• Integration Policy;
  • Citizenship;
  • Subnational Comparison;
  • Swiss Cantons


  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices

Abstract:  Subnational varieties of immigrant integration policy, which are particularly salient in federal states, remain largely neglected by migration studies. Following Lijphart, who long demanded to verify international research at the subnational level, this study aims at capturing subnational policy variations using the example of Swiss cantons. In line with international approaches, cantonal integration policies are conceptualized and measured in terms of immigrants’ ease or difficulty of access to civic, political, socio-structural, as well as cultural and religious rights and obligations. The transfer of an international concept to the subnational level facilitates a validation of the former, which constitutes a second neglected research field. Finally, a look at the empirical evidence allows testing the construct validity of our measurement: in line with theoretical assumptions, our data reveal a clear linguistic divide, an institutionalised “Röschtigraben”, with German speaking cantons exhibiting overall more restrictive policies than Latin cantons.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices

Comparative studies analysing policies of immigrant integration or related policy fields focus commonly at the national or cross-national level. Among the classical contributions to the migration literature we find many comparative case studies, such as Brubaker’s (1992: 55) analysis of citizenship and nationhood in France and Germany, or Favell’s (1998) study on philosophies of integration in France and Britain. Another cross-country comparison which, contrary to the aforementioned references, analyses policies at the local instead of the national level, is Ireland’s (1994) comparison of immigrant politics in France and Switzerland. More recently, the list of countries has been extended by more quantitatively oriented approaches of Koopmans et al. (2005), who compare integration policies in France, Britain, Germany, Switzerland and the Netherlands, or Howard, who developed a citizenship policy indicator for the former 15 EU countries (Howard, 2006). Worth mentioning is finally the Migrant Integration Policy Index (MIPEX)1, an instrument led by the British Council and the Migration Policy Group, comparing integration policies in all current EU countries plus Canada, Switzerland, Norway and the United States.

At the same time, subnational variations of integration policy, which are particularly salient in federal states, remain largely neglected by the scientific literature (cf. Akgün and Tränhardt, 2001, Henkes, 2008). Keeping in mind that a majority of post-war immigration countries are federal states this disregard is astonishing. Subnational variations of integration policy have been reported for the United States (Schmitter Heisler, 2001), Canada (Schmidtke, 2001), or Germany (Henkes, 2008), and they are particularly pronounced in strongly federal Switzerland. Swiss cantons enjoy a high degree of autonomy which results in a heterogeneous puzzle, a “Mini-Europe” (Cattacin, 1996: 69) of subnational integration policies (cf. Cattacin and Kaya, 2005, D’Amato and Gerber, 2005, Efionayi-Mäder, 2006, Ireland, 1994, Lavenex, 2006, Soysal, 1994). Thus, one might argue that the focus on national policies constitutes a rough simplification of subnational policy variety.

Considering ongoing political debates within Switzerland, opinions vary on whether the cantonal variety of integration policies is rather beneficial or detrimental. On the one hand, proponents of cantonal autonomy argue that adapted, context specific solutions for the local issue of immigrant integration are better than a “one size fits all” national framework (cf. Federal Council 2010: 32), or that Switzerland’s federalist laboratory facilitates the evolution of cantonal best practices. Opponents, on the other hand, contend that subnational policy variations constitute a potential source of structural discrimination (cf. Kübler and Piñeiro, 2010, Tripartite Agglomeration Conference TAK, 2005: 11), and that the heterogeneous puzzle of cantonal integration policies challenges the formulation of a coherent national strategy in the field.

Assessing subnational policy variety appears less delicate taking a scientific perspective, as the subnational analytical level provides clear benefits regarding research design. Subnational units offer the best approximation to a most similar cases setting (Przeworski and Teune, 1970). A similar degree of comparability is barely reachable in migration studies at the international comparative level, where more often than not a vast amount of control variables poses a serious methodological challenge (Fennema and Tillie, 2001: 38, IMISCOE, 2008). Thus, leading comparativists have long been demanding that the results of international comparative research should be verified at the subnational level (cf. Lijphart, 1971). Considering that Lijphart (1999: 38) classified Switzerland as one of the “most federal and decentralized countries worldwide”, it is hardly surprising that he explicitly recommends comparative subnational analyses for the Swiss case (Lijphart in Vatter, 2002: 14).

Drawing on the analytical benefits of the subnational comparative level, and paying tribute to Lijphart’s demands, the present study follows a twofold purpose: at an empirical level, we intent to capture subnational variations of integration policy in order to make them visible, using the example of Swiss cantons. In order to do so, we rely on an established international framework (Koopmans, 2010, Koopmans, et al., 2005) which we transfer to the subnational level. Thus, unlike the rather exploratory approach of a preceding study on the topic (Cattacin and Kaya, 2005), conceptualization and empirical measurement of cantonal integration policy used in this study draw directly on an established international approach. At the conceptual level, this transfer facilitates a quantitative and qualitative validation of the international concept. Considering that the validity-question has so far been largely neglected by authors and creators of integration policy indices alike, the suggestions offered in this study should also be read as an impulse to address this crucial topic more thoroughly.

The study is organized as follows. Section two elaborates on the two-dimensional conceptualization of integration regimes. In section three, this conceptual framework, which classifies integration policies in terms of immigrants’ ease or difficulty of access to comprehensive citizenship rights, is transferred to the cantonal level. Section four contains an exploratory factor analysis of the cantonal data, resulting in more subtle theoretical categories of immigrant rights and obligations than the two-dimensions of citizenship proposed by the international framework. The univariate illustration of the cantonal data in section five reveals that cantonal integration policies are shaped by traditional citizenship understandings of adjacent countries Germany and France, resulting in more restrictive policies in German speaking cantons than in their Latin counterparts. Finally, we discuss cantonal evidence as well as the results of the factor analysis in section six, and argue how the latter may be read as a suggestion of how to improve the theoretical coherence of the conceptual framework, and thus, its overall validity.

2. Two-dimensional typology of integration regimes

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices

Apart from a normative and target-oriented understanding of integration as equality of opportunity in the most important areas of integration (cf. art. 2, par. 1 VIntA; art. 4, par. 2 AuG), the Federal Council, as well as the Swiss parliament, refrain from further legal specifications of the term and its measurability (BFM, 2006: 8). Taking a scientific perspective, comparative international studies dealing with immigration and integration policies often refer to the concept of citizenship (Brubaker, 1992, Favell, 2001a, Koopmans, 2010, Koopmans and Statham, 2003, Koopmans, et al., 2005, Tilly, 1995). In this context, the meaning of citizenship is not restricted to the narrow traditional understanding referring merely to naturalization. Instead, citizenship is defined in an open and comprehensive manner, as an interaction of rights and obligations toward any given state, thereby creating an area of legal equality between native and new citizens (Brubaker, 1992: 21, Kleger and D’Amato, 1995: 260, Koopmans and Kriesi, 1997: 297, Tilly, 1995: 8). By focusing on legal equality, this broad definition of citizenship incorporates the aspect of equality of opportunities emphasized in Switzerland’s official integration debate. Furthermore, the concept of citizenship implies the feature of inclusion, and exclusion respectively, appearing as something exclusive from the outside, while it is inclusive from the inside (Eggert and Murigande, 2004: 127, Giugni and Passy, 1997: 13, Kleger and D’Amato, 1995: 260).2 Assuming that integration policies can be more or less inclusive or exclusive, variations of integration policy can therefore be captured in terms of varying degrees of ease or difficulty of immigrants’ access to broad citizenship rights and obligations.

