Introduction
The issue of immigration regularly sparks contestation in receiving societies. The native population is often reluctant to embrace the social change of becoming a country of immigration. In this context, many liberal democracies face a tension between a continuous economic need for foreign workers and public opposition to immigration (Hampshire, Reference Hampshire2013; Hollifield, Reference Hollifield2004). An old and prominent idea to ease this tension are temporary migration schemes where immigrants are allowed to enter the country as a flexible workforce but not allowed to stay on a permanent basis or to receive equal social and political rights as native citizens (Martin, Reference Martin, Chiswick and Miller2015; Ruhs, Reference Ruhs2006). A case in point for this rationale are the guest-worker policies in the 20th century that allowed labour-starved economies to recruit large numbers of foreign workers that were needed to sustain economic growth (Castles, Reference Castles1986). More recently, a revival of temporary migration policies can be observed, highlighting the continuous popularity of such models of immigrant admission (Castles, Reference Castles2006; Wright and Clibborn, Reference Wright and Clibborn2020). Other scholars have argued that immigrant-receiving countries have converged towards a ‘market model’ that prioritises temporary migration in an attempt to escape the liberal paradox between economic needs and political opposition (Boucher and Gest, Reference Boucher and Gest2018). In a similar vein, various governments have sought to restrict welfare benefits for immigrants to address popular discontent with immigration and to send control signals to voters (Emmenegger and Careja, Reference Emmenegger, Careja, Emmenegger, Häusermann, Palier and Seeleib-Kaiser2012; Slaven et al., Reference Slaven2021). Indeed, governments have been found to be more likely to restrict immigrant rights than immigrant numbers when facing successful anti-immigration parties (Lutz, Reference Lutz2019). An implicit assumption of such policies is that public opposition to immigration can be mitigated if immigrants are entitled to fewer rights and, hence, some privileges of the native population are preserved. While restricting immigrant rights appears to be both popular and contentious, we know surprisingly little about the implied mechanism of appeasing domestic opposition. Are people indeed more willing to admit immigrants with an inferior legal status?
The burgeoning literature on immigration preferences focuses on explaining the general support of (or opposition to) immigration or seeks to identify which immigrants are preferred by the native population (see Hainmueller and Hopkins, Reference Hainmueller and Hopkins2014, for an overview). Thereby, the focus is on entry policies, i.e. who should be allowed to immigrate, while neglecting another core part of immigration policy, namely the post-entry rights that are part of every immigration decision. What residence permit and associated rights will immigrants receive upon arrival? These rights define the terms of residence and define the conditions for permanent settlement and access to social membership in the receiving country (cf. Bjerre et al., Reference Bjerre2015; Tichenor, Reference Tichenor2002: 36). Immigrant admission involves thus not only a decision about which and how many immigrants to allow entry to the country but also what rights should be associated with the residence status of admitted immigrants. A comprehensive understanding of immigration preferences should therefore take into account the external dimension of entry selection as well as the internal dimension of associated rights and the acknowledgement of immigration policy preferences as multidimensional (cf. Helbling et al., Reference Helbling2023; Reference Helbling2024). So far, only a few studies seek to explain citizens’ preferences on the provision of immigrant rights. Existing research in this area has studied preferences on immigrants’ access to welfare benefits (e.g. Czymara and Schmidt-Catran, Reference Czymara and Schmidt-Catran2016; Hjort, Reference Hjort2016) or their access to citizenship (e.g. Hainmueller and Hangartner, Reference Hainmueller and Hangartner2013; Turper et al., Reference Turper2015). However, these studies do not allow us to assess the interdependence of preferences on post-entry rights with preferences on admission.
This study seeks to shed light on immigration preferences by taking into account the multi-dimensionality of immigration policies and assessing the idea of public support for immigration being conditioned by immigrants’ post-entry rights. First, I synthesise the theoretical argument of such a rights-conditionality in immigration preferences by focusing on two important immigrant rights that are characteristic for contemporary immigration policies: the right to permanent residence and social rights of access to welfare benefits. I theorise how the restriction of these rights may attenuate natives’ opposition to immigration. Restricting permanent residence allows receiving countries to exclude immigrants from long-term settlement and restricting welfare rights allows for the exclusion of immigrants from social solidarity and the access to welfare benefits. The logic of such a rights-conditionality in public preferences suggests that if immigration is only temporary and bears limited fiscal risks, then there is less to fear from immigration and the public should, as a result, be more willing to accept immigrants. Second, I derive a series of empirical implications for the structure of immigration preferences and the circumstances under which immigrant rights should be the most likely to influence natives’ willingness to admit immigrants. Considerations of economic burden and symbolic threat, as well as historical policy legacies, should moderate the effect of immigrant rights on opposition to immigration. Third, I test these theoretical expectations with a survey experiment among citizens in the United States and Switzerland. Using a factorial design, survey respondents were asked to judge the admission of fictitious immigrants conditional on individual characteristics and prospective post-entry rights. The mixed results reveal that the restriction of welfare rights does significantly mitigate citizens’ opposition to immigrant admission, whereas this is not the case for residence rights. These findings provide novel insights into the nature and dynamic of immigration preferences in receiving-countries.
