Do Integration Policies Affect Immigrants' Voluntary Engagement? An Exploration at Switzerland's Subnational Level

Abstract

This paper investigates whether integration policies influence immigrants' propensity to volunteer, the latter being an important element of immigrants' integration into the host society. By distinguishing different categories of integration policies at Switzerland's subnational level and applying a Bayesian multilevel approach, our results suggest varying policy effects: while policies fostering socio-structural rights enhance immigrants' propensity to volunteer, we observe a negative curvilinear relationship between cultural rights and obligations and immigrants' volunteerism implying that a combination of cultural entitlements and obligations is most conducive to immigrants' civic engagement.

Keywords:

• Integration Policy
• Immigrant Rights and Obligations
• Immigrant Volunteering
• Social Capital
• Subnational Comparison

Acknowledgements

We thank the anonymous reviewers, Marco Giugni, Marc Helbling, Rainer Bauböck, George Vasilev and Tijana Prokic-Breuer for their helpful comments on earlier versions of this article.

Notes

[1] Big cities in Switzerland often formulate their own integration strategies. Yet, cantonal law prevails when it comes to define the competences of municipalities and thus, cities, in this policy field (TAK 2005, 26). The considerable autonomy of Swiss cantons in Swiss integration policy, which rests on Switzerland's political principles of executive federalism and subsidiarity, has been further corroborated in article 57 of the new aliens' law.

[2] For a detailed discussion of our theoretical conceptualisation of cantonal integration policy based on the international framework (Koopmans et al., 2005; Koopmans 2010), as well as our empirical measurement see Manatschal (2011).

[3] Volunteering can of course take place in many different organisations, whereby some of these organisations (e.g. immigrant associations) may flourish as a result of ‘cultural distance’ and therefore result in an increase of such organisations and eventually in immigrant volunteering. Based on our data, we cannot explicitly know how many immigrants actually engage in this kind of organisations. But most recent data from the Swiss Volunteering Survey show that only 5% of all (native and immigrant) respondents indicate that they are members in an immigrant association.

[4] Of the 1814 immigrants interviewed, 78 show missing values for the dependent and/or independent variables and are therefore excluded from the analysis. The number of immigrant respondents per canton are as follows: Zurich (285), Bern (104), Lucerne (49), Uri (10), Schwyz (35), Obwalden (13), Nidwalden (9), Glarus (30), Zug (25), Fribourg (47), Solothurn (45), Basel-Town (64), Basel-Country (63), Schaffhausen (34), Appenzell Outer Rhodes (13), Appenzell Inner Rhodes (17), St. Gall (82), Grisons (27), Argovia (100), Thurgau (58), Ticino (114), Vaud (207), Valais (47), Neuchâtel (56), Geneva (184) and Jura (18).

[5] We should note here that our migrant sample is not representative: immigrants account for ca. 13% of all respondents in the survey, whereas their share amounts to ca. 23% of Switzerland's overall population. Furthermore, the survey includes only immigrant respondents who know one of the Swiss languages (German, French or Italian), meaning that our immigrant sample is rather specific. However, the sample of immigrant respondents included in the Swiss Volunteering Survey represents fairly well the socio-demographic characteristics of migrants in Switzerland (Stadelmann-Steffen et al. 2010, 140f). As the aim of our analysis is to estimate relationships and not representative shares of immigrant volunteers, we accept this lack of representativity in this context.

[6] The focus on Islam is justified by the fact that it is currently the largest religion in Switzerland besides the two Christian confessions (see census data 2010 on http://www.bfs.admin.ch/bfs/portal/de/index/news/04/01.html, last accessed: September 20 2012). A similar focus on Islam is applied by Koopmans et al. (2005, 52) who argues that Muslims are a relevant religious minority, which is furthermore most affected by contemporary controversies on cultural differences.

[7] We refrain from further integrating the size of the foreign population into the models. First and foremost, it is theoretically not clear whether a large share of foreigners would hinder or rather promote the likelihood of immigrant volunteering. Further analyses not presented here also suggest that a large immigrant population is not correlated with the probability of immigrant volunteering, nor does this variable influence the estimation of other parameters.

[8] The Bayesian Deviance Information Criterion (DIC) amounts to 1344 for Model 2 and to 1343 for Model 3. In order to speak of a significant improvement, the DIC should decrease by approximately five when adding two additional variables.