ABSTRACT
Numerous studies have shown that even after controlling for relevant socio-economic background variables, the labour market position of immigrant minorities lags considerably behind that of natives. The label ‘ethnic penalties’ is often used to denote these gaps and reflects the idea that differences between natives and immigrants that cannot be explained by demographic and human capital variables must be due to discrimination by employers. I challenge this interpretation by looking at the role of sociocultural variables such as language proficiency, interethnic social ties and gender values as alternative sources of unexplained ethnic group differences. I use the data from the cross-national ‘Eurislam’ survey of four immigrant ethnic groups of predominantly Muslim belief—Turks, Moroccans, former Yugoslav Muslims and Pakistani—as well as native ethnics. The results indicate that once sociocultural variables are taken into account, differences in rates of labour market participation and unemployment between native ethnics and the Muslim groups are strongly reduced and in many cases become statistically insignificant. Using mediation analyses, I demonstrate that the findings do not fit a scenario that assumes that the causality primarily flows from labour market participation to sociocultural assimilation rather than the other way around.
