Movement without integration? An analysis of labor market attainment in Sweden among its post- war immigrants from Estonia, Latvia, Lithuania and Poland

Following the end to Soviet occupation the question of an integrated Baltic Sea region has been high on the policy agenda. There is now talk of the European Union adopting a Baltic Sea Strategy in 2009 in order to promote deepened integration and prosperity in the region. In this paper a truly fundamental aspect of such integration is approached, namely that of labor market attainment and integration. By offering an economic analysis of the fate of previous immigrants, using individual data, we seek to address the question of incentives for current and future labor market migration and integration in the region. In this study we thus investigate the labor market attainment of all immigrants of working age from the neighboring Baltic countries on the Swedish labor market in 2004. As neighboring Baltic countries is here understood Poland, Estonia, Latvia and Lithuania, henceforth alternatively referred to as either PELL countries or Baltic countries. Data is a special delivery from Statistics Sweden and comprise all immigrants to Sweden from these countries from 1944 onwards. Comparisons are made with a stratified random sample of persons born in Sweden. Stratification is made on age and gender. The analysis is conducted in three steps. In the first step we investigate labor force participation at the time of observation. This is done by estimating the probability of being part of the labor force in 2004, defined as either being employed or actively looking for employment. The group not belonging to the labor force thus contains both those permanently out of the labor force (e.g. early retirees) and those temporary out of the labor force (e.g. students and those on parental leave). In the second step we estimate the probability of employment among those included in the labor force. Employment is here defined as having a registered income in excess of one basic amount (SEK 39,300; app. €4,000). In the third and final step where we analyze the earnings of those employed we control for both types of selection described above. This means that we estimate a model with double selection. Preliminary results show that immigrants are less likely to be part of the labor force; that they are overrepresented among the unemployed and that the positive effects from higher education is less for immigrants than for native Swedes. However, among those in the labor force and being employed the income differences are relatively minor. Nevertheless, the fact that we find large negative effects relating to educations remains an issue of concern.