The Role of Local Voting Rights for Non-Naturalized Immigrants: A Catalyst for Integration?

Abstract

Recent decades have seen a strong trend among democratic countries to extend voting rights at subnational levels to non-naturalized immigrants, creating substantial variation across countries in terms of voting eligibility rules for non-naturalized immigrants. Our knowledge of the consequences of these different systems for immigrant political integration is, however, limited. This article seeks to shed new light on this important issue by using Swedish data to study whether immigrants who face shorter residency requirements for voting eligibility in local elections are more likely to integrate politically. We find little compelling evidence that such is the case. The results suggest that immigrants who became eligible to vote after six to seven years were as likely to naturalize and vote in future elections in both the short and long run as those who received the right to vote after only three years of residency. Thus, although expanded franchise can be of symbolic, as well as practical, value, it is unlikely to be a panacea for immigrant political inclusion. The argument that early voting rights for non-naturalized immigrants is desirable since it helps speed up immigrant political integration should, therefore, be used with some care by those advocating for such reforms.

Keywords

local election voting rights non-naturalized citizens integration naturalization turnout

Introduction

Since the 1970s, there has been a strong trend among democratic countries to extend voting rights at subnational levels to non-naturalized immigrants (i.e., immigrants who lack citizenship in their country of residence) (Earnest 2003). Today, non-naturalized immigrants have some voting rights in at least 65 countries (Immigrant Voting Project Website 2017). As a consequence of this development, there is substantial variation across countries in terms of voting eligibility rules for non-naturalized immigrants. In some countries, non-naturalized immigrants are allowed to vote in local elections upon arrival or after a few years of residency, whereas in others they must wait much longer or are not allowed to vote at all (Earnest 2003). Our knowledge of the consequences of these different systems for immigrant political integration is limited. This knowledge gap is unfortunate, since one important argument in favor of introducing local voting rights for non-naturalized immigrants is that doing so facilitates immigrant political integration (e.g., Harper-Ho 2000; Munro 2008; Vernby 2012). This article seeks to shed new light on this important, but understudied, issue by investigating the importance of the length of the residency requirement for immigrant political integration in Sweden.

A first challenge in studying this issue is that political integration is a nebulous concept that lacks a universally accepted definition. Previous research is replete with discussions about “the complexities and current muddiness of our definitions, conceptions, and theorizing about immigrant political incorporation” (Morales 2011, 11, see also Wolbrecht and Hero 2005; Goodman 2015). Thus, in studying political integration, it is important to be clear on the term’s meaning. In this article, we use Jimènez’s (2011) definition of political integration as containing both formal and informal dimensions. Whereas formal political integration refers to “legal status and citizenship, as well as participation in the political process, which primarily includes voting and holding elected office” (Jimènez 2011, 11), informal political integration, for Jimènez, includes various forms of political activism and associational engagement. This article focuses on formal political integration measured in terms of naturalizations and voting. Informal political integration is, of course, also interesting but lies outside this article’s scope, mainly due to a lack of adequate data.

The focus on the length of the residency requirement is inspired by economic research showing that favorable labor-market conditions at entry (Åslund and Rooth 2007) and early labor-market access (Hainmueller, Hangartner, and Lawrence 2016) are important for speeding up newly arrived immigrants’ economic assimilation. There is also a widespread belief that early interventions (i.e., introduction programs for refugees) positively affect the labor-market trajectories of newly arrived immigrants (Konle-Seidl and Bolits 2016). Similarly, research on native-born residents’ political participation shows that individuals who have their first opportunity to vote at 18 rather than at 22 typically have higher future turnout (Meredith 2009; Dinas 2012; Coppock and Green 2016), giving further support to the notion that the timing of voting eligibility is important.

Reasoning by analogy, we may expect a similar logic to apply to the process of immigrant political integration: immigrants could become more likely to integrate politically if they receive important political rights relatively soon upon arrival. A number of theoretical reasons have been suggested for why early voting rights could affect integration (e.g., Munro 2008). First, voting is frequently argued to be an important means of becoming incorporated into society and engaged in politics (e.g., Vernby 2012): granting non-naturalized immigrants the right to vote can, therefore, be viewed as a “way of educating future citizens in civic responsibilities and preparing them for citizenship” (Harper-Ho 2000, 297). Short residency requirements for voting eligibility may, thus, help ease and speed up the naturalization process. These earlier opportunities to vote may also encourage the development of a voting habit among newly arrived immigrants, which may affect future turnout even after they become citizens in their new host country, as earlier empirical results from Finland and Norway suggest (Wass et al. 2015; Ferwerda, Finseraas, and Bergh 2020).

