"Theoretical background and objectives
This project explores how policies regarding immigrant rights and welfare state regimes have affected the socio-economic integration of immigrants. Most of the literature on immigrant integration assumes that the granting of easy access of immigrants to citizenship rights and government recognition and support for cultural diversity promote the socio-economic integration of immigrants. At the same time, existing work (e.g., Borjas, van Tubergen) has shown that immigrants with low human capital resources tend to migrate preferably to countries with equal income distributions and extensive social security protection. This raises the question whether immigrant integration policies that grant easy access to citizenship rights, and thus also full access to welfare state rights, might have the unintended consequence that they produce a high rate of dependence of immigrants on welfare state arrangements and attendant socio-economic marginalisation in other domains. If integration policies in addition do not demand cultural assimilation (e.g., in the domain of language) the risk of lower-skilled immigrants to become dependent on welfare benefits may further increase. This hypothesis of an interaction effect between integration policies and welfare state regimes is confronted with cross-national data on labour market participation, residential segregation, and imprisonment of immigrants. Where possible, these comparisons are controlled for cross-national differences in the composition of immigrant populations by drawing on comparative data for particular ethnic groups. The analysis includes eight West European countries that have turned into immigration countries at roughly the same time in the 1960s and early 1970s, where institutions have therefore had several decades to affect integration outcomes. They vary both strongly regarding integration policies (including the highest, Sweden, and the second lowest scoring country, Austria, in the 2007 Migrant Integration Policy Index) and regarding welfare state regimes (with Sweden and the United Kingdom at the extremes).
Research design, data and methodology
The study relies on various indicators of immigrant rights, prevalent typologies and indicators of welfare state regimes, and data from the European Labour Force Survey, International Prison Statistics, as well as results from a large number of previous studies on immigrants' labour market participation, residential segregation and imprisonment. To control for composition effects, the labour market data refer to immigrants from non-EU countries, and for specific country contrasts specific ethnic groups (Turks and ex-Yugoslavs). Residential segregation data refer to a few dozen European cities, partly referring to specific ethnic groups (e.g., Turks, Maghrebians, Caribbeans, Pakistani) and partly to more general categories (Muslims, foreigners, immigrants).
Across the three domains of socio-economic integration a consistent cross-national patterns is found (with the exception of residential segregation in the United Kingdom) in which the gap or the degree of segregation between immigrants and the native population is largest in the countries that combine easy access to citizenship rights and a large degree of accommodation of cultural differences with a relatively encompassing and generous welfare state (Sweden, the Netherlands, Belgium). Both the United Kingdom, which combines inclusive integration policies with low welfare state provision levels, and the three Germanophone countries (Germany, Austria, Switzerland), which combine restrictive policies with – at least in the German and Austrian cases – moderately strong welfare states, show relatively small gaps between immigrants and natives. These findings are confirmed for contrast comparisons for specific ethnic groups. For instance, compared to the native population, Turks in the Netherlands have much lower rates of labour market participation than German Turks, and similarly ex-Yugoslavs in Austria perform much better than those in Sweden. Because the results are mostly based on aggregate data – although some of the studies that are used do control for individual-level variables – they need to be further tested by taking individual and local context data more systematically into account. This will be one of the aims of the analyses in the context of project 6.3 further below."