive effects. The lack of consensus motivates the need for further exploration of the impact of recreational marijuana dispensaries on local communities.