The rights-conditionality of immigration preferences
When and why citizens in receiving countries support or oppose immigration remains an intricate research question. The matter is complicated due to likely tensions and trade-offs resulting from the multi-dimensionality of immigration policy choices (Helbling et al., Reference Helbling2024). In the following, I systematise the argument that immigration preferences are not only conditional on who the prospective immigrants are but also on the rights that immigrants obtain as residents of the receiving country: Providing immigrants with an inferior legal status may attenuate perceived threat among citizens and increase their willingness to admit immigrants.
There are two strands of literature that can serve as inspiration for the argument of a rights-conditionality in immigration preferences: The literature on the political economy of migration policies and the literature on immigration attitudes. First, the reasoning of a rights-conditionality follows a common rationale of immigration politics and reflects the idea of a policy trade-off between the numbers and the rights of immigrants. Prominently theorised by Martin Ruhs, the idea suggests that ‘insisting on equality of rights for migrant workers can come at the price of more restrictive admission policies’ (Ruhs, Reference Ruhs2013: 9). The underlying argument is that immigrant rights are costly for receiving countries. The resulting expectation is an inverse relationship between the number of immigrants that countries admit and the extent of immigrant rights that they provide. While this original argument is primarily a political-economic theory about the rational behaviour of domestic economic agents, there is the assumption that the ‘price of rights’ comprises also of a stronger public opposition to immigration (see Ruhs, Reference Ruhs2013: 49). The implication of this idea is that support or opposition are not unconditional preferences but depend on the conditions of immigrant admission.
The literature on immigration attitudes provides us with some support for the idea that immigration preferences are multi-dimensional and that many citizens might not oppose immigration per se, but their attitudes are rather shaped by the perceived economic and social costs of immigration (e.g. Dolmas and Huffman, Reference Dolmas and Huffman2004; Helbling and Kriesi, Reference Helbling and Kriesi2014; Hanson et al., Reference Hanson2007; Ilias et al., Reference Ilias2008, Neureiter, Reference Neureiter2022; Turper et al., Reference Turper2015; Wright et al., Reference Wright2016). Most recently, a number of studies employ survey experiments to test the effect of different policy features including rights provisions on public support of immigration policies (Helbling et al., Reference Helbling2023; Reference Helbling2024; Jeannet et al., Reference Jeannet2021). They find mixed evidence with effects varying across different rights and obligations assigned to immigrants, providing a first empirical indication for the argument that I systematise in the following: The idea that the perceived costs of immigration in the public opinion results at least in part from immigrants’ post-entry rights, and that immigrants with fewer such rights should be conceived as a lesser burden to the receiving country. What is more, immigrants with an inferior legal status should also appear less threatening as they are less likely to evoke the perception of a zero-sum group competition between natives and immigrants (Esses et al., Reference Esses2001) as well as attenuate social status fears of natives whose citizenship privileges are preserved. The unequal status of immigrants and natives signals that social membership is reserved for citizens and that immigrants will not impinge on the dominant position of natives. As immigrant rights shape the effect of immigration on the receiving country, they could potentially be employed to address the material and symbolic concerns of citizens. Restricting post-entry rights could allow governments to set limits to what extent immigration impinges on the receiving society and may thereby attenuate natives’ opposition to immigration. Such a rights-conditionality in the public’s immigration preferences is a common implicit assumption of scholars and policy-makers.
To further develop the idea of a rights-conditionality in citizens’ immigration preferences, I discuss two main dimensions of immigrant rights: access to permanent settlement and access to welfare benefits. These rights provide access to resources and social membership and are, therefore, of crucial importance for the lives of immigrants.Footnote 1 Moreover, they are regularly the object of public immigration debates in terms of the conditions under which immigrants should be allowed to settle on a permanent basis and in terms of what welfare solidarity immigrants should be entitled to.