In line with the traditional understanding of citizenship, earlier studies often applied a formal, one-dimensional perspective along the axis between the jus-soli and jus-sanguinis type of citizenship (Brubaker, 1992). Many Germanic countries such as Switzerland or until recently also Germany belong to the jus-sanguinis group, where citizenship is traditionally transmitted by inheritance (the “blood”). In such countries immigrants’ access to nationality is more difficult than in jus-soli countries such as France, where nationality is bound to the soil on which a person is born. In the course of more recent debates on multiculturalism and the rising importance attached to cultural rights, this formal legal understanding of citizenship has increasingly been criticized as being to reductive. Koopmans et al. (2005: 8) contend that Brubaker’s large neglect of the cultural rights dimension of citizenship causes him to overstate the “openness” of the French citizenship regime: while the French model may provide for easy formal access to citizenship, it couples this with the expectation that new citizens of migrant origin will assimilate to a unitary, national political culture. Accordingly, the authors suggest that the two dimensions, cultural group rights from the multiculturalism debate (see for example Kymlicka, 1999) and the formal criteria for individual access to citizenship, which is central to Brubaker’s analysis, should be combined (Koopmans, et al., 2005: 8).

This claim for a comprehensive conception is in the same vein with Marshall’s (1950) classical and influential formulation of citizenship, which postulates a progression from civic to political to social citizenship. Accordingly, several scholars base their concepts of integration policy on such complex notions of citizenship, which cover, besides the narrow civic aspect referring to the process of naturalization, political, social, economic, cultural or religious aspects of societal life.3

In the following, we will draw on the two-dimensional framework of integration regimes developed by Koopmans et al. (2005), more specifically its extended version, where the components of the first dimension are complemented by additional categories of the MIPEX (cf. Koopmans, 2010). The first dimension refers to Brubaker’s distinction and captures the aspect of equality of individual access. The second, cultural dimension captures cultural and religious rights attributed to immigrants as a group, as well as cultural obligations the state expects immigrants to meet in order to obtain full citizenship rights.

Figure 1 shows the resulting four ideal types of integration regimes based on the two dimensional typology, called assimilationist, segregationist, universalist and multicultural. By requiring a high degree of cultural assimilation with only scarce concessions to cultural pluralism, and by hampering the access to nationality through a jus sanguinis tradition, the assimilationist model can be seen as the most exclusive or restrictive type. Along with Germanic countries, Switzerland is considered strongly assimilationist (Kleger and D’Amato, 1995, Koopmans and Kriesi, 1997, Koopmans, et al., 2005, Skenderovic, 2009). The multicultural model can be read as the counterpart of the assimilationist model. It is prevalent in Anglo-Saxon and former settler states and represents the most inclusive and least restrictive type. In multicultural integration regimes immigrants enjoy considerable cultural or religious rights, and access to the nationality of the host country is comparatively easier.


Figure 1. : Two-dimensional typology of integration regimes. Sources: Own illustration, based on Koopmans (cf. Koopmans and Kriesi, 1997, Koopmans and Statham, 2003, Koopmans, et al., 2005).

Download figure to PowerPoint

Once more, figure 1 illustrates that integration regimes are commonly classified at the national level. As argued above, we assume that this national focus constitutes a rough simplification of subnational policy variety. Thus, behind the national classification of Switzerland as assimilationist we would rather expect a scatter of Swiss cantons in the lower left quadrant of figure 1.

3. Transfer of concept and measurement to the cantonal level

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices

The conceptual framework elaborated by Koopmans and colleagues will serve as the frame of reference for our cantonal measurement of integration policy. The concept exhibits the typical complex, multidimensional structure which Goertz (2006) attributes to most social science concepts. Thereby, the basic concept integration policy is conceptualized along two dimensions, namely individual equality and cultural difference, which are captured by several components, which in turn comprise various indicators and subindicators. Table 1 offers an overview on all components of the two dimensions.

Table 1.   Measuring integration policy along the individual and cultural dimensions of citizenship
Koopmans et al. (2005)MIPEX II (Koopmans (2010))
  1. Notes:* = components exhibiting variation at the cantonal level and accordingly included in the measurement of cantonal integration policies. Many components in the cultural dimension are not applicable at the cantonal level. However, this dimension has been amended by an additional component measuring the “tendency for legal recognition of minorities’ religions” in the cantons.

Individual equalityAccess to nationality*yesyes
Political participation*   partlyyes
Labour market access*yes
Family reunion*yes
Cultural differenceCultural requirements For naturalization*yes
Religious rights outside public institutions*yes
Cultural rights in public institutionsyes
Political representation rights (specific cultural groups)yes
Group specific affirmative action (labour market)yes

The components of the first dimension address the question to what extent access to citizenship is colour-blind, meaning to which extent individuals, regardless of their race, ethnicity or cultural background, have equal access to these rights (Koopmans, et al., 2005: 34). According to Koopmans et al.’s original concept from 2005, this dimension captures the following components: access to nationality, civic political rights, and anti-discrimination. The components of this dimension correspond largely to the categories of the MIPEX. Yet, the MIPEX additionally includes crucial aspects of integration policy such as family reunion or access to the labour market which are not accounted for by Koopmans et al. (2005). Thus, in his more recent work, Koopmans (2010: 5) himself refers to the MIPEX in order to measure the individual equality dimension. Therefore, and in line with Koopmans (2010), we complement the components of the individual equality dimension from Koopmans et al.’s (2005) original operationalization of this dimension by additional aspects of the MIPEX such as labour market access and family reunion. Furthermore, we include the indicator “immigrant commission” from the MIPEX to the component political participation, since we consider it, besides immigrant voting rights, a relevant indicator for immigrants’ political participation and representation.4 By contrast, we did not include the MIPEX category long term residence, as this is the only category which does not vary at the cantonal level. The last two columns in Table 1 illustrate, which components are present in both data sources and which are additionally adopted from the MIPEX (Koopmans 2010).

Basically, the components of the individual equality dimension cover the aspects of citizenship delineated by Marshall (1950): access to nationality refers to the narrow civic aspect of citizenship, whereas its political aspect is captured by the component political participation. Certain scholars may argue that access to the labour market captures Marshall’s social aspect of citizenship. Castles and Davidson (2000: 110), for instance, consider the right to work and equal opportunities in the labour market a social right. Others expanded Marshall’s trichotomy by specific economic rights in order to account for this aspect (i.e. Mackert, 1999). The case seems less clear for the remaining components, family reunion and anti-discrimination, which might also constitute elements of social citizenship. However, as our empirical validation in the following section suggests, they might also be considered cultural rights or obligations.

