Introduction

Does alien enfranchisement, i.e., policies extending voting rights to non-citizens, shape the perceptions of non-citizens regarding their ability to influence politics in host countries? Does such an enfranchisement entail (unintended) spill-over effects on citizens regarding their political attitudes? While often characterized as a tool to strengthen democratic principles and the political incorporation of immigrants (Blatter et al., 2017), we still know little about the political consequences of alien enfranchisement on the attitudes of the individuals directly targeted by this policy, i.e., non-citizens,Footnote 1 and those who are indirectly affected, i.e. citizens. In this paper, we are interested in studying the link between alien enfranchisement and external efficacy perceptions among non-citizens and citizens. Political efficacy is the self-evaluated attitudes of individuals commonly understood in two dimensions (Anderson, 2010; Craig, 1979; Scotto et al., 2021). In one part, external efficacy refers to an individual’s self-perceived ability to influence political processes based on government responsiveness. In another part, internal efficacy addresses how people perceive themselves as having the ability to make sense of politics through the self-evaluation of capabilities (Acock et al., 1985; Clarke & Acock, 1989). Here, we focus on investigating whether specific policy interventions play a role in improving the external efficacy perceptions, i.e., attitudes of system responsiveness rather than self-assessments of political competence. While it is no doubt fruitful to advance the study of both dimensions of efficacy, these two political attitudes have clearly been shown to address distinct latent concepts which deserve a devoted study of their antecedents and outcomes separately (Geurking et al., 2020; Scotto et al., 2021; Sulitzeanu-Kenan & Halperin, 2012).

Our study seeks to make three main contributions. First, we add to the field of political behavior by studying the consequences of alien enfranchisement policies on political attitudes, extending the literature on representation and migration and citizenship studies, which have, thus far, primarily focused on policy preferences (such as on welfare state and immigration policy) and turnout (Bergh & Bjorklund, 2011; Ferwerda et al., 2020). Higher perceived subjective responsiveness of the government and democratic institutions among residents begets stronger legitimacy of the democratic system and decreases the likelihood of violent demonstrations of resident grievances (Finifter, 1970; Geurking et al., 2020). As noted in earlier works, this has crucial social integration and political consequences for societies, particularly for citizen and non-citizen residents with immigration backgrounds (Dancygier, 2010). Yet, there has been no study on how public policies concerning electoral rules may influence this key political aspect in diverse societies. Thus, the question of what these policies are substantially doing to integrate non-citizens remains disputed (Engdahl et al., 2020; Ferwerda et al., 2020; Ruedin, 2018). In this paper, we shift the attention from behavioral outcomes to attitudes about perceived political representation by testing whether non-citizen enfranchisement policies alter external efficacy attitudes, a demonstrably important predictor of political integration and mobilization, across the broader electorate.

Second, we contribute to the broader public opinion and efficacy literature by studying how perceptions of government responsiveness and political influence are shaped by contextual factors, i.e., local electoral institutions enfranchising non-citizens. Higher external efficacy is often characterized as a function of higher levels of education, age, better socio-economic status, being a member of a dominant group in the society (i.e., white, male, non-immigrant), and personality traits (Campbell et al., 1954). However, policy factors explaining such differences across local or national contexts have been widely debated without much consensus in public opinion and political behavior studies (Wolak, 2018; Chamberlain, 2012).

Third, we examine how policy interventions targeting immigrants play a role in perceptions of political representation among citizens. In that sense, we are interested in the possible spill-over effects or externalities of expanding political rights to non-citizens on the native population, i.e., individuals who are not directly addressed by these policies. To our knowledge, this study is the first to address the implications of non-citizen voting rights on the broader population. Even though alien enfranchisement seemingly targets and should have a positive external efficacy boost on non-citizens, to fully comprehend the implications of such policies, it is essential to examine both the intended and the unintended externalities of their implementation from a public policy evaluation perspective (Oliver et al., 2019).

Our main argument is that non-citizens living in contexts that extend voting rights to these groups will have enhanced material and symbolic resources, making them perceive that their voice matters in politics. While there seems to be a limited effect of alien enfranchisement on electoral participation among non-citizens (Engdahl et al., 2020; Ruedin, 2018), robust evidence demonstrates that more inclusive electoral institutions yield more active political environments (Hayduk, 2006). Moreover, enfranchisement boosts external efficacy even if individuals do not actively use such electoral rights and are not explicitly mobilized (Shineman, 2020). This means that alien enfranchisement as a marker of a more inclusive conception of the demos could increase efficacy among non-citizens (Bauböck, 2009; Blatter et al., 2017). Next, considering the potential reactions among citizens given such inclusion of non-citizens in the electoral supply (Stutzer & Slotwinski, 2021), we assess whether such an expansion is associated with a subsequent influence on efficacy among citizens.

Empirically, while most studies in the political efficacy literature have concentrated on the North American context (Anderson, 2010; Wolak, 2018), we turn our attention to Europe. Using a within-case comparison, we focus on the Swiss case and exploit the sub-national variation in foreign enfranchisement policies. Breaking with earlier studies, we observe the impact of alien enfranchisement on the entire resident population by combining non-citizen and citizen resident samples. We employ a longitudinal research design using data from the Swiss Household Panel (SHP) from 1999 to 2014,Footnote 2 investigating the external efficacy differences within and across respondents due to variations in voting rights regulations. We find robust evidence showing that non-citizens living in municipalities with alien enfranchisement display higher external efficacy levels than those in contexts without such inclusive electoral policies. Moreover, our analysis does not reveal any evidence that citizens (either native or naturalized) perceive a decline in their level of external efficacy when alien enfranchisement extends the limits of formal political membership.