Second, voting and the right to run for office might stimulate (and offer opportunities for) non-naturalized immigrants to become involved in various associations and forums in civil society (Munro 2008). One example is the possibility that political leaders from immigrant communities will emerge; another is that the incentive to organize and mobilize immigrant groups will increase. This higher participation level is, in turn, likely to increase immigrants’ knowledge about the political system and other institutions (Munro 2008, 73).

A third argument for early voting rights revolves around the negative consequences of not permitting non-naturalized immigrants to vote. Proponents of this argument stress that if immigrants must wait until after they are naturalized to vote, they might internalize the idea that political decisions should be left to others, which could lead to low levels of political and/or civic engagement or beliefs that other means of raising concerns are more effective, perhaps even violent ones (Munro 2008). Longer residency requirements for voting could, therefore, have detrimental effects on immigrants’ subsequent political participation.

As Munro (2008) notes, these arguments for early voting rights are built on the idea of integration through participation (i.e., allowing non-naturalized immigrants to vote soon upon arrival will help speed up political integration because it provides them with additional incentives and opportunities to learn more about their new country). In the end, early voting rights may, therefore, make immigrants more likely to naturalize and become politically active community members.

A potential counterargument to this view, however, is that the extension of voting rights to non-naturalized immigrants can be detrimental to immigrant political integration by contributing to the devaluation of citizenship (Schuck 1989; Pickus 1998). The desire to partake in elections is known to be an important motive for naturalization (Kahanec and Tosun 2009; Street 2017); thus, an opportunity for immigrants to vote without being citizens in their country of residence may weaken their incentives to naturalize (Lenard 2015, 129). Consequently, earlier voting rights for immigrants could have a negative effect on their political integration by reducing citizenship’s benefits and, thus, the desire for it. As citizenship status is known to be highly correlated with many types of political participation (Hainmueller, Hangartner, and Pietrantuono 2015), a reduction in the naturalization rate could be expected to have negative repercussions on other aspects of immigrant political integration. Moreover, to the extent that voting is a habit, it could also be problematic if immigrants are given the opportunity to vote before they feel ready to exercise this right, since in the same way that voting can become habitual, so, too, can not voting.

As this short literature review has made clear, it is not theoretically obvious whether we should expect shorter residency requirements for voting eligibility to enable or hinder immigrant political integration. The only existing empirical study aimed at answering (parts of) this question is one by Ferwerda, Finseraas, and Bergh (2020), who use Norwegian data to investigate whether immigrants who, due to the timing of their arrival in relation to the election cycle, were allowed to vote after three (instead of seven) years were more likely to vote in the next election.¹ The results of this Norwegian study show that early voting eligibility is associated with higher subsequent turnout, at least for immigrants from non-democratic source countries. To the extent that these results are more generally valid, they have important implications for the ongoing debate on how immigrant political integration is affected by different voting eligibility rules for non-naturalized immigrants (e.g., Seidle 2015; Ferris et al. 2020). In particular, the results indicate that shorter residency requirements for voting eligibility in local elections may help increase voter turnout among (certain) immigrant groups, at least in the short run.

Despite its merits, Ferwerda, Finseraas, and Bergh’s study (2020) has some important limitations that warrant further investigation. First, due to its design, the study cannot say anything about the long-term effects of granting non-naturalized immigrants voting rights relatively soon upon arrival, which is somewhat unfortunate, since there is no guarantee that the effects will look the same in the long run. Second, the study has a narrow focus on one aspect of political integration — voting in local elections — whereas other equally important indicators of political integration, such as naturalization or voting in national elections, do not receive attention.² Third, as is always the case when studying data from a single country and one election, it remains an open question to what extent the findings generalize to other contexts and countries. To decide whether the findings from Norway are more generally applicable, it is necessary to study this issue in other institutional settings.

This article seeks to extend and complement the Norwegian study (Ferwerda, Finseraas, and Bergh 2020) in all three regards. Although our study employs a similar empirical strategy, we analyze additional outcomes, including naturalization, and follow immigrants over a considerably longer time period. Our design, therefore, enables us to investigate political integration more broadly and, unlike the previous study, to examine long-term and short-term effects of the policy in question.

Moreover, the research presented in this article is situated in Sweden, rather than in Norway. Although similar in many respects, there are also important differences between the two countries (Pettersen and Østby 2013). The share of immigrants in Sweden’s population is higher, and there are also notable differences in the composition of immigrants between the two countries (Karlsdottìr et al. 2018). Most importantly, refugees account for a much larger share of recent immigration to Sweden compared to Norway (Pettersen and Østby 2013), and there is a fair degree of variation in immigrants’ origin countries in the two countries (Karlsdottìr et al. 2018). If the integrative effects of early voting rights depend on immigrants’ previous experiences and reasons for migration, as Ferwerda, Finseraas, and Bergh (2018) seem to indicate, it is fully possible that the relationship between early voting rights and political integration could look different in Sweden and in Norway.