Koopmans et al.’s (2005) cultural difference dimension, by contrast, is not accounted for by the MIPEX (cf. Table 1). An interpretation of the single components of this dimension is less straightforward, as it becomes only evident in the light of the different integration regimes (cf. figure 1). The first component, cultural requirements for naturalization, corresponds to the assimilationist demand for cultural adaptation. The two following categories, religious rights outside public institutions, and cultural rights in public institutions, stand for poly-ethnic rights, which are favoured by multiculturalists, and generally opposed by assimilationists (Koopmans, et al., 2005: 51). The distinction between cultural and religious rights inside versus outside public institutions, finally, is necessary to capture the nature of the universalist integration regime. The strict neutrality of the state regarding cultural and religious practices of its citizens, demanded by the universalist regime, implies that the state should tolerate such practices outside public institutions, for instance Muslim cemeteries, whereas it should reject these practices within public institutions such as the education system and public broadcasting (Koopmans, et al., 2005: 52).

Together with the remaining two components of the cultural rights dimension, political representation rights and affirmative action in the labour market, cultural rights inside public institutions relate to group specific rights which do not exist in Switzerland. The asterisks after the components in Table 1 thus indicate that there are many equivalents for measuring the individual dimension of citizenship at the cantonal level, while there are only few counterparts for the cultural dimension. This is not surprising, considering Switzerland’s assimilationist tradition with scarce concessions to cultural pluralism.

In the remainder of this section, the transfer of the single components to the subnational level will be discussed. The time span covered by our data goes from 2004 to 2008. The reason why we consider a time span instead of one single point in time is that not all indicators are available for one particular year. However, a look at single indicators reveals that they represent rather stable policy patterns over time. We will discuss each component, its indicators and subindicators. Since there is not enough space for a detailed discussion of the coding of all subindicators, we will provide coding examples only for selected subindicators. Additionally, Table 4 in the appendix contains the details on operationalization and coding for all subindicators. Whenever possible, we preferred a more differentiated coding than the dummy solution in order to capture the variation on the subindicators at the most subtle level (cf. Goertz, 2006), whereas in most of the cases values between 0 (restrictive pole, high requirements and few rights) and 1 (liberal pole, extensive rights and minimal requirements) were attributed. For additive index creation, we aggregated the standardized values step by step to the next higher level, i.e. from subindicators to indicator, from indicators to component and so on up to the comprehensive integration policy index level.

To varying degrees, all components of the individual equality dimension could be measured at the cantonal level. Some components such as anti-discrimination measure the mere presence or absence of a respective clause in cantonal laws or constitutions. We attributed the value 0.75 if anti-discrimination was covered by one of the legal sources and 1 for both. Other components such as access to nationality or political participation were captured by additive indices based on a broader set of indicators or subindicators. For the component access to nationality we considered cantonal requirements regarding fee, period of residence, facilitated procedure, and right of appeal, whereas the component political participation embraces cantonal provisions regarding immigrant voting rights (active/passive, cantonal / local, additional requirements regarding period of residence and residence permit) and the cantonal provision of an immigrants’ commission.

Supposedly, the component access to the labour market can only be captured to a limited extent at the cantonal level, as in Switzerland, this access is broadly regulated at the national level.5 Some authors take this assumption as a justification to exclude this aspect from their measurement of cantonal integration policy (Cattacin and Kaya 2005: 293). We do not agree with this assessment, as there is an area of employment where cantons vary considerably regarding labour market access which should not be neglected: when it comes to cantonal employment, such as the cantonal administration, the police service, the judiciary or teaching at public schools, cantons are the main regulating instances, which allows for a measurement of the respective variances. Unrestricted access for immigrants to an employment in the cantonal administration is for instance provided in Bern, which we coded 1. Several cantons restrict this access when it comes to specific official positions (“Ausübung hoheitlicher Funktionen”), such as Zürich (coded 0.75). Even more restrictive regulations can be found in Ticino (TI). In this canton, immigrants’ access to employment in the public administration is basically denied, whereas the cantonal government decides on exceptions to this rule (coded 0.5). In a similar vein, we measured access to teaching positions, the cantonal police service and the judiciary (source: EKM, 2007b, own survey in cantonal migration offices). Access to the labor market thus constitutes an additive index based on the values of these four subindicators. The component family reunion was finally covered by the varying extent of facilitation for EU-nationals and requirements regarding the housing situation of the applicant (cf. Achermann, 2004).

Although only weakly developed in the Swiss case, the cultural dimension can be amended by an additional legal component: the tendency for legal recognition of minorities’ religions, which may be considered of particular interest for Switzerland’s growing Muslim community (Cattacin, et al., 2003). We adopted the coding for this component from Christmann (2010), who attributed values from 0 to 1, with 1 as the most liberal cantonal predisposition in this respect. Considering that Islam is Switzerland’s second biggest religion after the Christian confessions (Hunter, 2002, Mahnig, 2000), the component religious rights outside public institutions will, similarly to Koopmans et al. (Koopmans, 2010,2005), be defined in terms of Muslim specific rights. Koopmans and colleagues consider the allowance of ritual slaughtering according to the Islamic rite, allowance of Islamic prayer in public, and provisions for Muslim burials (Koopmans, et al., 2005: 55). In Switzerland, only Muslim cemeteries exist, and therefore this indicator has been used.

Finally, we used two indicators to measure cultural requirements for naturalization: the requirement for cultural integration or absence thereof, as it is outlined in cantonal citizenship laws, as well as the cantons use and implementation of integration agreements. The first indicator, requirement for cultural integration, was assessed in a relative manner, i.e. compared to the national level. The national citizenship law requires applicants to comply with four conditions in order to be eligible for Swiss citizenship: they need to be integrated into the Swiss community, be acquainted with the Swiss way of life, norms and customs, should respect the legal order and must not pose a threat for the internal or external security of the country (art. 14, BüG). To capture cantonal variations, we classified cantonal requirements exceeding the national prescription, for instance regarding language knowledge or moral disposition of the applicant, with 0, less demanding requirements at the cantonal level with 1, and cantonal requirements equal to the national with 0.5.

While Swiss cantons provide general political representation rights for immigrants (captured by the component political participation on the individual equality dimension), no such rights are allocated to specific cultural groups, so this component was discarded. Similarly, no group specific affirmative action programmes exist in the cantons and cultural rights in public institutions are so far very weakly developed,6 and thus the respective components are irrelevant for the time period considered here. Table 2 presents the cantonal data for all components or indicators. As there is not enough space here to show the values for all subindicators, we decided to display the values for the first aggregated level, which corresponds to the indicator or component level, depending on the complexity of the measurement (detailed information on index formation, indicators and subindicators can be found in Table 4 in the appendix).

Table 2.   Values for the first aggregation level (indicators/components)
Indicators/componentsAccess to nationality*Right to vote*Immigrant commiss.*Anti-discriminationLabour market access*Family reunion*Islamic cemetery*Legal recognitionCultural obligations*
  1. Note: time span covered: 2004–2008. * indicates that indicator / component itself represents an additive index based on subindicators.


