THE POSITIONING OF TRANSNATIONALLY MOBILE EUROPEANS IN THE GERMAN LABOUR MARKET

ABSTRACT

This article analyses the economic integration of transnationally mobile Europeans into the German labour market. The measure used for the degree of integration is their relative income. Specifically, the article attempts to answer the question, whether and to which degree the integration of transnationally mobile Europeans into the labour market depends on their national origins or on their social origins. Based on the concept of transnational social mobility which goes beyond classical migration research, we will point out the factors that position mobile Europeans in the labour market and the feedback effects on national class systems. The results show that it is not national origin but social-structural characteristics that explains the positioning of EU-transnationals. The empirical basis of this article is primary data on Berlin's labour market gathered in the beginning of 2002.

Notes

¹Transnational mobility is not a broadly established concept in the research. Recently, the term is used at the interface between labour market research and migration research. Jordan and Düvell (2002) or Fassmann (2003) for example conceptualise migration as transnational mobility because they perceive migration as one form of transnational labour market mobility or as a special mode of inner-European migration, respectively.

²Super-Stratification is not a widely used sociological term. In historical research super-stratification refers to invasions that installed a ruling class in high civilizations (Allsen 1987). We believe that the notion of super-stratification (i.e., the installation of a (ethnic) upper class ‘on top’ of the existing class structure) applies quite well to our conceptualisation of the possible effects of transnational mobility on national class systems.

³The mobility of Polish is regarded as a special ‘reference case’. Poland did not belong to the EU at the time of the survey. However, it is a full member now. In addition, Poland and Germany are connected by mobility streams since the immigration of Polish workers into the mining fields of the Ruhr District in the nineteenth Century.

⁴Based on the thesis of the importance of national origin (which derived from the ‘classical’ migration research) one would assume that individuals, depending on their national origin, have varying degrees of success in integrating themselves into the German labour market (e.g., French integrate with better results than Italians). Our theoretical model claims that rather social status influences the course of integration of transnationally mobile Europeans. As the empirical results will show, these effects of social status develop in different ways.

⁵The empirical results are based on a ‘pilot study’ in Berlin. For the next step it is planned to extend the research to other metropolitan areas in Germany in order to be able to depict the different regional labour markets in a more sufficient manner. The sample of the survey is comprised of 501 people from the German control group, 409 Britons, 426 French, 344 Italians, 263 Poles and 100 Danes. In order to reach a good return rate, a questionnaire in German and in the native language was sent to all interviewees. The address sample had a volume of 11,200 individuals. The net sample of 10,500 individuals was smaller because of address errors.

⁶It is assumed that postal surveys have advantages and disadvantages, like other survey techniques do. However, a comparison between postal surveys and telephone or face-to-face surveys shows that mail surveys have specific qualities and are unjustly underrated in sociological research (Dillman 2000).

⁷A further examination of the data quality, e.g., with regard to education, is not possible, for none of the usual comparison methods (census data, official statistics) are available for the investigated ‘national groups’.

⁸The operationalisation and description of all variables can be found in the Appendix.

⁹One could also have chosen a linear regression model (which would have produced results similar to an ordered-probit model) based on a conversion of the original categorised variable into a continuous variable of hourly wage plus using a natural logarithm (Petersen 1989; Rosenfeld and Kalleberg 1990). Such an income conversion corresponds to the procedures in this type of research and is justified by the unequal distribution of earned income and by the interpretability of the coefficients as percentage effects (Petersen 1989). We preferred an ordered-probit model, mainly to be able to apply to the categorised character of the available income variable.

¹⁰However, contradicting this second assumption is the fact that the women surveyed indicated relatively frequently that they followed their male partner to the new country (Verwiebe 2004).

¹¹We only report the significant interaction effects. The complete tables are available upon request.