First, whether immigrants are granted access to permanent settlement determines the security and duration of their stay. Restricting residence rights has been a common policy in immigrant-receiving countries, the classic example being ‘temporary migration’ schemes that allow the entry of immigrant workers but seek to prevent their long-term settlement. The underlying rationale is that the right to permanent settlement is likely to heighten material and symbolic concerns of natives. In the material dimension, restricting permanent settlement may reduce the perceived economic costs of immigration. The longer the presence of immigrants the higher the pressure to grant them membership and the associated bundle of rights. The right to permanent settlement typically provides immigrants with a series of social and economic rights that may place costs on the society through schooling, housing and health care provision for permanent immigrants and their families (Green, Reference Green2004: 33; Winters et al., Reference Winters2003). Accordingly, restricting immigrants’ right to permanent residence minimises these fiscal risks and allows countries to export unemployment and social problems in bad economic times by sending immigrants ‘back home’ (Afonso, Reference Afonso2005). In the symbolic dimension, restricting permanent settlement may reduce the perceived cultural threat of immigration. Many citizens hold an exclusive national identity and have a preference for preserving a homogenous society and social cohesion. If the stay of immigrants is only temporary, then they are less likely to integrate and establish bonds with the society, and ultimately do not change the social fabric of the country. It is the permanent settlement with immigrants becoming members of the receiving society that demands openness from the native population and that is thus more likely to create social conflicts (El-Mafaalani, Reference El-Mafaalani2018). Therefore, treating immigrants as guest-workers who will return home after a while allows natives to preserve the idea that the society remains unaffected in the long-term by immigrants and avoids the adaptation to multicultural realities.
In sum, for both economic and cultural considerations, one might expect that permanent settlement increases the perceived threat by immigration compared to short-term residence of immigrants. Accordingly, citizens should be more supportive of immigrant admission if the latter’s residence is only temporary compared to immigrants who are settling in the country and becoming new members of society.
The existing research provides mixed indications on whether temporary migration is preferred over permanent immigration. Blinder (Reference Blinder2015) has shown that perceiving immigration as permanent is associated with a preference for more restrictive immigration policies. The study of immigration policies in OECD countries has further revealed that states have a systematic preference for temporary mobility to permanent migration (Lutz, Reference Lutz2024). In contrast, Aalberg et al. (Reference Aalberg2012) found that Norwegian citizens, on average, are more willing to grant a longer stay than a shorter stay. Diehl et al. (Reference Diehl2018) show that the intention of the stay duration does not significantly affect the support of Swiss citizens for immigration restrictions. Similarly, the prospect of permanent settlement did not reduce popular support for refugee admission in the United States (Stanley et al., Reference Stanley2022). These mixed results suggest that the existence of a rights-conditionality in immigration preferences based on the duration of stay remains an open question. Following the outlined theoretical considerations, citizens should be more supportive of immigration if the right to permanent settlement is restricted.
Hypothesis H1: Citizens are more willing to admit immigrants if the latter’s access to permanent residence is restricted.
A second dimension of immigrants’ post-entry rights is access to welfare benefits. Welfare rights determine whether and to what extent immigrants are included into the institutions of the welfare state and the national community of mutual solidarity. As with residence rights, immigrants’ access to welfare is likely to evoke material and symbolic concerns among natives. First, immigration entails fiscal risks for receiving countries, which evokes fears of higher welfare costs and drained public services (Sainsbury, Reference Sainsbury2006). Generous welfare rights create fiscal exposure to immigration and hence incentives to restrict the admission of foreign labour. Various studies have shown that many natives indeed perceive immigrants to be a financial burden for the welfare state and fear that immigration could erode their own social benefits as well as lead to a higher tax burden (Alesina and Glaeser, Reference Alesina and Glaeser2004; Hainmueller and Hiscox, Reference Hainmueller and Hiscox2010). Studies have identified such fiscal concerns as powerful predictors of anti-immigration attitudes (Fietkau and Hansen, Reference Fietkau and Hansen.2018; Gerber et al., Reference Gerber2017). Second, natives often question the legitimacy of welfare benefits for immigrants as a result of the common perception that immigrants are less deserving of social solidarity than fellow citizens (van Oorschot, Reference Van Oorschot2006). Welfare rights imply social membership and welfare states are understood accordingly as a ‘system of social protection for those who belong to the ethnically defined community’ (Kitschelt, Reference Kitschelt1997: 22). In this perspective, restricting welfare access for immigrants helps preserve the boundaries of the national community based on mutual solidarity.