The list of indicators included in our measurement is far from exhaustive. Several additional aspects of integration policy such as financial indicators (cantonal integration budgets) or the existence of cantonal integration laws, constitutional articles or specific cantonal guiding principles for integration, play a role at the cantonal level.7 There are two reasons why we exclude these aspects from our measurement: first, purpose of this study is to validate an international instrument at the subnational level, which requires the highest possible degree of conceptual consistency with the international framework. If we include a range of additional aspects in our measurement, we would hamper this comparability. Second, we are particularly interested in subnational policy variations, which can be observed by focusing on certain aspects and does not require an exhaustive measurement of these policies.

This is not to say that these additional factors are not relevant, on the contrary: certain indicators such as cantonal integration budgets are probably much more decisive for cantonal integration policies than the provision of specific minority rights such as Islamic cemeteries. In order to test whether these additional aspects are in line with the general orientation of cantonal integration policies captured by our index or not, we considered bivariate correlations (operationalization, coding and sources for the two additional factors are reported in Table 4 in the appendix). It turns out that additional integration provisions (e.g. legislation and guiding principles) correspond to the general integration policy strategy of a canton, as they are positively related to our cantonal integration policy index, with Pearson’s r amounting to 0.57 significant at the one percent level. Thus, cantons with more liberal or multiculturalist policies are also more likely to foster integration by means of additional laws or guiding principles. By contrast, there is only a weak correlation between cantonal integration policies and integration budgets (Pearson’s r = 0.3, not significant), which for obvious reasons are rather related to the degree of urbanization (Pearson’s r = 0.57, significant at the one per cent level), and even stronger to a canton’s immigrant share (Pearson’s r = 0.63, significant at the 0.1 per cent level).

4. Concept validation: four categories instead of two dimensions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices

As should have become clear from the preceding elaboration, Koopmans et al.’s framework constitutes a very comprehensive attempt to measure integration policy, accounting for the most essential aspects of integration without being too complex. Yet a closer look at the theoretical conceptualization of this framework reveals certain inconsistencies, which we will address in the following. Most important, the conceptualization of the single components along two dimensions is not completely convincing. The components of the individual dimension cover such different issues as socio-structural (labour market access), political (political participation) or civic aspects (access to nationality) of integration. Not surprisingly, alternative conceptualizations suggest treating these aspects as separate dimensions of integration policy. Entzinger (2000), for instance, distinguishes three dimensions of integration policy, which he calls political / legal, cultural, and social / economic.

Equally, the cultural dimension comprises a mix of cultural and religious elements. Furthermore, the conceptualization of the cultural dimension as group sensitive appears inconsistent: while this applies to the aspect of cultural or religious rights, cultural obligations, i.e. the component cultural requirements for naturalization, are addressed to immigrants in general and may therefore not be considered group specific. The same applies for the component affirmative action, which is not in the same way group specific as, for instance, religious rights, as Koopmans et al. (2005: 68) correctly argue. In this respect, the individual level dimension is more consistent as all of its components operate at the level of individual equality.

At a more general level, one might finally question the use of the term citizenship in order to refer to integration regimes, as suggested by Koopmans et al. (Koopmans, 2010, 2005). The fact that many scholars share the broad understanding of citizenship applied by Koopmans and colleagues does not prevent that the term is still strongly associated to its narrow traditional understanding referring to naturalization. Thus, the question arises, whether it is indeed appropriate or necessary to talk of dimensions of citizenship in the context of integration regimes, considering that this terminology might be misleading.

In search of a more solid theoretical base and conceptualization of the measurement for cantonal integration policies, we carried out a first inspection of the data by means of exploratory factor analysis. Therefore, and since most, but not all indicators have scales from 0 to 1 (cf. Table 4 in the appendix), data for all components have been z-transformed (cf. Backhaus, et al., 2008: 332). As expected, the factor analytical results for the single components revealed a pattern which does not correspond to the two-dimensional model presented in Table 1.

Instead, four factors resulted from the analysis presented in Table 3. The factor loadings suggest that the components measure more subtle theoretical dimensions. A closer look at the scores of the components on the single factors reveals that political participation, religious rights I and labour market access load high on their respective factors, suggesting that these components indeed represent different aspects of integration policy. Slightly lower are the scores for access to nationality and religious rights II. However, intuitively, their highest loading on the first and second factor respectively seems reasonable. Cultural requirements for naturalization load on a fourth factor, implying that they are different from religious rights.

Table 3.   Factor analysis based on single components
  1. Note: Principal factor analysis, varimax-rotated (orthogonal). Extraction of factors according to scree-test. Religious rights I stands for the “tendency for legal recognition of minorities’ religions”, religious rights II for the “allowance for Islamic burials”.

Political participation0.857−0.1110.0370.114
Access to nationality0.5980.3620.024−0.314
Religious rights I0.1850.679−0.045−0.065
Religious rights II0.2570.492−0.410−0.014
Labour market access0.104−0.0120.7100.041
Cultural requirements for natural.0.2630.1340.2860.540
Family reunion0.288−0.0020.0740.476
Explained variance43.8%28.5%22.8%21.3%

Probably most surprising are the loadings of anti-discrimination and family reunion on the fourth, cultural factor. However, one has to note that the factor scores for this last two components are low, and thus, not very trustworthy. Nevertheless, one could still find arguments to interpret this pattern: Bauböck (1996: 230), for instance, considers anti-discrimination a cultural minority right. A closer look at existent legal provisions on the topic reveals that anti-discrimination is mostly defined in group specific terms, i.e. as the prohibition of any discriminatory acts based on race, origin, gender, religious background etc. (cf. art. 8, par. 2 BV; or art. 11, par. 2 constitution of canton Zürich), which would allow for a group specific interpretation of this component. The elevated loading of the family reunion component on the cultural factor, finally, might be caused by our group sensitive operationalization of this variable, since we distinguish between different groups of nationalities (EU versus third country nationals).

The weak loadings of single factors point to the limits of our quantitative validation by means of exploratory factor analysis based on only 26 cases. In spite of this limitation, there is a clear picture emerging from our factor analytical results favouring a more subtle theoretical categorization of integration policies. Therefore, we suggest capturing single aspects of integration policy along substantive or thematic lines, and in terms of categories of rights or, as in the case of the cultural component, rights and obligations, instead of two abstract dimensions. The thematic categories of political, socio-structural, cultural and religious rights correspond more closely to areas of integration which are considered central in the official debate on integration (cf. BFM, 2006, Cattacin and Kaya, 2005, TAK, 2009). Furthermore, they match alternative scientific conceptualizations which distinguish between social / economic, cultural and political /legal aspects of integration policy (cf. Entzinger, 2000, Henkes, 2008).

