ABSTRACT Since the 1980s, policymakers have implemented dispersal policies to prevent ethnic spatial concentration. However, the effectiveness of these policies is questionable. For this reason, and to enforce the dispersal policy, the Danish government introduced the Danish Integration Act in 1999, which required newly arrived refugees to remain in the placement municipality for three years to remain eligible for social assistance. Leveraging Danish register data, we provide quasi-experimental evidence by evaluating the effects of the 1999 Danish Integration Act on the mobility of refugee households and ethnic concentration. Our results suggest that this policy had a retaining effect for up to 5 years. However, we find that the estimates are no longer significant after 6–7 years, suggesting that resettlement decisions were mainly delayed by the enforcement policy. In addition, ties (in terms of employment, education, and school enrolment) play an important role, as the constraining effect of the policy is greater for households with stronger ties to the placement municipality. Once sanctions were lifted, households that moved opted for neighbourhoods with lower ethnic concentration.