The desire to exclude immigrants from access to welfare benefits could therefore be motivated by material concerns about a fiscal burden and symbolic concerns about immigrants’ non-deservingness. Such ‘welfare-chauvinist’ attitudes are common among citizens of Western democracies, suggesting that many citizens prefer immigrants to be excluded from welfare benefits or support immigrants’ welfare access only under certain conditions (Van Oorschot and Uunk, Reference Van Oorschot and Uunk2007). These public preferences render welfare chauvinism an electorally attractive option for political parties, a development that led to welfare retrenchment targeting immigrants (Chueri, Reference Chueri2021; Schumacher and Van Kersbergen, Reference Schumacher and Van Kersbergen2014). Such a restriction of immigrants’ social rights is an attempt of governments to ease the tension between the economic need for foreign labour and the public opposition to immigration (Emmenegger and Careja, Reference Emmenegger, Careja, Emmenegger, Häusermann, Palier and Seeleib-Kaiser2012; Slaven et al., Reference Slaven2021). The underlying micro-level mechanism reflects the idea of a rights-conditionality: restricting immigrants’ access to welfare benefits lowers the perceived burden and threat, thereby attenuating citizens’ opposition to immigration. In a similar way, a recent study by Helbling et al. (Reference Helbling2024) on public support for immigration policies in Germany, argues that voters who are concerned about immigration are willing to trade-off higher immigrant intakes for restricted welfare rights. Thus, it seems plausible that restricting immigrants’ access to the welfare state should also decrease natives’ opposition to immigrant admission.
Hypothesis H2: Citizens are more willing to admit immigrants if the latter’s access to welfare benefits is restricted.
The outlined arguments for a rights-conditionality in public immigration preferences bear further implications for the immigrant type and country context where such a pattern of preferences is most likely to be observed. In the next step, I discuss these moderating factors on the individual and contextual levels. First, different characteristics of immigrants come with different costs (and benefits) for the receiving country and not all immigrants evoke the same threat sentiments. Attitudinal studies show the consistent pattern that low-skilled immigrants are conceived as an economic burden and that the public uniformly prefers high-skilled immigrants (Hainmueller and Hopkins, Reference Hainmueller and Hopkins2014; Helbling and Kriesi, Reference Helbling and Kriesi2014; Ruedin, Reference Ruedin2020). Ruhs and Martin (Reference Ruhs and Martin2008) formulated the ‘numbers vs. rights’ trade-off specifically in the context of low-skilled immigrants whose rights are perceived as most costly for the receiving country as these immigrants are more likely to rely on public services and to receive welfare benefits. Moreover, it has been argued that low-skilled immigrants are conceived as having a lower work ethic and therefore as less deserving (Helbling and Kriesi, Reference Helbling and Kriesi2014). For these reasons, the provision of immigrant rights should primarily reduce natives’ openness to admit low-skilled immigrants as the group that they are most concerned about. Hence, the rights-conditionality in citizens’ immigration preferences should be moderated by immigrants’ skill -levels.
Hypothesis H3: The rights-conditionality in immigration preferences is larger for low-skilled immigrants than for high-skilled immigrants.
In addition, concerns of native citizens about immigration are often of symbolic nature based on a perceived threat to the values, culture and identity of the national community. Given clear ethnic hierarchies in the minds of people as well as varying perceptions of cultural proximity, not all immigrants should evoke the same level of cultural threat (Ford, Reference Ford2011). Hence, a greater perceived cultural distance should make citizens more likely to perceive immigrants as threatening and increase their inclination to support immigrant exclusion (see Hainmueller und Hangartner, Reference Hainmueller and Hangartner2013; Ruedin, Reference Ruedin2020). Therefore, the rights-conditionality in citizens’ immigration preferences should be moderated by immigrants’ country of origin.
Hypothesis H4: The rights-conditionality in immigration preferences is larger for immigrants from countries that are perceived as culturally more distant.
In addition to the individual characteristics of immigrants, immigration preferences may also differ across national contexts. Different countries tend to have different national models in their approach to immigration depending on how and when they have first experienced large-scale immigration in their recent past (Freeman, Reference Freeman1995; Reference Freeman2006; Lutz, Reference Lutz2024). I focus on two models with an opposite approach on the integration of immigrants into society. The first model is the ‘traditional immigration country’ or ‘settler nation’ found in countries such as the United States or Australia. They are characterised by the experience of mass immigration during their nation-building process that shaped the national self-conception and their collective identity: These countries’ understanding of society and their immigration policies are based on the idea that immigration is a natural and permanent feature of societal development (Freeman, Reference Freeman2006: 231). This contrasts with the second immigration model of (former) guest-worker countries that are found primarily in Western and Northern Europe. These countries’ exposure to large-scale immigration occurred primarily when they were already developed nation states with collective identities based on ethnic and cultural homogeneity, which explains their reluctance to admit immigrants that would change the social fabric of their nations (Cornelius and Tsuda, Reference Cornelius, Tsuda, Cornelius, Tsuda, Martin and Hollifield2004: 24–39). Consequently, immigration is in this context understood as a temporary necessity rather than a permanent feature and defining characteristic of the nation.