In line with these theoretical and empirical considerations, we aggregated the components along their highest factor loadings, creating thereby the following categories8: Factor one stands for political and civic rights, factor two represents religious rights, factor three denotes socio-structural rights, and factor four comprises cultural rights and obligations. In a next step, the four categories were aggregated to a comprehensive additive index, measuring cantonal integration policies.

5. Empirical results: testing construct validity

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices

The present section contains some descriptive illustration of the empirical evidence found in the cantonal data on integration policies. The uni- and bivariate evidence provided here facilitates a second type of validity test. While the exploratory factor analysis presented above contributed to enhance the concept validity, i.e. the consistency of the theoretical categories underlying the complex concept of integration policy, a look at the empirical evidence allows for a kind of construct validity test. We speak of construct validity, if a concept relates to other observable constructs in a way that is consistent with theoretically derived predictions (Schnell, et al., 1999: 150).

In this case, the theoretical prediction would be related to the cultural linguistic background of Swiss cantons, and their respective understanding of citizenship and belonging. More specifically, German-speaking cantons are assumed to be influenced by Germany’s jus sanguinis tradition, exhibiting thereby more restrictive integration policies than French-speaking cantons, which are in turn expected to be influenced by Frances more inclusive jus soli understanding of citizenship (cf. Figure 1). Such a transboarder shaping of subnational, regional citizenship cultures is not only assumed to affect cantonal integration policies (cf. Cattacin and Kaya, 2005: 290, D’Amato, 2010: 143). Even more prominently it reflects in the larger public opinion, for instance people’s voting behaviour. French speaking cantons vote generally in a more liberal and open minded way when it comes to foreign or social policy than German speaking cantons (Kriesi, et al., 1996), revealing a linguistic divide of mentalities which is better known as “Röstigraben”. While this pattern has been particularly salient in national votes regarding the European Union (Linder, 2005: 90), it is also observed in votes on citizenship issues and immigrant rights (Danaci, 2009).

Does this regional understanding of citizenship also reflect in cantonal integration policies? Our data lend strong support to this assumption. Figure 2 offers a first inspection of the univariate distribution of the integration policy index. Negative numbers denote restrictive policies, while positive numbers denote more permissive policies towards cultural pluralism. At first glance, a clustering of negative values for central, mainly rural, German speaking cantons and a similar clustering of positive values for French speaking cantons (FR, VD, NE, GE, JU) plus Italian speaking Ticino (TI) are striking. The only French speaking canton below the mean value is Valais (VS), which is also geographically closest to central German speaking cantons. Thus, the evidence found in the data on cantonal integration policies suggests that the linguistic divide is not restricted to the individual behavioural level, but it even shaped cantonal policies.


Figure 2.  Cantonal integration policies, comprehensive index. Note: Entries are z-transformed values.

Download figure to PowerPoint

In order to compare the cantonal variations with Koopmans international typology (cf. figure 1), a two-dimensional illustration of the data is presented in figure 3. Yet, instead of Koopmans individual and cultural dimensions, the respective modified categories, civic-political rights and cultural rights and obligations will be applied here, keeping in mind that they measure similar aspects but in a more consistent and narrowed way. Compared to figure 1, where Switzerland was located uniformly in the assimilationist quadrant, figure 3 reveals considerable variations in the cantons along the two categories of rights. Even more pronounced than in figure 2, a “Latin cluster” in the upper right quadrant denoting the more permissive corner is discernible. Again, rural German speaking cantons lie closest to the lower left quadrant standing for the restrictive and assimilationist pole of citizenship.


Figure 3.  Scatter of Swiss cantons along civic-political and cultural rights and obligations categories. Note: Entries are z-transformed values.

Download figure to PowerPoint

A look at the scatter for our two remaining policy categories, socio-structural and religious rights, still reveals traces of a linguistic pattern, particularly on the socio-structural rights category, although it is less pronounced (see figure 4 in the appendix). Only for the religious rights category, finally, we cannot claim that French speaking cantons have more liberal policies than German-speaking cantons.

The linguistic pattern, which we could observe at the comprehensive integration policy index and for three out of four policy categories, is furthermore corroborated by multivariate analyses accounting for several control variables such as right populism, foreign population and urbanization. The respective cross-sectional OLS regression analyses show that the cultural-linguistic background of a canton, and related to this factor, citizens’ attitudes towards immigrants, turn out to be the most important predictors of cantonal integration policies (Manatschal, forthcoming).

6. Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices

While subnational varieties of integration policy remain largely neglected by the scientific literature, the present study shows that a closer look at the policy variety below the national level can be quite revealing. Following Lijphart’s invitation to verify the results of international comparative research at the subnational level, we transferred an international framework to the Swiss cantons in order to capture varying cantonal integration policies. Our cantonal data not only corroborate the assumption that subnational variations of integration policy are particularly pronounced in strongly federal Switzerland. As suggested by Lijphart, they furthermore lend themselves to verify results of international research at the subnational level. The verification facilitated by the empirical evidence presented in this study is twofold: first, at a conceptual level, regarding the consistency of an internationally established framework, and second, at an empirical level, regarding the determinants of cantonal integration policy.

As for the conceptual level, the results of the exploratory factor analysis suggest that the single components measure more subtle theoretical categories than the rather vague two dimensions of citizenship used by Koopmans et al. (2010, 2005). The four substantive categories of immigrant rights and obligations we derive from the factor analytical results – civic-political, socio-structural, cultural and religious – correspond to areas which are commonly considered crucial for integration in the scientific as well as in the official political debate on integration. Finally, the more neutral terminology of immigrant rights and obligations might be less ambiguous than the term citizenship dimension and prevents us from a potential over-extension of the meaning of citizenship.

At the same time, the low factor loadings of single components clearly revealed the limits of our quantitative validation based on only 26 cases, and in a subnational setting, where cultural and religious rights are only weakly represented. It seems not surprising, thus, that the lowest factor loadings could be found on these two categories. While the small number of 26 cases limits any generalization of the results of the factor analysis beyond the Swiss case, they should nevertheless be read as a suggestion on how the conceptual framework of Koopmans et al. (2005) could be refined, both, at inter- and subnational, or even local levels.

There is a small but increasing number of studies analysing outcomes of integration or citizenship policies comparing policies of cities (cf. Cinalli and Giugni, 2011, Fennema and Tillie, 2001, Fennema and Tillie, 2004, Ireland, 1994) or also municipalities (cf. Helbling, 2008). While the small size of Swiss cantons, which are furthermore considered crucial actors in Swiss integration policy (Art. 57 AuG), justifies an analysis at the regional level for the Swiss case, it might particularly in larger federal states than Switzerland be reasonable to transfer the international concept not only to the regional, but also to the local level.

