The migrant mortality advantage (MMA) has been observed in many immigrant-receiving countries, but its underlying factors remain poorly understood. This paper examines the role of return migration selection effects (“salmon bias”) in explaining the MMA, using a rich, unique data set from France. This dataset consists of native-born and foreign-born pensioners who are tracked worldwide until they die, providing a rare opportunity to assess return migration selection effects and their impact on the MMA, with a focus on males aged 65+.
Results give evidence for the existence of a large and systematic amount of negative return migration selection among foreign-born males in France. Post-retirement returns, in particular, appear highly selective; however, they are not frequent enough to explain the MMA at ages 65+. Pre-retirement returns, by contrast, are much more frequent, and the MMA at ages 65+ essentially disappears once these earlier returns are accounted for. This study expands the concept of the salmon bias by providing evidence that negative selection at return may operate not only at older ages but throughout the entire life course of immigrants, with impacts on the MMA that are larger than previously suggested.