These different historical legacies of nation-building of ‘traditional’ and ‘reluctant’ immigration countries are likely to shape citizens’ immigration preferences. While the immigration mood may become more positive or more negative over time, the way in which immigration is understood in the context of the nation state is more likely to follow an institutional path-dependence from countries’ immigration regimes. Despite these models being shaped long in the past, they continue to shape contemporary immigration policies (see Ellermann, Reference Ellermann2021). In line with this reasoning, Turper et al. (Reference Turper2015) found that citizens’ preferred duration of stay of immigrants is a short-term stay in the Netherlands (former guest-worker country) but not in the United States (former settler nation). It seems therefore plausible that in the latter context of a traditional immigration country the restriction of immigrant rights should thus be less likely to be treated as an admission condition by the public than in the former context of reluctant countries of immigration with a guest-worker tradition.
Hypothesis H5: The rights-conditionality in immigration preferences is larger in former guest-worker countries than in former settler nations.
Research design
To put the stated hypotheses to an empirical test, I conduct a survey experiment among citizens in the United States and Switzerland.Footnote 2 These two countries are both wealthy liberal democracies with high levels of ethnic and cultural diversity, and as countries of immigration with liberal labour markets they rely strongly on immigrant workers. At the same time, they also share a political context where immigration is a salient and polarising issue in domestic politics. They differ however in their immigration history and policy legacy: While the United States represent a classic settler nation with a self-identity of a country of immigration where permanent settlement is seen as the default mode of migration, Switzerland, on the other hand, is a paradigmatic case of a country with a guest-worker tradition that is reluctant to accept immigrants and where migration is conceived as a temporary necessity rather than an integral part of the nation (Afonso, Reference Afonso2004; Ellermann, Reference Ellermann2013; Reference Ellermann2021). The two countries are thus ideal representations of the two opposite types of immigration regimes that were discussed above. Estimating the rights-conditionality in public immigration preferences in both countries allow for a comparison across the context of two different immigration regimes.
Population-based survey experiments offer important advantages as they facilitate the identification of causal effects through random assignment (internal validity) and the generalisation to the population that is studied by using representative samples (Mutz, Reference Mutz2011). It is therefore not surprising that they have become increasingly the method of choice in the study of immigration preferences (e.g., Hainmueller and Hiscox, Reference Hainmueller and Hiscox2010; Hainmueller and Hopkins, Reference Hainmueller and Hopkins2015). In this study, I employ a factorial design that asks respondents in the two countries to evaluate different immigration cases and decide about their admission to the country. Thereby, I manipulate the post-entry rights of immigrants and their individual characteristics to identify the causal effect of these factors on citizens’ willingness to admit immigrants. This design is motivated by the theoretical expectation that multiple attributes jointly influence immigration preference formation. Thus, it allows for multi-dimensional representations of immigrant profiles and to embed them in a concrete and realistic context (Auspurg and Hinz, Reference Auspurg and Hinz2014; Wallander, Reference Wallander2009). The decision task resembles that of immigration caseworkers in the sense that they have to decide for individual immigrants whether their admission request for a particular permit is granted or not.Footnote 3 Moreover, the admission of immigrants with different characteristics is likely to be influenced by social norms. The experimental variation of different factors reduces potential social desirability bias compared with the alternative of asking people directly about their willingness to discriminate against certain immigrants.
Each respondent receives ten vignettes that consist of a text description of a fictitious admission request with details on the post-entry rights the immigrant would receive if the request were approved as well as a series of the applicant’s characteristics. The vignettes vary by whether the immigrant would receive access to permanent residence (yes/no) and access to welfare benefits (yes/no) upon admission to the country. As individual immigrant characteristics, I include gender (male/female), age (young/old), family status (single, married with two children), skill-level (low/high) and country of origin (close/distant in terms of cultural proximity, Canada/Mexico for the United States; Germany/Kosovo for Switzerland) as vignette factors. I select factors that have been commonly applied in factorial experiments on immigration preferences, see Ward (Reference Ward2019) for age and gender, Valentino et al. (Reference Valentino2019) for family status, and Hainmueller and Hopkins (Reference Hainmueller and Hopkins2015) for the country of origin.Footnote 4 The inclusion of broader socio-demographic factors offers more realistic and complete profiles that provide sufficient variation for respondents to evaluate multiple vignettes. As people have specific images in mind when they think of immigrants (Blinder, Reference Blinder2015), having more complete profiles should reduce the influence of unsolicited perceptions of immigrant profiles. Moreover, socio-demographic factors are used in most survey experiments of this kind and will allow comparing their effect size with the effect of rights-restrictions. The resulting vignette universe of N = 128 immigrant profiles in text form allows for a design where the complete vignette universe is used in the assignment to respondents. Based on a sample of N = 476 in the United States and N = 420 respondents in Switzerland, the effective sample size is N = 8’960 observations.Footnote 5 This means that in the overall sample, each immigrant profile is assigned to around seven different respondents following the recommendations in the literature (Auspurg and Hinz, Reference Auspurg and Hinz2014: 48–49).Footnote 6 The sample uses quota for age, gender and education to approximate socio-demographic representativeness (see the online Appendix for the details).