Another important difference between the countries is that in Sweden, in contrast to Norway, local and national elections are held on the same day, which is known to increase local level turnout (Garrman 2016). Consequently, turnout among both natives and immigrants in local Swedish elections are considerably higher than in Norway (Kleven 2015). Whereas the Norwegian study reports that only around 20–25 percent of immigrants who arrived in 2008 voted in the 2015 local election (Ferwerda, Finseraas, and Bergh 2020), the comparable Swedish figure is above 50 percent (see Figure 1).

Figure 1.

Naturalization rate and voter turnout in 2010 by time in country.

Apart from extending the work of Ferwerda, Finseraas, and Bergh (2018), this article also contributes to the growing literature on immigrant electoral participation (de Rooij 2012; Wass et al. 2015; Ruedin 2018). In particular, investigating whether early access to voting fosters immigrant political integration can help shed new light on how and why political integration varies with the amount of time spent in the host country (Adman and Strömblad 2000; de Rooij 2012; Duleep 2015; Dancygier et al. 2015).

To preview our findings, we did not find evidence that the timing of voting eligibility matters for immigrants’ naturalization process or future turnout. Our results, thus, suggest that the positive short-run effects on turnout of early access to voting rights found in previous research (Ferwerda, Finseraas, and Bergh 2020) are more context-dependent than the supporters of early voting rights for non-naturalized immigrants would have hoped. Similar to Seidle (2015), we find that although expanded franchise can be of symbolic, as well as practical, value, it is unlikely to be a panacea for immigrant political inclusion. To be clear, our findings do not invalidate the case for extending voting rights to non-naturalized immigrants, but they suggest that the main argument for doing so should not be that immigrants’ early access to voting rights will foster their subsequent political integration.

The Institutional Setting

In Sweden, non-naturalized immigrants first gained the right to vote and run for office in local elections in 1976, conditional on having resided in Sweden for about three years before the election (Hammar 1990). In many countries, local or regional elections are considered less important than national elections (e.g., Cancela and Geys 2016), and Sweden is no exception in this regard (Oscarsson 2001). Nonetheless, elections to municipal and county councils carry substantial weight, as Swedish municipal and county councils have independent taxation rights and play a crucial role in the provision of vital government goods and services such as health care, education, and social assistance (Bäck 2003). Consequently, the decision to extend local election rights to non-naturalized immigrants was of great importance and strengthened immigrants’ political clout in Sweden (Vernby 2012).

This article uses quasi-experimental variation in voting eligibility in local elections for all elections that took place between 1976 and 2010 (11 elections, listed in Table 1). Before 1998 Swedish general elections were held every third year; thereafter, they have been held every fourth year. The elections we study all took place on the third Sunday of September, and there are always three separate elections on the same day: municipal, county, and national. The registration process prior to Swedish elections is automatic, and all citizens aged 18 years or above on election day can vote in all three elections. Non-naturalized immigrants aged 18 years or above on the election day, who satisfy the residency requirement, can vote in municipal and county elections. About one month before the election, eligible individuals receive a letter from the Swedish Election Authority with information on their voting rights and a voting eligibility certificate that informs individuals of their right to vote (Bernitz 2013). In elections from 1976 to 1994, non-naturalized immigrants needed to have been registered as residents in Sweden by November 1 three years prior to the election year (Prop. 1975/76:23). For later elections, exactly three years of residency before election day was required, except for EU nationals because of Swedish EU membership in 1995 (Bevelander 2015).³

Table 1.

Elections and Immigration Cutoff Dates for Voting Eligibility.

(1) Election	(2) Latest immigration date for voting eligibilityin the next election(cutoff dates)	(3) Affected immigrants	(4) Wait time forimmigrants who arrive just before the cutoff	(5) Wait time for immigrants who arrive just after the cutoff
1976	November 1, 1973	All	2.9 years	5.9 years
1979	November 1, 1976	All	2.9 years	5.9 years
1982	November 1, 1979	All	2.9 years	5.9 years
1985	November 1, 1982	All	2.9 years	5.9 years
1988	November 1, 1985	All	2.9 years	5.9 years
1991	November 1, 1988	All	2.9 years	5.9 years
1994	November 1, 1991	All	2.9 years	5.9 years
1998	September 19, 1995	Non-EU	3 years	7 years
2002	September 14, 1999	Non-EU	3 years	7 years
2006	September 16, 2003	Non-EU	3 years	7 years
2010	September 18, 2007	Non-EU	3 years	7 years

Note: For data reasons we can only use variation in voting eligibility for the elections in 1976–2006 when we use turnout in the election in 2010 as the outcome.

In the main analysis, we focus on immigrants from non-EU countries that were subject to the “three-year rule” throughout the study period (the sample used in the analysis is discussed in the next section). Table 1 shows the immigration cutoff date for immigrant voting eligibility in each election (column 2) and the impact on wait time to the first opportunity to vote in local elections for immigrants who arrived just before versus just after the relevant dates (column 4 and 5). For example, an immigrant who arrived on November 1, 1973, could vote in the local elections that took place in September 1976. Thus, the wait time between immigration and the first opportunity to vote in local election was only around 2.9 years. Immigrants who instead arrived on November 2, 1973, got their first opportunity to vote in local elections in September 1979. Thus, their wait time was around 5.9 years. In this way, arriving just before or after a cutoff creates a significant difference in wait time before an immigrant can cast a vote in a local election for the first time.

Before we turn to a description of our data, it is worth noting that the turnout of eligible non-naturalized immigrants has decreased over time. About 60 percent of eligible non-naturalized immigrants voted in the 1976 municipal elections, while the corresponding figure in 2014 was only about 35 percent (SCB 2019). It is also interesting to visualize the general context in which we perform our analyses. Figure 1 displays how the naturalization rate (solid line) and voter turnout in municipal elections (dashed line) varied over the years since immigration among immigrants in our main sample (i.e., immigrants from non-EU countries). As is clear, both the naturalization rate and voter turnout increase with time in the country. Among immigrants from non-EU countries who had lived in Sweden for 30 years, about 90 percent were naturalized citizens, and 70 percent voted in the municipal election in 2010.

Data

To study the integrative effects of early, rather than late, voting eligibility for non-naturalized immigrants, we combined register data from Statistics Sweden with information on voting eligibility cutoff dates. The main data source is a population register (Historiska FBR) that, among other things, records all immigrations, emigrations, births, and deaths for the full Swedish population during the period 1969–2014, which means that it covers all migrations in the periods surrounding the voting eligibility cutoff dates analyzed. This register also records changes in citizenship status.⁴ The dataset can be linked to other registers at Statistics Sweden through the personal identifier. In particular, it can be linked to a register containing information on gender, birth year, and birth country. For confidentiality reasons, some birth countries are grouped so that it should not be possible to identify immigrants from less common source countries. In Section 1 in the Supplemental Appendix, we describe the classification scheme and the consequences for the construction of the main sample, to which we now turn.

We made five important restrictions to our sample. First, all immigrants born in EU or Nordic countries were excluded from the main analysis because of Sweden’s EU membership since 1995. As a consequence of Sweden’s EU membership, the waiting time for EU nationals to vote in local elections was drastically shortened (see the above section). Thus, we dropped this group because their eligibility rules for participating in local elections changed over time. By keeping the source countries constant over time, interpretation of our results becomes more straightforward.⁵ The non-EU immigrants are, arguably, also the most interesting group to study, since they (with some exceptions, such as immigrants from the United States and Canada) have the poorest integration outcomes. Almost 50 percent of immigrants in our main sample come from the Middle East or Northern Africa.

Second, to keep the identification strategy as clear as possible and to avoid having some individuals appear in the data more than once, we only included immigrants who had exactly one recorded immigration to Sweden. This restriction rules out circular migrants and reduces the sample by about 8 percent. Third, we required that immigrants be at least 16 years of age at the time of immigration. This restriction ensures that they were over 18 years of age at the time of the relevant election, which is a basic criterion for voting eligibility. Fourth, we dropped immigrants who died, out-migrated, or became Swedish citizens within three years of immigration. We imposed this restriction, since we think about the treatment as occurring when eligible immigrants receive their voting eligibility certificate by mail, which happens about one month before the election (i.e., around three years after immigration for those who immigrated close to a voting eligibility cutoff). Immigrants who died or out-migrated within three years were obviously never eligible to vote, and immigrants who managed to become Swedish citizens within three years would have had the right to vote no matter whether they immigrated just before or just after a voting eligibility cutoff. Thus, keeping these immigrants in the sample would risk attenuating any effects. In a robustness check we relaxed this restriction and obtained very similar results. Immigrants who out-migrated after spending three years in Sweden, however, were not dropped from the analysis, since we consider out-migration as a potential outcome. Fifth, we dropped immigrants for whom birth country was not available, since this information is needed to know what eligibility rules applied to a person. About 2 percent of our sample was dropped for this reason.

We analyze the effect of early, rather than late, voting eligibility on subsequent voter turnout, using a separate but very similar dataset.⁶ Just like the data described above, this dataset contains full information on all cases of immigration since 1969, as well as detailed information on various demographic and socioeconomic characteristics. Unfortunately, this second dataset includes information on citizenship status only starting in 1990, which is why we do not use it when analyzing the potential effects of voting rights on naturalizations. On the other hand, thanks to a recent effort to scan and digitize the complete electoral rolls for the 2010 general Swedish election, these data include validated individual-level turnout information for 95 percent of the electorate. The digitized data’s reliability is very high. When validated against a manually coded subset of these data, there is agreement in 99.7 percent of cases (see Lindgren, Oskarsson, and Persson 2019).

Empirical Strategy

The fundamental question we ask in this article is how the length of residency requirements for voting eligibility in local elections affects immigrant political integration. Ideally, from the viewpoint of causal inference, this question could be answered by conducting an experiment in which we randomly assign immigrants arriving in a country to groups with different residency requirements. In such a scenario, where some individuals would be eligible to vote shortly after arrival and others would have to wait longer, we could obtain the eligibility rules’ causal effects on political integration simply by comparing the means of political integration in the different groups. Obviously, such an experiment cannot be carried out for ethical reasons, but in countries where immigrants are eligible to vote in local elections after a certain period of residency, there will sometimes be large differences in the wait time to the first opportunity to vote among immigrants arriving almost at the same time. These differences arise because some immigrants fulfill the residency requirement just in time for an election while others barely miss it.

As previously explained, in Sweden, immigrants are eligible to vote in local elections after three years of residency. However, some immigrants satisfy this criterion just before an election and some just after. Thus, although they immigrated to Sweden at almost the same time, the first group needs to wait only three years for their first opportunity to vote, whereas the latter group must wait seven years. Consequently, the situation resembles an experiment where immigrants are randomly placed in two groups with three- and seven-year residency requirements, respectively. The differences in wait time created by the timing of immigration vis-à-vis the election cycle might appear perfect for the use of a regression discontinuity design, since it seems unlikely that immigrants would make special efforts to immigrate just before rather than just after a voting eligibility cutoff simply to be eligible to vote at an earlier date. Thus, we could gauge the causal effect of the length of residency requirements on political integration by comparing naturalization and turnout rates of immigrants who satisfied the three-year requirement just before and just after an election, since these two groups should be identical in all relevant regards.

A careful examination of our data, however, reveals that there is substantial variation in observable characteristics between immigrants on either side of the voting eligibility cutoffs, which makes it difficult to achieve balanced covariates around the cutoffs using a standard regression discontinuity design. These imbalances are most likely driven by seasonal variation in immigration, since we find very similar patterns in years where there are no wait-time discontinuities. To clarify, consider the voting eligibility cutoff on September 16, 2003 (see Table 1). Non-EU immigrants who arrived on this date at the latest were eligible to vote in the local elections in September 2006. Immigrants who arrived just after this date got their first opportunity to vote in September 2010. The problem is that immigrants who arrived in August and early September 2003 are observationally different from immigrants who arrived in late September and October. However, when we examine the immigration data from 2002 and 2004 in the same way, we find very similar differences between immigrants on the two sides of September 16. Even in adjacent years, when the September 16 cutoff is artificially constructed and has no bearing on wait time to the first opportunity to vote, we find imbalances in observed covariates (e.g., age and gender). Thus, there seems to be permanent differences between late-summer and early-fall immigrant arrivals in Sweden.

To handle this problem, we utilize the fact that the seasonal variation in immigration is stable over time and estimate the effect of early voting eligibility on political integration by means of a so-called “difference-in-discontinuities” design (see Grembi, Nannicini, and Troiano 2016, for the method’s original implementation). With this method we look at data from years both with and without voting eligibility cutoffs. In the years without cutoffs, we use artificially constructed cutoffs that would fall on the same dates of the year as the real cutoffs (see the example above). The method’s idea is to purge permanent differences between immigrants arriving before and after the relevant dates from the analysis. Equation 1 gives the formal empirical model:

Y_{i} = β_{0} + β_{1} I {[B e f o r e]}_{i} + β_{2} I {[B e f o r e]}_{i} \times I {[R e a l]}_{t} + γ_{t} + ε_{i} (1)

Y_i is some outcome (e.g., having naturalized at some point after immigration or voter turnout). $I [B e f o r e]$ is an indicator variable taking the value 1 if immigration occurred before a voting eligibility cutoff (artificially constructed or real). In the 2003 example, immigrants who arrived on September 16 or earlier in 2002, 2003, or 2004 would have the value 1 on this variable; later arrivals would have 0. $I [R e a l]$ is an indicator variable taking the value 1 if immigration occurred during a year with a real voting eligibility cutoff. In the 2003 example, immigrants who arrived in 2003 would have 1 on this variable, whereas immigrants who arrived in 2002 or 2004 would have 0. We always include two baseline years that surround the year with the real cutoff (1972 and 1974 for 1973 and so on). $γ_{t}$ represents year fixed effects, and $ε_{i}$ is an error term. The inclusion of the year fixed effects is why $I [R e a l]$ does not appear by itself in the model. $β_{1}$ captures the difference in the studied variable between immigrants on the two sides of the artificially constructed cutoffs. In the 2003 example with late-summer and early-fall immigrant arrivals, $β_{1}$ represents the mean difference (e.g., in age at immigration) between late-summer arrivals in 2002 and 2004 and early-fall arrivals in 2002 and 2004.

The parameter of most interest, $β_{2}$ , captures the difference in outcomes between immigrants on either side of the cutoffs in real cutoff years compared to the difference between immigrants on either side of the cutoffs in years with artificially constructed cutoffs. For instance, if age at immigration is the studied variable, $β_{2}$ . will be zero if the age difference between late-summer and early-fall arrivals in 2003 is identical to the mean age difference around the artificially constructed cutoffs in 2002 and 2004. Essentially, the estimate of $β_{2}$ tells us if the pattern around the cutoffs looks different in years when there actually are real voting eligibility cutoffs compared to the baseline years.

If $β_{2}$ is insignificant and close to zero when predetermined variables are used as outcomes, it supports the assumption that the empirical design can purge the analysis of imbalances in immigrant composition around the real cutoffs and isolate the potential effects of wait time to the first opportunity to vote on subsequent voting and citizenship acquisition. The predetermined variables to which we have access are gender, age at immigration, marital status at immigration, and origin country.⁷

Table 2 shows estimates of $β_{1}$ (Before) and $β_{2}$ (Before × Real) from models where the predetermined variables have been used as outcomes. Three different data windows have been used: immigration during the full year, immigration during the second half of the year (where all cutoffs appear), and immigration during a window consisting of 30 days on either side of the cutoffs. The overall picture from Table 2 suggests that the “difference-in-discontinuities” strategy is valid, since the estimates of $β_{2}$ generally are small and insignificant. However, the fact that two (weakly) significant estimates appear when using the full-year data window indicates that this specification should be used with some caution. Moreover, the estimates of $β_{1}$ show that those immigrants who immigrated before and after the artificially constructed cutoffs in the baseline years are significantly different from each other. This difference becomes particularly apparent when using the small data window in Panel C, which highlights the problem with using a regression discontinuity design, as discussed above.

Table 2.

Validity Check on Predetermined Variables.

Column:	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Outcome	Male	Age	Married	Bosnia	Asia	Sub-Saharan Africa	LatinAmerica	Canada/USA/Oceania	Middle Eastand N. Africa
Panel A: Data window is month 1–12
Before	0.008	–0.252	–0.034***	0.028	0.026*	–0.014	0.005	0.005*	–0.050***
	(0.006)	(0.232)	(0.006)	(0.021)	(0.014)	(0.009)	(0.007)	(0.003)	(0.012)

Before ×	–0.021*	0.414	0.007	–0.025	0.008	–0.000	0.005	0.009**	0.003
Real	(0.011)	(0.307)	(0.017)	(0.022)	(0.022)	(0.014)	(0.010)	(0.004)	(0.022)

N	391,042	391,042	391,042	391,042	391,042	391,042	391,042	391,042	391,042
Mean	0.507	31.370	0.584	0.072	0.195	0.120	0.110	0.044	0.458
Panel B: Data window is month 7–12
Before	0.013	–0.825***	–0.063***	0.006	0.063***	–0.021*	0.004	0.015***	–0.067***
	(0.008)	(0.209)	(0.016)	(0.008)	(0.021)	(0.011)	(0.003)	(0.002)	(0.013)

Before ×	–0.015	0.110	–0.001	–0.014	0.001	0.003	0.005	0.008*	–0.003
Real	(0.014)	(0.441)	(0.033)	(0.009)	(0.035)	(0.013)	(0.008)	(0.004)	(0.026)

N	193,232	193,232	193,232	193,232	193,232	193,232	193,232	193,232	193,232
Mean	0.509	30.948	0.572	0.046	0.212	0.123	0.106	0.049	0.464
Panel C: Data window is +/– 30 days from cutoff
Before	0.035***	–0.869***	–0.069***	0.003	0.077***	–0.026***	–0.010*	0.010***	–0.054***
	(0.013)	(0.291)	(0.023)	(0.007)	(0.022)	(0.008)	(0.005)	(0.003)	(0.012)

Before ×	–0.018	–0.248	–0.004	–0.011	–0.004	0.013	0.012	–0.001	–0.009
Real	(0.017)	(0.650)	(0.043)	(0.009)	(0.038)	(0.012)	(0.008)	(0.004)	(0.032)

N	74,134	74,134	74,134	74,134	74,134	74,134	74,134	74,134	74,134
Mean	0.532	30.599	0.542	0.045	0.240	0.121	0.102	0.048	0.444

Note: Asterisks indicate that the estimates are significantly different from zero at the ***1% level, **5% level, and *10% level. Standard errors clustered on year of immigration are in parentheses. The years included are 1972–1992, 1994–1996, 1998–2000, 2002–2004, and 2006–2008. See Section 1 in the Supplemental Appendix for a list of the included birth country groups. The outcome in column (1) takes the value 1 if the immigrant is a man. The outcome in column (2) is age at immigration measured in years. The outcome in column (3) takes the value 1 if the immigrant was married at the point of immigration. The outcomes in columns (4–9) take the value 1 if the immigrant has a birth code belonging to the respective categories.

We have chosen to cluster the standard errors on the year of immigration level (33 clusters), on the logic that those who immigrate in a given year are likely to be subject to the same set of institutions and general environment. Clustering on a lower level (e.g., year × month of immigration) would, of course, lead to even more similarity in immigration environment within clusters, but here we follow the conventional practice of clustering on the highest level where some correlation in error terms can be expected to exist.⁸

Results

Our presentation of results starts by discussing our findings regarding naturalizations and then proceeds to voter turnout.

Effects on Naturalizations

Table 3 presents the effects of being able to vote early rather than late on immigrants’ naturalization rate within 4 to 10 years after immigration. As for the predetermined variables, the model is estimated using three different data windows. All models, however, yield similar results: namely, that a shorter wait time to the first opportunity to vote in local elections is not significantly related to subsequent naturalization behavior (see the point estimates for “Before × Real”).

While the point estimates are generally negative and statistically insignificant, some effects are quite large, in particular the negative effect on becoming a Swedish citizen within seven years from the point of immigration (column 4). A 95-percent confidence interval around the estimate in column (4) of Panel C, for example, includes negative effects of up to 12 percentage points (the baseline naturalization rate after seven years is about 45 percent). Thus, we cannot rule out the existence of substantial negative effects on naturalization. Importantly, however, none of the results are consistent with the existence of large positive effects from earlier opportunities to vote on the subsequent likelihood of naturalization. Ninety-five percent confidence intervals around the “Before × Real” estimates generally only include rather small positive effects. The main message from Table 3, therefore, is that it is unlikely that earlier voting opportunities, on average, speed up the naturalization process.⁹

Table 3.

Main Results on Naturalizations.

Column:	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Outcome:	Becomes Swedish citizen within:
	4 years	5 years	6 years	7 years	8 years	9 years	10 years
Panel A: Data window is month 1–12
Before	–0.004	–0.002	–0.017	–0.010	–0.004	–0.007	–0.008
	(0.005)	(0.010)	(0.011)	(0.016)	(0.015)	(0.015)	(0.015)

Before ×	–0.001	–0.025*	–0.023	–0.031	–0.018	–0.015	–0.019
Real	(0.010)	(0.014)	(0.021)	(0.023)	(0.022)	(0.023)	(0.021)

N	391,042	391,042	391,042	359,646	297,203	297,203	297,203
Mean	0.083	0.210	0.405	0.497	0.549	0.600	0.636
Panel B: Data window is month 7–12
Before	–0.007	–0.022**	–0.058***	–0.048**	–0.038**	–0.042**	–0.044**
	(0.006)	(0.010)	(0.016)	(0.017)	(0.015)	(0.018)	(0.018)

Before ×	–0.007	–0.021	–0.020	–0.040	–0.016	–0.013	–0.017
Real	(0.009)	(0.013)	(0.031)	(0.031)	(0.022)	(0.025)	(0.025)

N	193,232	193,232	193,232	175,194	142,319	142,319	142,319
Mean	0.084	0.206	0.394	0.484	0.536	0.586	0.622
Panel C: Data window is +/– 30 days from cutoff
Before	–0.015**	–0.025*	–0.055**	–0.043**	–0.035*	–0.038	–0.038*
	(0.007)	(0.013)	(0.022)	(0.018)	(0.020)	(0.022)	(0.022)

Before ×	0.001	–0.017	–0.029	–0.053	–0.029	–0.027	–0.024
Real	(0.010)	(0.020)	(0.040)	(0.036)	(0.027)	(0.031)	(0.032)

N	74,134	74,134	74,134	66,577	54,038	54,038	54,038
Mean	0.074	0.190	0.363	0.454	0.517	0.565	0.601

Note: Asterisks indicate that the estimates are significantly different from zero at the ***1% level, **5% level, and *10% level. Standard errors clustered on year of immigration are in parentheses. This table includes the years 1972–1992, 1994–1996, 1998–2000, 2002–2004, and 2006–2008. Year 2008 was dropped in column (4). Years 2006–2008 were dropped in columns (5–7). These years were dropped because we can only follow naturalizations through 2014. See Section 1 in the Supplemental Appendix for a list of the included birth country groups. The outcome in column (1) takes the value 1 if the immigrant has acquired Swedish citizenship within four years from the point of immigration, and the other outcomes in columns (2–7) follow the same logic.

