Refugees placed in Dutch regions with more Syrian residents formed more connections with other Syrians, while areas with more Dutch residents were linked to more contacts with Dutch people.
Researchers used the Netherlands’ refugee dispersal policy, which allocates refugees across regions in a largely unpredictable way, to study whether housing location affects social contacts. They combined a longitudinal survey of Syrian refugees with register data describing the ethnic makeup and urbanity of their housing regions.
Regions with a larger share of Syrian-descent residents led to more contacts with other Syrians. Regions with a larger share of Dutch residents led to more contacts with Dutch people, with tentative evidence of a positive effect on contacts within refugees’ own origin group as well. The effects did not diminish over five years, and urbanity had no robust effect.
Where contacts formed
The study analyzed 7,387 observations from 3,137 Syrian refugees in the Netherlands. Refugees placed in regions with a higher share of Dutch residents developed more social contacts with Dutch people. There was also tentative evidence that these regions increased contacts within refugees’ own origin group, but they did not affect contacts with other migrants.
Refugees placed in regions with a higher share of residents of Syrian descent developed more contacts with other Syrians. These regional effects remained over a five-year period. The researchers found no robust effect of how urban a region was on refugees’ social contacts.
Evidence and limits
The study uses a representative longitudinal survey and register data, together covering 7,387 observations from 3,137 Syrian refugees. The Netherlands’ dispersal policy quasi-randomly allocated refugees across regions, allowing the researchers to estimate the effects of housing location more convincingly than a simple comparison of places where refugees chose to live.
The evidence concerns Syrian refugees in the Netherlands and the social contacts measured in this study, so the findings may not apply in the same way to other refugee groups or countries. The evidence for increased contacts within refugees’ own origin group in regions with more Dutch residents is described as tentative, and the study found no robust effect of urbanity.