At an empirical level, the evidence emerging from our cantonal data points to potential determinants of the subnational policy variety, allowing us to test the construct validity of our measurement. A look at the uni- and bivariate distribution of the cantonal values in figures 2 to 4 highlight the significance of the cultural-linguistic background of a canton. According to this pattern, which is furthermore confirmed by multivariate regression analyses, German-speaking cantons, influenced by Germany’s jus sanguinis tradition, exhibit more restrictive integration policies than French-speaking cantons, which are in turn influenced by Frances more inclusive jus soli understanding of citizenship. Our newly developed categories allow for an even more subtle construct validity test, not only at the comprehensive index level, but also at the categorical level: Following the outlined theoretical reasoning, we would expect a similar nation-specific impact regarding religious traditions, i.e. France’s strict separation of state and church called laïcité (Favell, 2001b: 75, Fetzer and Soper, 2005: 127, Koopmans, et al., 2005: 168–173). Our data as well as multivariate regression analyses not reported here support this assumption: the religious category is the only category where French speaking cantons do not have more liberal policies than German speaking cantons (cf. figure 4 in the appendix).

An analogous cultural-linguistic pattern has been observed in another multilingual country, Belgium, where Walloon and Flemish integration policies are said to be influenced by French and Dutch understandings of citizenship (Favell, 2001a: 382, Ireland, 2006: 146ff, Koopmans, 2010: 6). These findings verify results of international research inasmuch as they confirm the scholarly consensus that integration policies are path-dependent. According to this line of thought, integration policies are considered stable policy frameworks which are shaped by traditional notions of citizenship and belonging (Brubaker, 1992, Favell, 2001b, Ireland, 1994). However, the fact that we observe a transborder impact of French and German citizenship traditions at Switzerland’s subnational level further questions the focus of migration scholars, which remains restricted to the national level.

What are the implications of our findings for future analyses based on this kind of data? Or more specifically: on which level, the categorical or the comprehensive index level of integration policy, should comparative analyses operate? There is no categorical answer to this question, as it depends purely on the research question and thus, theoretical considerations. The empirical evidence presented in this study suggests that when it comes to explain integration policy as an output, analyses on both levels may be revealing. By contrast, when analysing outcomes of integration policy single policy categories may yield differing and even contrasting outcomes which neutralize when aggregated to the comprehensive index level.9 As a consequence, analyses at the level of clear cut categories are generally preferable, as they allow testing more concise hypotheses resulting in more subtle empirical findings.

  • 2

    Similarly, Helbling (2008), based on Weber’s (1978) idea of “open and closed social relationships”, uses the expression of citizenship as a “closure mechanism”.

  • 3
  • 4

    Koopmans et al. (2005) also consider political representation rights, but only for specific cultural groups (cultural difference dimension, cf. Table 1 below). In Switzerland, there are no group specific immigrant commissions but they address all immigrants equally. Therefore, and similar to the MIPEX, we consider them an instrument for political participation (individual equality dimension).

  • 5

    The most relevant legal document being here the new aliens’ law AUG, particularly chapter five “Zulassung zu einem Aufenthalt mit Erwerbstätigkeit”, articles 20 to 25.

  • 6

    Such as Islamic religious classes in state schools, the right of female teachers to wear the Islamic headscarf or Islamic religious programmes in public broadcasting (Koopmans et al. 2005:57; 2010:5f).

  • 7

    Respective articles in cantonal constitutions can be found in BL, BS, FR, GL, JU, NE, SG, SO, SH, VD and ZH. Integration laws are in force in GE, NE, VD, BS and BL. The following cantons and cities drafted own integration guiding principles (“Integrationsleitbilder”): AG, AR, BS, LU, OW, SG, SH, SO, TI, VS and the cities of Bern, Biel, Burdorf, Thun, Lausanne, Luzern , St.Gallen, Winterthur, Zürich and Wil. Source: (EKM, 2007a); own investigations in cantonal constitutions and legal texts (reference year: 2008).

  • 8

    The results of the factor analysis did not prove to be stable enough for an index-creation based on factor scores. Instead, an additive index-creation has been applied.

  • 9

    Empirical evidence supporting the assumption of contrasting effects of single policy categories has been found in multilevel analyses regarding the impact of cantonal integration policies on immigrant civic engagement (results available upon request).