State of the Literature on External Political Efficacy in Diverse Democracies

While investigations of external efficacy that specifically concentrated on non-citizen populations and citizens with an immigration background have been scarce, there is a wealth of research on individual and, to a lesser extent, on contextual antecedents of external efficacy among citizens (Abramson & Aldrich, 1982; Campbell et al., 1954; Clarke & Acock, 1989). Ample evidence supports the finding that perceived influence on politics tends to be lower for those with lower income, less education, women, ethnic minorities, immigrants, and the youth (Campbell et al., 1954; Cohen et al., 2001). Specific to individuals with an immigration background, the acquisition of citizenship and the country of origin significantly affect how they engage with the host country’s politics (Wass et al., 2015).

Regarding the influence of participation in religious, civic, and other associational activities and memberships, the findings are mixed on whether civic engagement boosts people’s perception of their political power (Whiteley, 2011). This evidence has also been corroborated by the immigrant incorporation literature, advancing the idea that associational engagement factors have less to do with reported evaluations of external efficacy (Anduiza & San Martin, 2011). Instead, previous studies suggest that political interest, political trust, and satisfaction with democracy correlate strongly with higher external efficacy—although the causal directions of these potentially endogenous factors are not well explored (Anderson, 2010; Caprara et al., 2009). Concerning non-citizens, to the best of our knowledge, there has only been one comparative large-N study of immigrants’ external efficacy attitudes (Anduiza & San Martin, 2011). The analysis of Moroccan and Ecuadorian populations in four European cities shows that neither socio-demographics, material resources, nor civic engagement factors predict efficacy differences. Instead, expressed interest in the host country’s politics and social and institutional trust explain efficacy differences. However, the study’s research design lacks sufficient variation in immigrant populations and does not allow contextual factors to be assessed systematically, employing temporal, cross-national, or within-country comparisons.

Next, we turn to the scholars highlighting the necessity of contextualizing external efficacy (Ainsworth, 2000; Finkel, 1985). Evidence shows that political diversity and representation of non-majority opinions are linked to better external efficacy (Bowler & Donovan, 2002; Dyck & Lascher, 2009). Likewise, there seem to be modest effects of using direct democracy tools such as referenda and ballot initiatives (Schlozman & Yohai, 2008). However, studies suggest that it is unclear how existing public policies and political dynamics influence external efficacy (Catterberg & Moreno, 2006; Chamberlain, 2012). Paradoxically, place considerably matters when it comes to efficacy. It is generally agreed that there are substantial differences in efficacy outcomes between sub-national and regional units even when individual-level characteristics are controlled for—such as in the case of the United States (US) (Wolak, 2018). If we understand the external efficacy, even to a small extent, to be systematically shaped by the policy conditions, such efficacy outcomes cannot be studied in isolation from the context—particularly at the local level (Ainsworth, 2000). Therefore, here, we concentrate on the contextual predictors of external efficacy at the municipal level.

Extant work on the efficacy of underrepresented groups contests a solely individual-based view of external efficacy (Wolak, 2018). For instance, more female legislators boost female voters’ external efficacy (Atkeson & Carrillo, 2007). Likewise, there is a positive link between descriptive representation and higher efficacy among ethnic minorities in the US (Jeong, 2013; Pantoja & Segura, 2003). Notably, local conditions are increasingly demonstrated to matter extensively regarding political participation and integration outcomes (Bratsberg et al., 2021; Gonzalez-Ferrer & Morales, 2013; Manatschal et al., 2020). Following the comparative integration context theory, immigrants’ participation and belonging in diverse democracies strongly depend on the integration context and institutional arrangements, especially on the local level (Crul & Schneider, 2010). Yet, the implications of such evidence from the broader field of political science and migration studies rarely speak to each other. Most studies concentrate on citizens or non-citizens separately, unduly drawing boundaries between how we can explain the way individuals construct their views about external efficacy.

Theoretical Framework: Alien Enfranchisement and External Political Efficacy of Residents

Our argument starts with the idea that external efficacy is not merely an individual trait with little connection to the political environment (Wolak, 2018). Therefore, we embed our understanding and study of external efficacy within political and electoral contexts. Considering the inclusive electoral nature of alien enfranchisement policies, we expect that non-citizens will feel more externally efficacious when they have voting rights in their host country, untested in earlier work. Next, we advance that these policies may imply externalities on citizens even though they are not the target audience of such policies through spill-over effects that predict their political efficacy (Filindra & Manatschal, 2020; Kayran & Nadler, 2022). Therefore, we evaluate the external efficacy responses among citizens and non-citizens facing alien enfranchisement policies.

