Neighborhood quality and opposition to immigration: Evidence from German refugee shelters


This paper examines whether refugee immigration during 2014-2016 affected the quality of neighborhood amenities and support for anti-immigration political parties in Berlin, Germany. From 2014 to 2016, refugees made more than 70 thousand initial applications for asylum in Berlin, equivalent to around 2 percent of Berlin’s population of 3.6 million.

The author examines the impact of temporary refugee shelters on rental prices, neighborhood ratings, and support for anti-immigration parties in the immediate vicinity (within 100m) of the shelters, that is, buildings and amenities within the same block or across the street. The analysis is based on: (a) location and opening/closing dates of refugee shelters from the Berlin Office for Refugee Affairs; (b) listed prices for new rentals from the website; (c) ratings for neighborhood amenities from the website Foursquare; and (d) voting data at the level of electoral precincts.

Main findings:

  • After a refugee shelter is established, rental prices decline three to four percent within 100m of the shelter, relative to locations without a refugee shelter. There is also some evidence of an increasing number of listings when a shelter opens, indicating that previous residents may be leaving these areas.
  • Ratings for a venue are twelve percentage points less likely to be positive after refugee shelters have opened within 100m of the venue.
  • Overall, there is no significant effect on the voter share of anti-immigration parties in precincts which include residents living within 100m of a newly established refugee shelter, but there appear to be heterogenous effects. Areas with high-quality housing and areas with more senior citizens experienced an increase in votes for right-wing parties. In contrast, younger voters, and voters in areas with a more ethnically diverse population became less likely to support anti-immigrant parties after a shelter opened nearby.

Since refugee shelters are temporary, the author concludes that the decline in perceived neighborhood quality in the immediate vicinity of a refugee shelter probably stems from the disutility of living next to the shelter, rather than from long-term concerns about the local labor market or school quality. The author concludes that while there is a local effect of refugee shelters on neighborhood quality, this cannot explain the aggregate increase in support for anti-immigration parties in Berlin after the refugee crisis.