  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices
  • Achermann, A. (2004). Integration und Habitat: Die “angemessene Wohnung” als Voraussetzung für den Familiennachzug. Bern: EKA.
  • Akgün, L. and D. Tränhardt (2001). Integrationspolitik in föderalistischen Systemen. Münster: Lit.
  • Backhaus, K., B. Erichson, W. Plinke and R. Weiber (2008). Multivariate Analysemethoden. Eine anwendungsorientierte Einführung, 12. Auflage ed. Berlin: Springer.
  • Bauböck, R. (1996). Cultural Minority Rights for Immigrants. International Migration Review 30(1): 203250.
  • BFM (2006). Probleme der Integration von Ausländerinnen und Ausländern in der Schweiz. Bestandesaufnahme der Fakten, Ursachen, Risikogruppen, Massnahmen und des integration spolitischen Handlungsbedarfs. Bern: Bundesamt für Migration im Auftrag des Departementvorstehers EJPD.
  • BFM (2008). Anwendung der Integrationsvereinbarung 2008 und 2009 in den Kantonen. Bern-Wabern: Bundesamt für Migration.
  • Brubaker, R. (1992). Citizenship and nationhood in France and Germany. Cambridge, Mass. etc.: Harvard University Press.
  • Bundesrat (2010). Bericht zur Weiterentwicklung der Integrationspolitik des Bundes. Bern.
  • Castles, S. and A. Davidson (2000). Citizenship and migration. Globalization and the politics of belonging. New York: Routledge.
  • Cattacin, S. (1996). Il federalismo integrativo. In I come...identità, integrazione, interculturalità. Zürich: Federazione delle Colonie Libere Italiane in Svizzera (6782).
  • Cattacin, S., C. R. Famos, M. Duttwiler and H. Mahnig (2003). Staat und Religion in der Schweiz – Anerkennungskämpfe, Anerkennungsformen, Studie im Auftrag des Schweizerischen Forums für Migrations- und Bevölkerungsstudien (SFM) im Auftrag der Eidgenössischen Kommission gegen Rassismus (EKR). Bern.
  • Cattacin, S. and B. Kaya (2005). Le développement des mesures d’intégration de la population migrante sur le plan local en Suisse. In Histoire de la politique de migration, d’asile et d’intégration en Suisse depuis 1948. Zürich: Seismo (288320).
  • Christmann, A. (2010). Damoklesschwert Referendum? Die indirekte Wirkung ausgebauter Volksrechte auf die Rechte religiöser Minderheiten. Swiss Political Science Review 16(1): 141.
  • Cinalli, M. and M. Giugni (2011). Institutional Opportunities, Discursive Opportunities and the Political Participation of Migrants in European Cities. In Social Capital, Political Participation and Migration in Europe. Making Multicultural Democracy Work? Houndmills: Palgrave (4362).
  • D’Amato, G. (2010). Switzerland: a multicultural country without multicultural policies? In The multiculturalism Backlash. European discourses, policies and practices. London and New York: Routledge (130151).
  • D’Amato, G. and B. Gerber (2005). Herausforderung Integration. Städtische Migrationspolitik in der Schweiz und in Europa. Zürich: Seismo.
  • Danaci, D. (2009). Uncovering Cross-National Influences and Within-Nation Heterogeneity of National Identity, Working Paper. University of Bern.
  • Efionayi-Mäder, D. (2006). L’immigration choisie. terra cognita 9: 4447.
  • Eggert, N. and A. Murigande (2004). Modèles de citoyenneté et mobilisation politique des migrants en Suisse: Le rôle des cantons. Swiss Political Science Review 10(4): 125145.
  • EKM (2007a). Jahresbericht. Bern: Eidgenössische Kommission für Migrationsfragen.
  • EKM (2007b). Mitreden und Mitgestalten. Strukturelle Partizipation in den Kantonen. Bern: Eidgenössische Kommission für Migrationsfragen.
  • Entzinger, H. (2000). The Dynamics of Integration Policies: A multidimensional Model. In Challenging Immigration and Ethnic Relations Politics: Comparative European Perspectives. Oxford: Oxford University Press (97118).
  • Favell, A. (1998). Philosophies of integration. Immigration and the idea of citizenship in France and Britain. Basingstoke etc.: Macmillan etc.
  • Favell, A. (2001a). Integration Policy and Integration Research in Europe: A Review and Critique. In Citizenship today: global perspectives and practices. Washington D.C.: Carnegie Endowment for International Peace (349399).
  • Favell, A. (2001b). Philosophies of integration. Immigration and the idea of citizenship in France and Britain. Basingstoke etc.: Macmillan etc.
  • Fennema, M. and J. Tillie (2001). Civic Community, Political Participation and Political Trust of Ethnic Groups. Connections 24(1): 2641.
  • Fennema, M. and J. Tillie (2004). Do immigrant policies Matter? Ethnic Civic Communities and Immigrant Policies in Amsterdam, Liège and Zurich. In Citizenship in European Cities. Immigrants, Local Politics and Integration Policies. Aldershot: Ashgate (85106).
  • Fetzer, J. S. and J. C. Soper (2005). Muslims and the State in Britain, France and Germany. Cambridge: Cambridge University Press.
  • Giugni, M. and F. Passy (1997). Histoires de mobilisation politique en Suisse: De la contestation à l’intégration. Paris: L’Harmattan.
  • Goertz, G. (2006). Social Science Concepts. A User’s Guide. Princeton, N.J.: Princeton University Press.
  • Helbling, M. (2008). Practising citizenship and Heterogeneous Nationhood. Naturalisations in Swiss Municipalities. Amsterdam: Amsterdam University Press.
  • Henkes, C. (2008). Integrationspolitik in den Bundesländern? In Die Politik der Bundesländer. Staatstätigkeit im Vergleich Wiesbaden: VS Verlag für Sozialwissenschaften (113145).
  • Howard, M. M. (2006). Comparative Citizenship: An Agenda for Cross-National Research. Perspectives on Politics 3(4): 443455.
  • Hunter, S. T. (2002). Islam, Europe’s second religion. The new social, cultural, and political landscape. Westport, Conn.: Praeger.
  • IMISCOE (2008). The Future for Migration Research in Europe, IMISCOE Policy Brief .
  • Ireland, P. (1994). The policy challenge of ethnic diversity. Immigrant politics in France and Switzerland. Cambridge, Massachusetts etc.: Harvard University Press.
  • Ireland, P. (2006). Institutions, Political Opportunity Structures, and the Shaping of Migration Policies in Western Europe. In Dialogues on Migration Policy. Lanham: Lexington Books (137155).
  • Kleger, H. and G. D’Amato (1995). Staatsbürgerschaft und Einbürgerung - oder: Wer ist ein Bürger? Ein Vergleich zwischen Deutschland, Frankreich und der Schweiz. Journal für Sozialforschung 35 (3/4): 259281.
  • Koopmans, R. (2010). Tradeoffs between Equality and Difference: Immigrant Integration, Multiculturalism and the Welfare State in Cross-National Perspective. Journal of Ethnic and Migration Studies 36(1): 126.
  • Koopmans, R. and H. Kriesi (1997). Citoyenneté, identité nationale et mobilisation de l’extrême droite. Une comparaison entre la France, l’Allemagne, les Pays-Bas et la Suisse. In sociologie des nationalismes. Paris: Presses Universitaires de France (205324).
  • Koopmans, R. and P. Statham (2003). Challenging Immigration and Ethnic Relations Politics. Comparative European Perspectives. Oxford: Oxford University Press.
  • Koopmans, R., P. Statham, M. Giugni and F. Passy (2005). Contested citizenship. Immigration and Cultural Diversity in Europe. Minneapolis; London: University of Minnesota Press.
  • Kriesi, H., B. Wernli, P. Sciarini and M. Gianni (1996). Le clivage linguistique. Bern: Bundesamt für Statistik.
  • Kübler, D. and E. Piñeiro (2010). Integrationspolitik des Zufalls, Neue Zürcher Zeitung.
  • Kymlicka, W. (1999). Multikulturalismus und Demokratie. Über Minderheiten in Staaten und Nationen. Hamburg: Rotbuch Verlag.
  • Lavenex, S. (2006). Einwanderungs-, Asyl-, Integrations- und Einbürgerungspolitik. In Handbuch der Schweizer Politik. Zürich: Verlag Neue Zürcher Zeitung (653676).
  • Lijphart, A. (1971). Comparative Politics and the Comparative Method. American Political Science Review 65(3): 682693.
  • Lijphart, A. (1999). Patterns of democracy: government forms an performance in thirty-six countries. New Haven: Yale University Press.
  • Linder, W. (2005). Schweizerische Demokratie. Institutionen, Prozesse, Perspektiven, 2., vollst. überarb. und aktualis. Aufl. ed. Bern: Haupt Verlag.
  • Mackert, J. (1999). Kampf um Zugehörigkeit. Nationale Staatsbürgerschaft als Modus sozialer Schliessung. Opladen: Westdeutscher Verlag.
  • Mahnig, H. (2000). L’intégration institutionelle des Musulmans en Suisse: l’exemple de Bâle-Ville, Berne, Genève, Neuchâtel et Zurich. TANGRAM, Bulletin der Eidgenössischen Kommission gegen Rassismus 8: 102111.
  • Manatschal, A. (forthcoming). Path Dependent or Dynamic? Cantonal Integration Policies between Regional Citizenship Traditions and Right Populist Party Politics. Ethnic and Racial Studies.
  • Marshall, T. H. (1950). Citizenship and social class and other essays. Cambridge.
  • Penninx, R. (2005). Integration of Immigrants in Europe: Policies of Diversity and Diversity of Policies. In Herausforderung Integration: Städtische Migrationspolitik in der Schwez und in Europa. Zürich: Seismo Verlag (4555).
  • Przeworski, A. and H. Teune (1970). The logic of comparative social inquiry. New York: Wiley-Interscience.
  • Schmidtke, O. (2001). Zwischen Standortpolitik und regionalem Nationalismus: Immigrations- und Integrationspolitik in den kanadischen Provinzen. In Integrationspolitik in föderalistischen Systemen. Münster: Lit (219247).
  • Schmitter Heisler, B. (2001). Immigration and Federalism in the United States. In Integrationspolitik in föderalistischen Systemen. Münster: Lit (151198).
  • Schnell, R., P. B. Hill and E. Esser (1999). Methoden der empirischen Sozialforschung, 6 ed. München, Wien: Oldenburg Verlag.
  • Skenderovic, D. (2009). The Radical Right in Switzerland: Continuity and Change, 1945-2000 New York: Berghahn Books.
  • Soysal, Y. N. (1994). Limits of citizenship. Migrants and postnational membership in Europe. Chicago etc.: University of Chicago Press.
  • TAK (2005). Rechtliche Integrationshemmnisse. Auslegeordnung und Lösungsansätze. Bern: Tripartite Agglomerationskonferenz.
  • TAK (2009). Weiterentwicklung der schweizerischen Integrationspolitik. Rohmaterial zu den einzelnen Themen zuhanden der Teilnehmenden der TAK-Hearings. Bern: Tripartite Agglomerationskonferenz.
  • Tilly, C. (1995). Citizenship, identity and social history. International Review of Social History 40(Supplement 3): 117.
  • Vatter, A. (2002). Kantonale Demokratien im Vergleich. Entstehungsgründe, Interaktionen und Wirkungen politischer Institutionen in den Schweizer Kantonen. Opladen: Leske und Budrich.
  • Waldrauch, H. and C. Hofinger (2010). An index to mesaure the legal obstacles to the integration of migrants. Journal of Ethnic and Migration Studies 23(2): 271285.


