Reducing the Social Assistance Benefits of Refugees Affect Their Residential Integration? Evidence from a Natural Experiment

I exploit quasi-random assignment of social assistance benefits induced by a major Danish reform to obtain credible causal evidence for the effect of social assistance reductions on residential integra-tion (i.e., the extent to which refugees settle among natives). Comparing otherwise similar refugees, I find that reducing the benefits of refugees markedly deteriorated their residential integration espe-cially in the long-term. In particular, refugees’ residential segregation increased by about 28 percent as a consequence of the benefit reductions. I show that this effect most likely runs through a depri-vation mechanism, where refugees live on a subsistence minimum that significantly limit their loca-tion choice. Moreover, I demonstrate that the overall effect is concentrated among the low educated who face the largest resource constraints, most marginalization, and have the worst chances of inte-grating into the host society at the outset.