This paper focuses on the residential resettlement decisions of a sample of immigrants from Iran and Turkey living in Sweden between 1968 and 2001. Using the Swedish Longitudinal Immigrant database, we are able to link unique pre‐ and post‐migration data to understand whether region of origin is a better predictor of internal migration decisions than is country of origin, the more often used measure in existing research. More specifically, we test whether living in municipalities with a high number of individuals from the same country of origin is a similar phenomenon as a high number of individuals from the same region of origin. This is relevant, as large immigrant groups come from ethnically, religiously, and linguistically heterogeneous countries of origin where regional characteristics differ according to aforementioned aspects from that of the mainstream population. We indeed find that individuals are less likely to relocate from municipalities in which there is a large presence of other immigrants from the same region of origin. Instead, individuals residing in areas with a large number of individuals from their country of origin are observed with an elevated probability of resettlement. Copyright © 2016 John Wiley & Sons, Ltd.