Abstract. For most workers, access to suitable employment is severely restricted by the fact that they look for jobs in the regional labor market rather than the global one. In this paper we analyze how macro-level opportunities (regional market characteristics) and micro-level restrictions (the extent to which job searchers are restricted to the regional market) can help to explain the phenomenon of overeducation. We use a two-step procedure to control for selective access to employment. The results show that the size of the labor market is an important factor in avoiding overeducation.

The theory of career mobility (Sicherman and Galor, Journal of Political Economy, 98(1), 169-92, 1990) claims that wage penalties for overeducated workers are compensated by better promotion prospects. Sicherman (Journal of Labour Economics, 9(2), 101-22, 1991) was able to confirm this theory in an empirical study using panel data. However, the only retest using panel data so far (Robst, Eastern Economic Journal, 21, 539-50, 1995) produced rather ambiguous results. In the present paper, random effects models to analyse relative wage growth are estimated using data from the German Socio-Economic Panel. It is found that overeducated workers in Germany have markedly lower relative wage growth rates than adequately educated workers. The results cast serious doubt on whether the career mobility model is able to explain overeducation in Germany. The plausibility of the results is supported by the finding that overeducated workers have less access to formal and informal on-the-job training, which is usually found to be positively correlated with wage growth even when controlling for selectivity effects (Pischke, Journal of Population Economics, 14, 523-48, 2001).

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte.

and Austria to that of the respective indigenous population. The unit of analysis is the individual in the household context. This allows us to define immigrants' state of integration into the host society at the family level taking into account issues such as immigrant/native intermarriage. Economic performance is measured in terms of the country-specific pregovernment income position and change in the relative income position due to redistribution processes within the respective tax and social security systems. Our work is based on the premise that countries may be categorized -similarly to existing categorizations based on the type of welfare regime -according to the nature of their immigration policy. From an economic point of view, a successful and integrative immigration policy should result -at least when controlling for background characteristics such as education -in a non-significant differential between the economic performance of immigrants and that of the indigenous population. At first glance, our results indicate that this ''ideal'' is not attained in all of the countries analysed, particularly not in Germany and Denmark, where the economic performance of immigrants is much lower than that of the indigenous population. However, results from GLS random-effects models show that immigrants to these countries improve their economic situation rapidly with increasing duration of stay in the host country. This implies that these countries also do fairly well in fostering in the economic integration of immigrants. Our empirical results further reveal that the substantial cross-country differences in the immigrant/native-born performance differential persist even when controlling in detail for socioeconomic characteristics of the household and for indicators of the state of the immigrants' integration, such as years since migration and immigrant/ native intermarriage. This suggests that not only the conditions of entry to a country impact on immigrants' economic performance, but also country-specific institutional aspects such as restrictions on access to the labour market and parts of the social w This paper benefited substantially from discussions at the EEA 2003 Conference in Stockholm, Sweden, and the EPUNet 2003 Conference in Colchester, UK. We also acknowledge very helpful comments resulting from the reviewing process. (2005) 24: 175-212 Ó Springer 2005 Population Research and Policy Review