Labor market outcomes, capital accumulation, and return migration: Evidence from immigrants in Germany

In this paper I test the capital accumulation conjecture that is used to rationalize return migration decisions in the context of immigrants in Germany and examine how labor market outcomes influence return migration decisions, with particular attention to selection in these outcomes in return migration. I characterize the level and timing of return migration as well as the selection in it and derive a number of implications of these on the impact of immigrants on the host as well as source countries. Using a rich longitudinal dataset that has an over-sampled group of immigrants (German Socioeconomic Panel), I conduct a Cox proportional hazard analysis with alternative waiting-time concepts. That the sample contains immigrants from four different source countries allows me to utilize the variation in the source country characteristics as well as the time variation in them to identify the parameters of interest. I find evidence for the savings accumulation conjecture, in which return is motivated by higher purchasing power of accumulated savings in the home country. On the other hand, human capital accumulation conjecture is rejected. In the framework of savings accumulation, I examine the impact of an increase in German earnings whose theoretical impact on the return migration decision is ambiguous. In terms of labor market outcomes, both retirement and unemployment emerge as important determinants of return migration choices. Unemployment spell length determines the direction of selection with respect to unemployment in return migration. The data also reveal that the level of return migration is high and varies considerably across the source countries. The hazard function of Turkish immigrants displays a hump-shaped profile that peaks between the ages of 45 and 54 whereas EU immigrants are more likely to return at earlier ages and after retirement.