After the presentation of the vignettes, respondents are asked to decide about their willingness to approve a particular immigrant permit on a 1-to-10 scale of how likely someone is to support admitting a specific immigrant. Besides the experimental block, the survey includes a pre-treatment questionnaire that measures socio-demographic characteristics of respondents as well as their political views in terms of left-right self-placement and attitudes to immigration (economic and cultural concerns).Footnote 7 The socio-demographic measures serve as control variables to reduce the error variance at the respondent level and to adjust for possible confounding with respondents’ personal features. The two survey items on political predispositions serve as moderators of the treatment effect.Footnote 8
To analyse the result of the survey experiment, I estimate ordinary least squares (OLS) regressions with standard errors clustered on the respondent level. This design allows me to identify the causal effects of the different vignette factors and to account for potential clustering effects from repeated measurement (Auspurg and Hinz, Reference Auspurg and Hinz2014). The experimental evidence is based on two types of estimates, average treatment effects comparing the predicted values of different vignette factors conditional on the other factors, and interaction effects to assess the factors moderating rights-conditionality in public preferences. I estimate causal interaction effects between different treatment factors, immigrant rights and immigrant characteristics (skill-level/country of origin), and an effect modification in terms of the extent to which the rights-conditionality in immigration preferences varies by respondents’ political predispositions to capture effect heterogeneity (cf. Keele and Stevenson, Reference Keele and Stevenson2021). Estimating separate models for the United States and Switzerland allows for cross-country comparison to assess the role of the country-context on immigration preferences. Finally, the analysis is completed with a series of robustness checks that assess the stability of the estimated effects.
Results
The results from the vignette experiment in the United States and Switzerland are presented based on citizens’ willingness to approve immigrant applications and the estimates on what drives these preferences. Overall, respondents tend to support immigrant admission to a similar degree in both countries: The average support of immigrant admission is almost identical in Switzerland (6.47) and in the United States (6.44), but with a stronger polarisation in the US than in Switzerland (see Figure A1 in the Appendix for the detailed distribution). How the admission preferences depend on the vignette factors is shown in Figure 1 below, based on the average treatment effects (see Table A1 in the Appendix for the model output). The model of the US sample identifies the expected positive effects of rights restriction on respondents’ willingness to admit immigrants. The effect is large and statistically significant for restricting migrants’ access to welfare benefits, but small and not significant for restricting immigrants’ duration of stay. This result provides first tentative support for an appeasing effect of immigrants’ inferior legal status. In contrast, immigrants’ characteristics have a limited explanatory power: With the exception of a strong positive effect of a high skill-qualification, the socio-demographics of immigrants (nationality, family status, gender, age) do not affect respondents’ willingness to admit immigrants in a meaningful way.
Figure 1. Coefficient plot of vignette factors (willingness of immigrant admission).
Note: The coefficient plot is based on a linear regression model that includes all vignette factors and clustered standard errors (confidence intervals of 90% (inner) and 95% (outer) shown). Sample size is N = 4760 (United States) and N = 4200 (Switzerland). The complete model output can be found in Table A1 in the Appendix.
The same model is estimated for the Swiss sample yielding similar effects but also notable differences. The strong positive effect of restricting welfare rights is in line with the US results and of similar magnitude, whereas restricting permanent residence leads to a significantly lower admission willingness, a result that contrasts with the US model and that also runs against the theoretical expectations. Finally, the estimates for the effect of immigrants’ characteristics resemble those in the United States, but with a larger effect size. Apart from the positive skill-level effect, nationality also exerts a significant effect (Germans significantly more likely to be admitted than Kosovans) as well as family status (singles significantly less likely to be admitted than families). Younger immigrants and women have a slightly higher propensity of being admitted, but with a coefficient close to zero. These base models provide supportive evidence for H1 on the restriction of welfare rights with strong and consistent effects, but they reject H2 on residence rights as the coefficient is either a null-effect or pointing in the opposite direction than expected.
