Gothenburg's Early-Intervention Programme for Newly Arrived Refugees — Language Training, Work Practice and Job Search in an RCT

A randomized controlled trial in Gothenburg tested a programme starting soon after a residence permit is granted: three months of intensive language training, six months of supervised work practice, then caseworker job-search help.
The trial randomly assigned 140 newly arrived refugees, 90% with no education beyond elementary school, to treatment (70) or control (70); 44 accepted the offer. Treated refugees received about 160 more hours of language training. Around 30% of the treatment group were employed each month in the year after the programme ended, versus about 15% of controls (roughly 15 percentage points). Treated refugees were almost 30% more likely to have completed Swedish language courses, but documented language skills explain only about 7-10% of the employment effect.
A later IFAU follow-up reports the effect lasts about three years and then fades as the control group catches up. Limits: a small sample from a single city, a non-peer-reviewed working-paper evidence base, and cost estimates indicating the programme was cheaper than earlier labour-market programmes.

Read the full analysis: https://ifn.se/en/publications/working-papers/2011-2020/2020/1372/