Elise Stenholt Sørensen (p), Anders Holm

Gaetano Vecchione

"see document"

Kerstin Tanis (p)

Elise Stenholt Sørensen

This paper investigates new immigration inflows into local labour markets and their impact on the geographic mobility of natives and earlier immigrant cohorts. In the last decades, Germany experienced several immigration inflows due to substantial institutional changes, for example, the ongoing EU Enlargement since 2004. Growing literature has focused on the impact of immigration on local workers’ earnings and employment opportunities. So far, some studies find negative, some find positive effects, others find no effects at all. Latter may be caused by selective out-migration of local residents: inter-regional mobility might simply “arbitrage” the effect of immigrants locally away (Borjas, Freeman, & Katz, 1997). Therefore, this paper claims that adjustments in inter-regional mobility of former residents are the predominant issue for analysing further impacts of immigration on local labour market outcomes in Germany.
Theoretically, I exploit two key aspects of recent immigration inflows: the significant changes in new immigrants’ skill structures and the variation of immigrant inflow rates across regions (Card & DiNardo, 2000). Immigrant receiving areas have done well in some periods and poorly in others. This produces a spurious understanding of the impact of immigration on inter-regional mobility of local residents. Thus, it seems not sufficient to compare mobility patterns between immigrant and non-immigrant areas. Following the equilibrium approach of Borjas et al. (1997), new immigrant inflows will affect the relative wage and employment outcomes of the local population by the extent to which inflows change the proportion of the population in skill/occupation groups. If the skill/occupation structure of immigrants and natives matches, immigration will increase local labour supply. Consequently, local wages will fall and former residents might respond with inter-regional mobility.
The empirical analysis on county level (NUTS 3) includes OLS as well as IV regression using the multiple instrumentation approach for the relative inflow rate of new immigrants developed by Jaeger, Ruist, and Stuhler (2018). The relative growth in the native commuters’ population and the relative growth in the overall native population (movers) serve as age-adjusted dependent variables for each skill group. The primary explanatory variable is the relative growth of the immigrant population in the specific skill group. The analysis is based on the Integrated Employment Histories (IEB, 1975-2014) provided by the Institute for Employment Research (IAB) and includes a thirty percent sample for immigrants and a two percent sample for natives.

Annamaria Nifo , Domenico Scalera , Gaetano Vecchione (p)

Kerstin Tanis

See extended abstract