After having established evidence on the rights-conditionality in citizens’ preferences of immigrant admission, I assess, in the next step, how contextual and individual factors moderate that outcome. For the contextual moderation, the two models above allow comparing the effects between the two country samples, showing that the effect of welfare rights is about the same size in the United States and Switzerland, whereas the effect of residence rights points in the opposite directions and only in the US sample with the expected positive coefficient. Thus, the expectation that the rights-conditionality is higher in Switzerland than in the United States (H5) is not confirmed, but the results nevertheless point towards meaningful cross-country differences when it comes to the preference for a temporary or permanent residence permit.
The moderation of the rights-conditionality by immigrant characteristics is assessed in separate models using interaction terms of the rights-restriction with immigrants’ level of education and nationality (see Figure 2). The predicted probabilities in both countries are largely between the values of six and seven, meaning that immigrants are more likely to be admitted than not across all different characteristics presented in the experiment. For welfare rights, I find that restricting access to social benefits increases admission willingness regardless of immigrants’ nationality and skill-level. Welfare right restrictions appear to be a popular policy and to increase the admission willingness for all immigrant types in the experiment.
Figure 2. Predicted values of admission willingness (causal interactions).
Note: The coefficient plots are based on linear regression models that include all vignette factors, interaction terms between immigrant rights and immigrant characteristics (nationality, level of education), separate estimates for the two country samples. Predicted probabilities with 95% confidence intervals shown. For the display of the interaction terms see Figure A2 in the Appendix.
For residence rights, the finding in the US is that restricting residence rights increases admission willingness primarily for Mexicans (but less for Canadians) and for low-skilled immigrants (but not high-skilled immigrants), whereas in Switzerland we find the reverse pattern that restricting residence rights reduces the willingness of admission primarily for Kosovans. This means that applicants from Kosovo are preferred as permanent immigrants, whereas this preference is smaller for Germans. The pattern is consistent with the theory expecting rights-conditionality to be most likely for immigrant groups perceived more negatively. Overall, all moderation effects are small in substantive size and are not statistically significant (see Figure A2 in the Appendix). This is remarkable given the importance of education and nationality on immigration preferences shown above. In sum, the evidence does thus not provide robust support for the expectation that rights-conditionality is larger for low-skilled immigrants (H3) and immigrants from countries that are perceived to be culturally more distant (H4).
Finally, I conduct four types of robustness checks to assess the validity of the results. First, the base models are re-estimated including socio-demographic control variables of respondents’ gender, age and level of education. The main results remain unaltered in substance when accounting for potential confounding effects of these respondent characteristics (see Table A1 in the Appendix for the model output). Second, the base model is estimated for the full sample to maximise statistical power (see Table A2 in the Appendix). The results confirm the positive effect of welfare rights restrictions and the null effect of residence rights restrictions. Interacting the post-entry rights with the theorised moderators corroborates the finding that significant country differences exist for the effect of residence rights but not for welfare rights, and that neither immigrants’ education nor nationality moderates rights-conditionality to a significant degree. Third, robustness is assessed through resampling tests that include and exclude certain respondents: Including non-citizens and excluding respondents that provided the identical answer across all 10 immigrant profiles for potential inattentiveness, results in largely similar effects. Fourth, I assess the stability of rights-conditionality by looking at further effect heterogeneity. A model interacting welfare rights and residence rights with one another confirms their independent effect and reveals no significant interaction between them. The two types of post-entry rights are thus evaluated separately as we would expect from the distinct result patterns. As immigration is a polarising issue, we might expect the rights-conditionality be driven by respondents’ political views. This is tested based on respondents’ immigration attitudes and political orientation (see Figures A3 and A4 in the Appendix). There are no significant interaction effects with the exception of welfare benefits in the United States. The effect of welfare restrictions is much stronger among conservative citizens than liberal citizens. Similarly, anti-immigration respondents increase their admission support significantly when welfare rights are restricted, whereas this is not the case for pro-immigration respondents. This heterogeneity by political views does not exist in the case of Switzerland: The rights-conditionality for both welfare and residence rights appears for citizens across the political spectrum. In sum, the heterogeneity analysis confirms the existence of a rights-conditionality in immigration preferences across different segments of the population, while also corroborating the existence of cross-country variation in the nature of the rights-conditionality.
