IED Discussion Paper 89, September 1998

Many recent studies comparing the living standards of immigrants to the local population raise interesting questions concerning the underlying motivations for migration and the assimilation of migrants into the local population. There is some evidence of an overtaking phenomenon where the income of immigrant families eventually exceeds that of locals. There is also evidence concerning the apparent assimilation of local fertility rates by migrants. Berman and Rzakhanov argue that both these could be consequences of self-selection of migrants on the basis of intergenerational altruism, rather than assimilation. Specifically, parents may consider migration to a higher income region as a costly investment in their children’s human capital. An implication of this theory is that families with greater altruism migrate when the costs of migration are high. The theory generates a number of other testable implications for the number of children and investment in their education for migrants relative to local residents.

support for this self-selection thesis. Countries that have historically absorbed immigrants (the US, Canada, Australia and New Zealand) average half a child more than the Western European countries. However, the most compelling argument for the selection hypothesis is provided by the study of migration of East-European Jews to Israel, whose fertility rates exceed those of similarly secular native Israelis.

Immigrants who arrived before 1980, despite high costs, average 2.8 children per woman, 0.9 more than immigrants who arrived since 1989 when costs were lower! Adjusting for age on arrival reveals that at least half of that 0.9 child gap is due to selection (with the remainder due to assimilation). This large selection effect is consistent with the theory of heterogeneity in altruism and is difficult to explain otherwise.