Airbnb and Rents: Evidence from Berlin

By Tomaso Duso, Claus Michelsen, Maximilian Schaefer, and Kevin Ducbao Tran

Short-term rental platforms, like Airbnb, are often suspected of causing rent increases because they remove apartments from the housing market. However, empirically documenting this effect is difficult because neighborhoods that are more popular among tourists usually also have higher average rent levels for various other reasons. At the same time, cities around the world are now regulating short-term rental platforms. Therefore, understanding the effect of short-term rental policy and quantifying the impact of short-term rental platforms on rents is important to inform the policy debate.

BCCP Fellows Tomaso Duso, Claus Michelsen, Maximilian Schaefer, and Kevin Ducbao Tran use the introduction and subsequent update of short-term rental regulation in Berlin (the so-called “Zweckentfremdungsverbot-Gesetz”) to assess the effects of the law on Airbnb in the city. They find that both the introduction as well as its update (which introduced a mandatory registration number display for hosts on short-term rental websites) substantially reduced the number of Airbnb listings in the city. The reduction is largest for entire homes listed on the platform as opposed to shared or private rooms. Furthermore, the authors show that the introduction of the law mostly resulted in apartments leaving the platforms which were available for booking for more days in a year, resulting in a stark decrease in the average booking availability of listings remaining on Airbnb. However, the update of the law mostly caused listings that were only available for rent for fewer days to leave the platform, resulting in a slight increase in average booking availability.

Using these policy-induced changes in the short-term rental market in Berlin, the authors assess the causal impact of Airbnb on rents in the city. Their results suggest that each additional nearby entire home on Airbnb increases rents by at least seven cents per square meter on average. The authors further document effect heterogeneity in various dimensions. First, when focusing on high-availability Airbnb listings, the marginal effect of additional Airbnb listings on rents is larger at ten to 13 cents per square meter. Second, the effect size varies across districts of the city. In particular, the marginal effect tends to be larger in districts with a lower Airbnb density, suggesting decreasing marginal effects on rents of nearby Airbnb listings. Finally, in further calculations, the authors show that renters in high-Airbnb-density districts saved up to 38 euro per month on average due to the introduction of the law. As a caveat, note that the entire analysis in the paper is based on asked rents for apartments available for rent. The effects of Airbnb on average rents of the entire stock of apartments are likely lower.

The full paper “Airbnb and Rents: Evidence from Berlin” is available as DIW Discussion Paper No. 1890.

This text is jointly published by BCCP News and BSE Insights.