Political Efficacy of Non-citizens in Alien Enfranchised Contexts

There are potential objective and subjective channels through which non-citizen voting rights may arguably relate to higher external efficacy among non-citizens. First, non-citizens’ political interest and knowledge are improved if they are given the right to vote (Hainmueller et al., 2015). Yet, while political knowledge and interest strongly correlate with internal efficacy perceptions (Moeller et al., 2013; Reichert, 2016), the subjective competence and interest dimensions are separate from the perceived responsiveness of the government and institutions. This means that, while they are two related areas of inquiry, the external and internal dimensions of political influence should be understood distinctly.Footnote 3

Studies have shown that the act of voting itself can cyclically lead to higher efficacy (Clarke & Acock, 1989; Finkel, 1985). This is not to suggest that enfranchisement will automatically boost individuals’ perceived external efficacy. In fact, despite being enfranchised as a birthright, citizens widely vary in their perceived efficacy, highlighting that it should not be a straightforward assumption that the public policy of enfranchisement of non-citizens will produce their desired outcome. Importantly, non-naturalized immigrants often do not take advantage of their local voting rights (Ferris et al., 2020; Seidle, 2015). Yet, even without a behavioral change, such a shift in formal membership and acquisition of voting rights can potentially enhance an individual’s (perceived) role as a political agent (Shineman, 2020). We argue that individuals who receive democratic rights see themselves as more politically relevant not just because of the knowledge that they are included in the electoral processes but also because of increasing targeting by political actors, improving their external efficacy (Mettler, 2002). Previous empirical work has demonstrated that introducing alien enfranchisement is often a widely politicized and contentious process with significant media attention and party communications (Nadler, 2022; Piccoli, 2022). Being increasingly made targets and mobilized by political actors, such politicized discourse can improve non-citizens’ view of their role in politics.

Second, policy feedback effects of alien enfranchisement can reciprocally improve how non-citizens perceive the political institutions of the host democracy and their place in it (Gundelach & Manatschal, 2017; Manatschal et al., 2020). If non-citizens feel addressed by the policy and politics in the host country, they are more likely to engage in political affairs, improving their perceptions as political actors (Bevelander & Pendakur, 2009; Bloemraad, 2006; Cinalli & Giugni, 2011). Indeed, political inclusion is most likely to be achieved when immigrants are both attached to and perceive themselves as accepted by the host community, displaying higher levels of sense of community-predictive of higher external efficacy (Bilodeau et al., 2020; Just & Anderson, 2014; Simonsen, 2020). This also relates to signalling to immigrants a sense of symbolic “right to politics” that can enhance their confidence as legitimate political actors, boosting efficacy attitudes (Bloemraad & Schönwälder, 2013). Overall, we contend that there are both objective and subjective channels through which alien enfranchisement should be expected to be associated with non-citizens’ perceptions of better responsiveness:

Hypothesis 1

In municipalities with non-citizen enfranchisement, the external efficacy of non-citizens will be higher.

Political Efficacy of Citizens in Alien Enfranchised Contexts

Thus far, our theoretical framework has characterized the introduction of alien enfranchisement as a way of enhancing democratic principles and representation without many political risks (Ferris et al., 2020). However, one could also think these policies may affect citizens through spill-over effects even though they are not the target audience. Recent literature shows that immigration and integration policy regimes, which are not targeted at citizens, impact the political participation of citizens with an immigration background (Filindra & Manatschal, 2020; Kayran & Nadler, 2022). Likewise, integration policies for foreigners mitigate the negative impact of increasing diversity on social trust among citizens (Gundelach & Manatschal, 2017).

Regarding the extension of political rights, there is evidence suggesting that women, younger and dual citizens and, those with an immigration background, those identifying with the political left and displaying weaker national identity are more likely to support the enfranchisement of non-citizens (Michel & Blatter, 2021). Naturalized citizens support extending political rights to non-citizens more than native citizens (Kayran & Nadler, 2022; Michel & Blatter, 2021). Previous work on attitudes towards immigrants’ access to socio-economic rights has suggested that citizens who acquired citizenship through naturalization were more open to sharing rights with newcomers than native citizens (Just & Anderson, 2015). Identity dynamics and shared experiences of immigration are, among others, at play to explain these differences. Thus, introducing non-citizen voting rights could potentially imply some positive externalities on the external efficacy perceptions of these citizens where the political system is responsive to their preferences on alien enfranchisement.

Yet, such a positive spill-over may be unlikely if we consider the extensive literature demonstrating that some citizens are particularly skeptical towards the extension of socio-economic and civic rights to foreigners, previously exclusive to citizens (Gorodzeisky, 2013; van der Waal et al., 2013). Specific to the case of political rights, it is well known that there is difficulty in passing alien enfranchisement laws when the number of immigrants is higher, making them more consequential as political actors (Stutzer & Slotwinski, 2021). Even in countries where non-citizens are enfranchised, these rights are gained through lengthy and politicized processes of often unsuccessful attempts that eventually lead to—often quite narrowly—successful adoption of such policies (Koukal et al., 2021; Piccoli, 2022). Thus, there may indeed be a negative externality to enfranchising non-citizens (necessary for their integration), which can beget democratic grievances among citizens and fears over their representation (Klarenbeek & Weide, 2020; Stutzer & Slotwinski, 2021). In this respect, one considerable political risk of alien enfranchisement is the potential retention or decline in efficacy among citizens once non-citizens are accepted as political members. Therefore, considering such possible spill-over effects, this paper tests the following competing hypotheses regarding the influence of non-citizen voting rights on citizens’ external efficacy.