  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Two-dimensional typology of integration regimes
  5. 3. Transfer of concept and measurement to the cantonal level
  6. 4. Concept validation: four categories instead of two dimensions
  7. 5. Empirical results: testing construct validity
  8. 6. Discussion
  9. Literature
  10. Appendices



Federal Office for Migration


Federal Commission for Migration


International Migration, Integration and Social Cohesion in Europe


Tripartite Agglomeration Conference

Swiss cantons




Appenzell Inner Rhodes


Appenzell Outer Rhodes


























St. Gall





















Table 4: Operationalization and sources of cantonal integration policy

Cantonal integration policyAdditive index based on the components listed below 
Labour market accessAdditive index based on migrants’ access to cantonal employment in administration, teaching position, police service and judiciary. Coding for each indicator: 0 = no access, 0.5 = very restricted access, 0.75 = restricted access, 1 = unrestricted accessFederal Commission for Migration (EKM, 2007b), own survey based on expert information (cantonal migration offices)
Access to nationalityAdditive index based on26 cantonal citizenship laws, own survey (official documents)
(1) period of residence: 0 = > 12 years, 0.2 = 10 years, 0.4 = 8 years, 0.6 = 6 years, 0.8 = 8 years, 1 = 2 years
(2) fee: 0 = > 2000 CHF, 0.25 = < 2000 CHF, 0.5 = < 1000 CHF, 0.75 = 500 CHF, 1 = < 500 CHF
(3) facilitated procedure (dummy): 0 = no, 1 = yes
(4) right of appeal (dummy): 0 = no, 1 = yes
Political ParticipationAdditive index based on non-nationals right to voteFederal Commission for Migration (EKM, 2007a, EKM, 2007b), own survey (official documents)
(1) right to vote in cantonal votes or elections: 0 = no, 1 = yes
(2) right to run for cantonal office: 0 = no, 1 = yes
(3) compulsory vs. optional adoption of cantonal regulation by municipalities: 0 = optional, 1 = compulsory
(4) right to vote in local votes or elections: 0 = no, 1 = yes
(5) right to run for local office: 0 = no, 1 = yes
(6) required period of residence: 0 = > 10 years, 0.5 = < 10 years, 1 = no requirement
(7) required residence permit status: 0 = permit C, 0.5 = < permit C, 1 = no residence permit required
cantonal provision of immigrants’ commission. Coding: 0 = no commission, 0.5 = ad hoc commission, 0.75 = traditional permanent commission founded after 2002, 1 = permanent commission founded before 2002
Anti-discriminationAdditive index based on cantonal anti-discrimination regulation in constitution and/or laws. Coding: 0 = none, 0.75 = constitution or law, 1 = constitution and lawTripartite Agglomeration Conference (TAK, 2009), own survey of cantonal constitutions and laws
Family reunionAdditive index based onAchermann (2004)
(1) cantonal requirements regarding housing situation. Coding: cantonal definition of the criteria “appropriate living place” for family reunion, 0.25 = restrictive, 0.5 = moderate, 0.75 = permissive criteria, 1 = no criteria
(2) differing criteria for EU- and third country nationals. 0 = indifferent, 0.5 = differing procedure, 1 = more permissive towards EU-nationals
Cultural requirements for naturalizationAdditive index based on26 cantonal citizenship laws, Federal Office for Migration (BFM, 2008)
(1) cultural integration required for naturalization. Coding: 0 = more restrictive, 0.5 = equal, 1 = less restrictive than national citizenship law
(2) cantonal implementation of integration agreements. Coding: 0 = systematic, 0.5 = partial, 1 = no application of integration agreement
Religious rights ITendency for legal recognition of minorities’ Religions. Continuous variables from 0 to 1, adapted from ChristmannChristmann (2010)
Religious rights IICantonal disposition towards Islamic burials. Additive index based onOwn survey based on expert information and official documents
(1) number of Islamic cemeteries in cantons (0–2)
(2) time of existence of Islamic cemeteries. 0 = none, 0.5 = built between 2005–2008, 0.75 = built between 2000–2005, 1 = built before 2000
Additional components of cantonal integration policy (not included in measurement)
Cantonal integration budgetsAnnual integration budget of a canton in CHF.Federal Office for Migration (BFM, 2006)
Mean: 203’154
SD: 304’194
Min: 0
Max: 1’300’000
Additional prescriptions on integrationadditive index based onFederal Commission for Migration (EKM, 2007b), Tripartite Agglomeration Conference (TAK, 2009), own survey of cantonal legislation (via LexFind)
1) Integration provision in constitution. Coding: 0 = no, 1 = yes
2) Specific legislation on integration. Coding: 0 = no, 0.5 = integration article within law, 1 = specific integration law
3) Integration guiding principles. Coding: 0 = no guiding principle, 0.5 = local guiding principle(s) (city level), 1 = cantonal guiding principle
  • image(4)

[  Scatter of Swiss cantons along socio-structural and religious rights categories. Note: Entries are z-transformed values. ]

Anita Manatschal is a PhD candidate and research fellow. She works and teaches as a doctoral student for Prof. Dr. Adrian Vatter, first at the chair for Swiss Politics at the University of Zürich and since August 2009 at the Institute for Political Science at the University of Bern. Her research interests include integration policy, citizenship, radical right parties, subnational comparison, Swiss politics and multimethod research. Address for correspondence: Institut für Politikwissenschaft, Universität Bern, Lerchenweg 36, 3000 Bern 9. Phone: +41 (0)31 631 83 33. Email: