Long-run integration of refugees : RCT evidence from a Swedish early intervention program

This study uses a randomized control trial (RCT) to evaluate a new program for increased labor market integration of refugees. The program has immediate and substantial short-run effects on employment, corresponding to around 15 percentage points. The effect lasts for three years but eventually fades out, as the control group catches up and reaches the long-run employment level of about 50 percent. We show that the program boosts language skills in the short run, and that this channel explains an increasing share of the effect on employment. Using survey data, we finally measure if the program affects integration in other dimensions, such as psychological, social, political, and navigational integration. Our findings suggest that faster labor market integration in the short run does not lead to increased general integration in the long run.