Hypothesis 2a

In municipalities with non-citizen enfranchisement, the external efficacy of citizens will be higher.

Hypothesis 2b

In municipalities with non-citizen enfranchisement, the external efficacy of citizens will be lower.

Method and Data

Data and the Case of Switzerland

We use the Swiss Household Panel (SHP) data from 1999 to 2014 to explore the relationship between external efficacy and alien enfranchisement (FORS, 2019). The SHP is a high-quality longitudinal panel study of households residing in Switzerland that surveys samples of native Swiss citizens, naturalized citizens, and non-citizen immigrants.Footnote 4 We concentrate on the case of Switzerland because it allows us the best empirical leverage in testing the political consequences of alien enfranchisement reforms while holding various country-level idiosyncrasies constant (Kayran & Nadler, 2022). Notably, the Swiss case is unique in allowing multiple over-time variations in passing these reforms and within sub-national unit (canton) variation across local residential areas.

In Switzerland, cantons and municipalities can enfranchise non-citizens since the Swiss federal constitution leaves legislation on electoral rules to the discretion of cantonal authorities. In most cases, if a canton passes laws making alien enfranchisement legal within the cantonal boundaries, this applies to all municipalities. In contrast, in a few others, municipalities within each canton may also be free to opt in or out of these rules. This means, within Switzerland, there is variation both between and within cantons in addition to the temporal variation of alien enfranchisement rules. Notably, except for Jura, all enfranchisement reforms have been enacted within the observation period of our study, see Table A1. Our research design allows for the precise assessment of the impact of alien enfranchisement while potentially confounding national and regional level differences are held constant between respondents.

At the individual level, we include adult citizens and non-citizens residing in Switzerland (18 + years old). Considering the potential systematic differences in efficacy based on civic and ethnic status among resident populations, as discussed in our theoretical framework in the previous section, we delineate the residential population in Switzerland into three categories. The first group refers to native citizens, i.e., Swiss citizens who acquired Swiss citizenship at birth through their parents. Next, we identify naturalized citizens, i.e., Swiss citizens who did not acquire Swiss citizenship by birth, meaning they went through the naturalization process and have an immigration background. Finally, our last group refers to immigrants who do not hold Swiss citizenship, i.e., non-citizens living in Switzerland.Footnote 5 We match our individual-level respondent data with information on households’ residential locations at the municipal level. We then create a new variable within the SHP by coding Swiss municipalities as positive or negative cases of alien enfranchisement.

Currently, out of the twenty-six Swiss cantons, eight have adopted some degree of local alien enfranchisement policies, see Appendix A. This means 599 municipalities in Switzerland out of 2205 have voting rights extended to non-citizen residents (see online appendix Fig. A1 for a visualization). Using Swiss federal statistics office data, cantonal constitutions, and secondary literature on the topic (Piccoli, 2022), we code municipalities with alien enfranchisement at the year of implementation as “1” and “0” if the municipality has no non-citizen voting rights. To relax the assumptions regarding the exact date on which the effect of the legislation should be observed, we estimate the effect using a lead (t + 1) impact of alien enfranchisement, revealing no change in our results (see online appendix Table G9). We use this binary measurement of enfranchisement rules to indicate the presence or absence of such voting rights rather than focus on incremental differences indicative of the rules’ conditions. While ongoing comparative data collection efforts exist, no measurement exists to capture such differences at the municipal level (Arrighi & Piccoli, 2018). Importantly, here, we are not interested in distinguishing between the degrees of alien enfranchisement rules, which fall beyond the scope of the present study.

Switzerland is a fitting case for our study because alien enfranchisement rules apply to all immigrants non-discriminately regardless of their countries of origin—given that the immigrants satisfy a specific set of residence duration conditions, see online appendix Table A1. Importantly, in the Swiss case, alien enfranchisement was introduced by widespread political campaigns, education efforts, and public information. Adopting such policies entails a constitutional change and a highly politicized popular referendum. Thus, Swiss residents are exposed to several instances of being made aware of these reforms (Piccoli, 2022; Stutzer & Slotwinski, 2021). While many of the enfranchisement legislation passed in the context of various constitutional changes, we hold that these other constitutional amendments were unlikely to be linked to increased external efficacy among residents—especially for non-citizen residents of each municipality, see Table A2 in the online appendix for details.

Next, since enfranchisement reforms are passed through popular vote, this raises a potential concern about whether the enfranchised municipalities constitute a set of distinct contexts regarding government responsiveness towards immigrants. However, the constitutional referendums required to pass these reform proposals were adopted by small margins, illustrating the contentious nature of these proposals even in areas where alien enfranchisement is present. The referendums that enfranchised non-citizens in Geneva, Vaud, and Fribourg passed with 52%, 56%, and 58% of “Yes” votes, respectively (Koukal et al., 2021). Thus, municipalities with alien enfranchisement are not unique in having an otherwise positively skewed environment of attitudes towards non-citizens. For instance, some relatively more immigrant-inclusive cities and cantons, such as Zürich and Basel-City, have not yet been able to enfranchise non-citizens. At the same time, some contexts tend to be more conservative and contentious on immigration, like the canton Appenzell-Ausserrhoden, which grants non-citizen voting rights.