Conclusion
Governments in liberal democracies face difficult choices regarding the openness to immigration, given widespread concerns among their domestic constituents. A common rationale for reconciling economic needs and political opposition is the idea of admitting immigrants into the country but restricting their post-entry rights to appease domestic opposition against their admission. Thereby, an inferior legal status of immigrants is conceived as a response to the concerns of natives by minimising the impact of immigration on the receiving society. The assumption is therefore that the public is more welcoming to immigrants if the latter’s rights are restricted.
This article theorised and analysed such a rights-conditionality in immigration preferences by using a factorial survey experiment in the United States and Switzerland – with mixed results. The argument finds strong and consistent support for immigrants’ welfare rights: In both countries the opposition to immigrant admission is lower when immigrants’ welfare rights are restricted. In contrast, the restriction of residence rights has a null effect in the US and even a negative effect on support of immigrant admission in Switzerland. These effects are of substantive size and consistent across different methodological choices. The common idea of a rights-conditionality – the mitigation of opposition to immigration by providing immigrants with an inferior and more precarious legal status with fewer rights – finds, thus, only partial support and its empirical nature deviates from how it has been understood so far.
These findings bear important implications for our understanding of immigration preferences as well as immigration policies. First, they demonstrate that public opposition to immigration is not only driven by who the immigrants are but also significantly shaped by the rights and benefits which immigrants are expected to receive after their arrival. This is, in particular, the case for welfare rights that have an even stronger effect on the admission preferences of citizens than salient characteristics of immigrants such as their country of origin. This means that citizens take into account the conditions under which immigration shall be permitted into their countries. For the legitimacy of policy-making, this means that one should not only eye at the numbers of immigrant intakes but also on the conditions under which immigrants enter the country.
Second, while the idea of a rights-conditionality is a widespread assumption, this study shows that the effect of immigrant rights is more complex than previously thought. By testing separately the effects of welfare rights and residence rights, I show that different rights are likely to have different effects in different national contexts, rather than a uniform effect in the sense of a trade-off between numbers and rights. What is more, the rights-conditionality appears also for culturally close and highly-skilled immigrants that should be the least threatening. The results indicate that the preferences may result from the perceived deservingness of immigrants to receive rights rather than the extent of threat sentiments, a motivation that has also been identified as an explanation for welfare chauvinist attitudes (Reeskens and van der Meer, Reference Reeskens and van der Meer2019).
Third, the evidence shows that the restriction of residence rights may have the opposite effect than commonly thought. The case of Switzerland suggests that citizens are more reluctant to admit immigrants that appear as ‘guests’ rather than as newcomers who come with the perspective of integration into society. This country-specific effect might be the result of negative experience with guest-worker policies in the past. It could however also result from the Swiss immigration model that values assimilation and that people prefer immigrants who commit themselves to the receiving country in the long-term (cf. Neureiter, Reference Neureiter2022). If residence is only temporary, one can expect that immigrants are less likely to undertake costly integration efforts. Permanent residence, on the other hand, is a commitment to integrate and by the expectation of settlement immigrants may become less threatening. This interpretation resonates with the finding that Swiss citizens prefer resident immigrants over cross-border commuters despite the former having a larger impact on society than the latter (Schaffer and Spilker, Reference Schaffer and Spilker2023). The widespread policies of temporary migration with the objective of addressing domestic labour needs, while avoiding a political backlash, may therefore not be an effective model. Public support for immigrant admission seems highest when immigrants are perceived to contribute economically (instead of enjoying welfare access) as well as socially by becoming an integrated part of society (instead of temporary residence).
This study has sought to advance the theory of a rights-conditionality in public immigration preferences – how post-entry rights shape the public support of and opposition towards the admission of immigrants – and provides novel insights into the way in which immigrant rights shape citizens’ preferences. Two important limitations deserve to be mentioned though: the motivation and mechanisms behind the observed rights-conditionality in immigration preferences remains an informed speculation and the empirical focus on two countries with different immigration contexts limits the generalisation that we can draw from the findings. Future research is thus warranted to expand this line of research across a wider number of countries and to uncover the different perceptions and motivations behind the way in which immigrant rights shape citizens’ openness to accept new immigrants.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1755773924000183.
Acknowledgement
I’m grateful to Sven Hegewald, Inge Hendriks, Jascha Grübel and Anna-Lena Nadler for their feedback on previous versions of the manuscript. I also thank Eva van Belle for the fruitful collaboration on the survey implementation in the context of the ‘nccr on the move’.
Financial support
This research was supported by the nccr – on the move funded by the Swiss National Science Foundation grant 51NF40-182897.
Competing interests
The author declares that he has no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Open access
