Wage assimilation of immigrants : which factors close the gap? - evidence from Germany

Using longitudinal employment register data this study analyzes the development of outcomes of male foreign workers from all important sending countries across time. Cohort analyses on persons entering the German labour market between 1995 and 2000 show significant differences in the assimilation processes between nationalities. We examine the sources of wage assimilation for migrants in Germany by estimating fixed effects regressions for migrants and Germans separately. Based on the estimated coefficients, we assess the contributions of the various right hand side variables to wage convergence between migrants compared to native Germans by generating predictions and by averaging them by year. This approach allows to decompose the wage adjustment of foreigners into three components: the adjustment due to observed time-varying characteristics as e.g. tenure or change of industries/occupations, the adjustment due to selectivity effects and a third component capturing the adjustment due to unobserved time-varying factors and age. If all migrants are tarred with the same brush (by pooling them into one group), we find that the raw wage gap (compared to native Germans) decreases from -50 percent in the year 2000 to -40 percent in 2008. According to the decomposition results this wage assimilation of 10 percentage points can almost completely be traced back to time-varying observable characteristics. Migrants increase their wages over time mainly by changing their workplaces into larger firms, by accumulating firm-specific human capital and by sorting themselves into better-paying industries and occupations. Furthermore, employment stability contributes to the wage convergence as well as changes to German citizenship. Selectivity- and age/trend- effects play only a minor role here. However, these results from a ‘one size fits all model' pooling all migrants into one group mask considerably heterogeneity between countries. The large size of our data allows us to differentiate between 30 nationality groups. We observe that the wage assimilation is significantly higher for countries with a small entry wage gap compared to German natives. It turns out that a large proportion of the total wage adjustment can be explained by selectivity effects especially for migrants from countries with large wage increases whereas adjustment is mainly driven by time-varying observable characteristics for countries in the midfield of the adjustment ranking.. In summary, the nationality-specific results reveal considerable heterogeneity regarding adjustment and its causes, indicating that effective migrant policy should account for that by creating nationality-specific integration measures.