Furthermore, canton-wide enfranchisement efforts occur in a top-to-bottom approach, expanding suffrage across all municipalities within canton boundaries. Hence, where cantonal referendums imposed all municipalities to enfranchise non-citizens regardless of the vote of the particular municipality, this constitutes a plausible exogenous event, allowing us to test our hypotheses as applied in previous research (Ferwerda, 2021; Ferwerda et al., 2020). In only two cantons, Appenzell-Ausserrhoden and Grisons, municipalities have the discretion to opt in for foreign enfranchisement, which can be argued to violate such an exogeneity assumption. In this respect, we also replicate our results by removing these two cases from our sample and reporting the same results, see online appendix Table G10.

Measurement of External Efficacy

We measure the external efficacy attitudes of respondents using the only suitable question item available in the SHP for our observation period formulated as follows: “How much influence do you think someone like you can have on government policy?” The answer scale goes from 0 to 10, where higher values indicate stronger perceptions of political influence. The question item is in every wave within our observation period except for 2010, 2012, and 2013, and our analysis covers all SHP waves from 1999 to 2014. While this formulation of measuring the latent concept of external efficacy is suitable in line with previous work (Craig, 1979; Geurking et al., 2020), we note that the lack of alternative or additional items that fit our purpose in the SHP is a limitation of our data. Regardless, this indicator provides us with a reliable question item that is shown to capture external efficacy, as illustrated in notable work on the topic (Scotto et al., 2021). In this way, we concede that an ideal measurement would be an indicator of external efficacy to include various dimensions of the latent concept. Yet, we hold that the SHP item we use is a valid measure of the perceived government responsiveness and excludes other closely related concepts, such as internal efficacy and political trust.

There is a significant gap in external efficacy in Switzerland if we compare non-citizens to citizens in Switzerland (see Figure D1 in the online appendix). In contrast, naturalized citizens do not exhibit any particularly striking feeling of being politically left behind compared with native citizens. Crucially, Switzerland seems to be a particularly prominent case where the gaps are among the highest when looking at other European democracies in the region, see online appendix Figure H1.Footnote 6 This makes the Swiss case a most likely case for a policy such as alien enfranchisement to impact such differences. If we do not find an external efficacy boost among non-citizens, considering the large gap and the low baseline, it may be even more unlikely to observe such an effect elsewhere where gaps may already be more modest.

Estimation Strategy

This paper is interested in the relationship between alien enfranchisement reforms and the external efficacy of non-citizen and citizen residents. Thus, we assess whether there is higher external efficacy reported among non-citizen groups living in municipalities with alien enfranchisement than those who do not and whether the efficacy boost in alien enfranchised municipalities is indeed substantively greater among the target group of the policy, i.e., non-citizens, to observe its intended impact more precisely (testing Hypothesis 1). We use an interaction term between alien enfranchisement (binary variable) and the residential group status of the respondents (distinguished into three categories) to investigate the relationships we are interested in. Following our theoretical discussion, we also investigate the spill-over effects among native and naturalized citizens (testing Hypothesis 2a and Hypothesis 2b). All our models are estimated with individual clustered standard errors to account for potential disturbances of heteroskedasticity and autocorrelation due to our repeated observation data structure.

Given the challenges in modelling the relationship between enfranchisement policies and efficacy in our research design, we use the following strategy to address the confounding factors while avoiding Type II errors in estimating our key coefficients of interest. First, presented in the findings section, we use a series of linear random-effects (RE) models with year and canton of residence fixed effects. This choice is based on several theoretical and empirical considerations. In this paper, we are primarily interested in the influence of alien enfranchisement on different resident groups in Switzerland. Therefore, using two-way (individual and year) fixed effects is empirically unhelpful, taking away all time-invariant “between variation” at the individual level. This is because our residential group categories are non-variant for most of our respondents (except for non-citizens who naturalize within our observation period), which will not be estimated if all between-individual variation is dropped. In fixed effects models, therefore, only time-variant variables could be estimated. This means that while RE models risk more bias in the coefficient estimates, here, it is preferred to lower variability (especially important where the sample includes small group sizes such as non-citizens) by partially pooling information across units (Clark & Linzer, 2014). Nevertheless, in all our models, we have time fixed effects to remove period effects and trends that may influence the efficacy attitudes that are systematically different from 1999 to 2014.

Furthermore, we control for age, sex, education level, household income, and employment status of the respondents as established material sources that seem to determine those who feel influential in politics systematically. Since the democratic institutions of the countries of origin and the extent to which respondents engage with the democratic processes in the sending country could matter for external efficacy, we control for the place of origin for those respondents in our sample with an immigration background, such as the non-citizens and the naturalized citizens.Footnote 7 Next, we add a second set of predictors of civic engagement, union membership, and religious attendance as two of the most evidenced factors of external efficacy. Third, we add political attitudes such as political interest, political trust, and partisanship demonstrated as potential correlates of external efficacy (see online Appendix C for details on the question items and online Appendix D for summary statistics) to isolate the relationship we are interested in.

We also specify our models with a series of alternative estimations that control for a set of individual-level factors that are relevant to external efficacy, such as the duration of stay in Switzerland for respondents with an immigration background, subjective reflections on one’s economic satisfaction, measures of civic engagement and associational activity (such as membership in a political party or a charitable organization), satisfaction with personal relationships in Switzerland, generalized social trust, and satisfaction with Swiss democratic institutions. We have not included these variables in our main models because they are not uniformly included in the SHP waves, and their inclusion reduces the sample size significantly. Our results are remarkably robust when including these covariates (see online appendix tables G1-G3).

Two other sources of confounding need to be considered: geographical location and individual-level characteristics. The first crucial confounding factor pertains to unobserved differences in the average external efficacy between cantons. To address this, we use canton fixed effects. By doing so, we isolate the impact of living in an alien-enfranchised municipality and the overtime change more narrowly (see online Appendix B). Overall, this means that the coefficients we present in our main findings reflect the effect of the policy reform for cantons that go from not having alien enfranchisement to having these rules (Vaud, Geneva, Fribourg, Graubünden) and from those cantons that have within canton variation between their municipalities (Appenzell-Ausserrhoden and Grisons). The disadvantage of canton fixed effects is that while it preserves the variation due to policy change within our sample, it removes much of the between-canton variation, increasing the risk of Type II error. Thus, we also report our coefficient of interest by removing the canton fixed effects (see Model 6 in Table 1).

As another way of dealing with the influence of residential location, we control for three theoretically important factors at the municipal level that could arguably be influential on both the introduction of alien enfranchisement and external efficacy simultaneously: the share of foreigners living in each municipality, the vote share of the vocally anti-immigration Swiss radical-right party, i.e. Schweizerische Volkspartei/ Union démocratique du centre (SVP/UDC), and the experience of failed referendums at the local level for alien enfranchisement. These three contextual factors do not change our findings (see online appendix Table G5 and G11). Thus, omitting these factors at the geographical location does not drive our results.

The second source of confounding is related to the time-invariant individual-level differences that remain, even when we control for some of the critical factors arguably related to external efficacy levels. This becomes a particularly crucial estimation challenge if individual characteristics correlate with the choice of the municipality of residence, introducing bias to our estimates. To resolve this, we estimate our models with two-way (individual and time) fixed effects with canton dummies and also in three subsamples of native citizens, naturalized citizens, and non-citizens that take away all differences between persons and isolate only the over-time variation in efficacy attitudes predicted by alien enfranchisement reforms in dynamic models. Given the similar levels of external efficacy differences between naturalized and native citizens, we inspect the external efficacy differences between non-citizen and citizen groups before and after the enfranchisement legislation in the short term (1-year pre and post the introduction of the voting rights), which we discuss in more detail below using a dynamic estimation (see Fig. 2). Overall, alternative estimation strategies and sensitivity checks corroborate our findings (see online Appendix F).

Finally, our modelling assumptions hold if individuals do not self-select into an alien-enfranchised municipality. In our SHP sample, when respondents’ status changes from living in a non-alien-enfranchisement to residing in an alien-enfranchisement municipality, it is in the overwhelming majority due to the policy changes, see online appendix Table G6. Only less than one per cent of respondents’ alien enfranchisement residential status change can be attributed to an actual move on the part of the respondents. Regardless, we re-estimate our models by removing respondents who have moved into a new municipality at t different from t–1, see online appendix Table G7, and report no substantive changes in our results.

Empirical Findings

We begin by presenting the results of our linear random-effects (RE) models. The first three models in Table 1 are focused on the overall correlation between alien enfranchisement and external efficacy levels, on average. Model 1 includes the first set of material resource controls, Model 2 adds civic engagement factors, and Model 3 is fully specified with political attitudes. We stepwise add our covariates of interest, as some control variables are missing in different waves.Footnote 8 Thus, we observe whether the reported effects are sensitive to such survey-year missingness. Next, to test our hypotheses on the relationship between alien enfranchisement and external efficacy, specifically for non-citizen and citizen respondents, we estimate an interaction term between alien enfranchisement and the three-category variable of the civic status of respondents. Model 4 includes this interaction term in a simple model specification (as in Model 1), and Model 5 presents the coefficient of this interaction effect on the fully specified model (as in Model 3). Model 6 uses the same specification as Model 5 but excludes canton fixed effects to estimate the results with the inclusion of between canton variation.

Alien Enfranchisement and External Efficacy

Does living in a municipality where all long-term non-citizen residents are enfranchised correlate with enhanced perceptions of political influence? Across the board, the results in Table 1 present that, on average, there are higher external efficacy perceptions among residents living in alien-enfranchised municipalities. Since our dependent variable is on a 0 to 10 scale, alien enfranchisement seems to be associated with at least 0.11 points of higher efficacy at p < 0.05 level.

Next, controlling for some of the most established alternative external efficacy explanations, we find that compared to a native citizen, being a non-citizen in Switzerland is correlated with lower efficacy by about 1.49 points (or 14.9% points) on our dependent variable scale. Likewise, compared to naturalized citizens, non-citizens in Switzerland are at least about 1.60 points (or 16% points) lower in external efficacy (not shown here). It is important to note that while the differences between citizens and non-citizens are statistically significant, the differences between native and naturalized immigrants are not.

Table 1 Alien enfranchisement and external efficacy attitudes, direct and interaction effects

Alien Enfranchisement and External Efficacy of Non-citizens and Citizens

Next, we test whether non-citizen voting rights, indeed, influence their target group of immigrants and examine whether an externality effect exists on citizens using an interaction term. If we are right in suggesting that alien enfranchisement is correlated with improving external efficacy among non-citizens, Model 4 and Model 5 should return significant interaction terms between being a non-naturalized immigrant and living in an alien-enfranchised municipality (Hypothesis 1). The results presented in Table 1 (Models 4–6) confirm the H1. While living in a municipality with or without alien enfranchisement does not systematically predict efficacy differences between citizens, alien enfranchisement is positively correlated with higher efficacy for non-citizens compared to the native citizen group. This means that there are efficacy gains in contexts with alien enfranchisement, specific to the non-citizen group. This result is robust to adding between-canton variation (by removing the canton fixed effects dummies) in Model 6. We note that the coefficient of our interaction term is statistically predictive only at p < 0.01 level in the simpler Model 4 with a larger number of observations up to 2014. Our interaction term is statistically significant at the p < 0.05 level in Model 5 and Model 6, with a relatively smaller sample of year-waves and observations (which are dropped due to missingness in some of the important control variables included in the models). To substantively interpret this interaction coefficient, Fig. 1 plots the average marginal effect of alien enfranchisement on external efficacy for the three groups using Model 4.

Figure 1 shows that, in line with our expectations, alien enfranchisement is correlated with higher external efficacy attitudes among non-citizens. The average marginal effect predicted is about 0.3 points on our efficacy scale (or 3% points), suggesting a small but non-negligible influence on efficacy for the non-citizen group. Figure 1 also reveals that alien enfranchisement does not correlate with negative external efficacy for naturalized or native citizens (at p < 0.05 level). This latter result concerning citizens is crucial because when investigating the efficacy impact of alien enfranchisement, it is key to test whether this immigrant-inclusive policy is negatively related to external efficacy perceptions among citizens. It is important to point out that Table 1 reports that the coefficients indicative of the link between alien enfranchisement and efficacy in these two citizen groups vary and are sensitive to model specifications. Regardless, Table 1; Fig. 1 show a positive relationship between immigrant-inclusive electoral rules and efficacy for non-citizens, finding evidence for Hypothesis 1. Regarding our competing hypotheses about citizens, at least from this analysis, there seems to be no evidence to suggest a decline in efficacy perceptions among citizens. Therefore, we conclude that our results support Hypothesis 2a, while there is less evidence to hold Hypothesis 2b.

Fig. 1
figure 1

Average marginal effects (AMEs) of alien enfranchisement on external efficacy. Predicted using Model 4 in Table 1. Estimates are presented with95% Confidence intervals

Changes in the Efficacy of Residential Groups Before and After Alien Enfranchisement

We use an alternative estimation strategy to look at changes in efficacy for citizens and non-citizens due to alien enfranchisement. Considering the lack of substantively meaningful efficacy differences between naturalized and native citizens, we group them here as “Swiss citizens” for brevity and an intuitive interpretation of our results. We interact this binary citizenship variable with alien enfranchisement and estimate two-way (individual and year) fixed effects models with year and canton dummies (see online appendix Table F2 in the appendix). In this way, we can both take out potential individual-level confounders in our model and estimate the changes in residential groups once their municipality goes from not having alien enfranchisement to giving non-citizens voting rights. Figure 1 illustrates the predicted efficacy levels for these two groups in municipalities before alien enfranchisement policies and after enfranchisement. In this way, it presents the change in efficacy across groups reflective of this policy change.

Looking at Fig. 2, we see a robust and positive effect of alien enfranchisement on non-citizens (going from about 2.31 to 2.55 points). Notably, the difference of such an increase for non-citizens is statistically significant from the difference that alien enfranchisement makes for Swiss citizens (going from about 3.83 to 4.02). Despite a change in efficaciousness in both groups, there is a statistically significant difference in such differences, about 0.05 points. This means that enacting alien enfranchisement policies leads to an increase in efficacy for non-citizens, which is statistically different and larger from the increase than that in efficacy for citizens. We conclude that the alien enfranchisement leads to small but robust efficacy gains on its intended residential group of non-citizens.Footnote 9

Fig. 2
figure 2

Linear predictions of the effect of external efficacy in Swiss and non-Swiss respondents. Predicted using Model 3 in online appendix Table F2 with individual, time, and canton fixed effects (full table of results is available in the online appendices). Estimates are presented with 95% Confidence intervals

Lastly, our main findings above showed no evidence of a negative backlash in the short term for citizens’ external efficacy in enfranchised municipalities. Yet, citizens are not equally willing to share rights with immigrants (Michel & Blatter, 2021). Citizens’ reactions to alien enfranchisement may, for instance, differ considering the size of the potential beneficiaries, i.e., the share of the foreign population in the municipality (Stutzer & Slotwinski, 2021). Likewise, we could expect that attitudes towards such alien enfranchisement may mitigate enfranchisement’s relationship to citizens’ efficacy after the passing of such policies. Given the lack of a relevant question item in the SHP, we cannot test this potential mitigating relationship directly at the individual level. Nevertheless, following previous research demonstrating a negative relationship between the share of foreigners in residential areas and the probability of enfranchisement (Kayran & Erdilmen, 2021; Stutzer & Slotwinski, 2021), we test the potential of this mitigating relationship (albeit indirectly). To do so, we estimate two-way and three-way interaction effects between alien enfranchisement, civic status group, and the share of foreigners in the residential area and find that, in our sample, such mitigating effects do not seem to be statistically significant, and inclusion of these interaction terms do not alter our findings presented here (see online appendix J).

Discussion and Conclusion

Our paper makes three main contributions to democratic representation, political integration of immigrants, and political behavior fields. Our study is the first to investigate external efficacy attitudes among non-citizen populations in democratic societies. We add to the literature by showing that the local context matters significantly in the political integration potential of non-citizens and that immigrant-inclusive policies at the sub-national level, such as alien enfranchisement, have a non-negligible positive influence on boosting non-citizens’ perceived importance as political actors (Gonzalez-Ferrer & Morales, 2013; Manatschal et al., 2020). Thus, our paper serves as an important evaluation of the intended consequences of the political impact of alien enfranchisement in diverse democracies. This is key because extant work has documented that non-citizen voting rights have not yielded the desired enhanced political participation outcomes among non-citizens, raising questions about its usefulness as a policy measure (Bevelander, 2015; Ruedin, 2018; Seidle, 2015). Our findings add further credence to the arguments pointing to the symbolic role of immigrant-inclusive policies in fostering democratic representation, integration, and a sense of belonging for immigrants in host countries, which is crucial in the long-term social cohesion outcomes in diverse societies (Bilodeau et al., 2020; Gundelach & Manatschal, 2017).

Second, previous work focused almost exclusively on the implications of voting rights extension on the targeted populations, i.e., non-citizens. However, the highly politicized context around extending political rights to non-citizens calls for a better understanding of citizens’ political attitudes and perceived representation in reaction to such policies regarding potential unintended consequences. Therefore, our study extended the research area to the policy externality effects of adopting alien enfranchisement among citizens. This is a critical aspect of evaluating the potential usefulness and applicability of alien enfranchisement policies, as a decline of external efficacy among citizens could have had serious knock-on effects on grievances and vote choices (Geurking et al., 2020; Sulitzeanu-Kenan & Halperin, 2012). Our analysis reveals that there seems to be no adverse effect of such inclusivity on citizens’ perceived political influence—despite their politically contentious nature.

Third, most research on political efficacy in political behavior examined the role of individual-level factors. Our study adds to these debates by suggesting that contextual factors can account for important cross-national and between-group variations of external efficacy. Notably, we have argued that the overall high gap in efficacy perceptions between citizens and non-citizens renders Switzerland a most likely case to study the potential of alien enfranchisement in reducing differences in efficaciousness between distinct groups of residents. Nevertheless, our proposed mechanisms linking alien enfranchisement to higher external efficacy perceptions also relate to other advanced democracies with large immigration populations and are independent of the Swiss case. For instance, there is reason to assume that despite the lower turnout among non-citizens even when they are enfranchised, we see that even in the case of Switzerland (with stark differences in efficacy between a foreigner and citizen residents), the ability to influence government decisions at the local level seems to beget increased perceptions of government responsiveness.

The mechanisms linking alien enfranchisement to better political participation among citizens and the spill-over effects of policies targeting non-citizens on citizens have been observed outside the Swiss context (Stadlmair, 2020; Manatschal et al., 2020). Yet, Switzerland has one of the largest number of foreign residents in advanced democracies. Thus, one could reasonably assume that in contexts with less heterogeneity, the likelihood of a negative backlash may even be lower, considering the lowered potential of electoral expansion. However, future research must investigate to what extent different compositions of non-citizen populations in terms of countries of origin and socio-economic characteristics across other host countries influence such results. Due to the limitations of the present data, the analysis here can disentangle neither the differential impact of heterogeneity of foreign populations on citizens nor explore more detailed country-specific aspects of the foreign-born population as a predictor of political attitudes. Overall, our findings in this paper advance existing evidence but also call for future research to investigate how alien enfranchisement relates to efficacy and other political attitudes in different European contexts from a comparative perspective.

Another limitation in the data is the inability to attribute whether non-citizens in our sample were, in fact, eligible for such voting rights immediately at the time of the reform or whether they had voted in the next local elections. Moreover, there is a lack of data for measuring citizens’ attitudes towards sharing political rights with non-citizens, which would be necessary for capturing citizens’ reactions pre- and post-enfranchisement. In addition, our paper focused on the short-term influence of alien enfranchisement on efficacy and provided the first insights into the absence of a negative externality of alien enfranchisement on citizens’ efficacy perceptions. However, evidence shows that alien enfranchisement boosts political parties’ efforts for descriptive representation of immigrant-origin populations (Nadler, 2022). Therefore, increasing immigrant descriptive representation may lead to a lowered feeling of political influence and representation among citizens in the longer term. Data limitation impedes us again from observing changes in efficacy in the long run. We believe these areas chart a fruitful further research agenda to investigate such links between alien enfranchisement, descriptive representation, and the political attitudes of citizens and non-citizen residents in diverse democracies.