So far, we have only looked at average outcomes; since there is substantial heterogeneity among immigrants in the studied sample, it is still possible that some immigrant groups reacted to earlier voting opportunity while others did not. Therefore, we have performed heterogeneity analyses along several different dimensions. We explore if the results vary by gender, age, civil status, birth country, period of immigration, and type of immigration (we attempt to single out refugees). The overall picture is that there is very little heterogeneity in the results (see Supplemental Tables S2 and S4 in the Supplemental Appendix, where we summarize the results).

The results in Table 3 have been subjected to several robustness checks in the form of variations of the model and the sample. First, we re-estimated the model on a completely different sample — immigrants born in EU or Nordic countries (see Supplemental Table S7). These immigrants were excluded from the main analysis since voting eligibility rules changed for this group during the period of analysis. The estimated effects for EU nationals of being able to vote early are generally small and insignificant and thus in line with the results of the main analysis. Second, we explored whether early voting rights affected out-migration decisions but found no evidence of bias due to selective out-migration (Supplemental Table S8).

Third, we re-estimated the model with all predetermined variables from Table 2 used as controls (Supplemental Table S9). Under this arrangement, the estimates’ precision is marginally increased by including the control variables, but generally, the results are very similar to the results from the baseline specification. If anything, the estimates move marginally in a positive direction when we include the control variables. Fourth, we re-estimated the model, using alternative standard errors. Results with standard errors clustered on year of immigration and birth country code (two-way clustering), as well as on the combination of the two variables, are shown in Supplemental Tables S11 and S12. In both cases, precision is marginally improved, but the general interpretation of the analysis remains unchanged. Fifth, we included immigrants who died, out-migrated, or became Swedish citizens within three years from the point of immigration (Supplemental Table S15). Again, the results are very similar to those from the baseline specification. In summary, then, the finding that a shorter wait time to the first opportunity to vote is unlikely to positively affect the naturalization process is robust to substantial variations of the model and the sample.

Effects on Voter Turnout

This section turns our interest to an alternative indicator of political integration — voter turnout. The first column of Table 4 shows how earlier access to voting opportunities affected participation in the 2010 municipal election for those immigrating between 1972 and 2004, whereas the second column shows the effect on turnout in the 2010 parliamentary election for the same group of immigrants. The second column, thus, estimates the effect of earlier access to voting opportunities on immigrants who chose to naturalize. Because these models include immigrants who immigrated between 6 and 38 years before the studied election, the results in the first two columns capture a combination of long- and short-run effects of a shorter wait time to the first opportunity to vote. Similar to before, we present the results for three different data windows, shown in Panels A to C.

Table 4.

Main Results on Voting.

Column:	(1)	(2)	(3)	(4)
Outcome:	Voted in the 2010:
	Municipal election	National election	Municipal election	Municipal election
Type of effect:	Overall	Overall	Short-run	Short-run
Panel A: Data window is month 1–12
Before	0.004	0.002	0.001	0.001
	(0.005)	(0.005)	(0.007)	(0.007)

Before × Real	0.010	0.006	0.031***	0.031***
	(0.009)	(0.006)	(0.011)	(0.012)

N	228,086	201,830	39,788	38,350
Mean	0.667	0.695	0.541	0.544
Panel B: Data window is month 7–12
Before	–0.006	–0.000	–0.028***	–0.030***
	(0.008)	(0.005)	(0.009)	(0.009)

Before × Real	0.007	0.002	0.014	0.013
	(0.012)	(0.009)	(0.015)	(0.015)

N	107,277	92,964	20,493	19,719
Mean	0.659	0.692	0.522	0.525
Panel C: Data window is +/– 30 days from cutoff
Before	–0.011	0.000	–0.047***	–0.050***
	(0.011)	(0.006)	(0.014)	(0.015)

Before × Real	0.003	–0.006	0.008	0.009
	(0.014)	(0.010)	(0.025)	(0.025)

N	39,065	33,721	7,200	6,914
Mean	0.653	0.693	0.477	0.478

Note: Asterisks indicate that the estimates are significantly different from zero at the ***1% level, **5% level, and *10% level. Standard errors clustered on year of immigration are in parentheses (the standard errors in columns 3–4 are not clustered since we have too few clusters). Included years in columns (1–2) are 1972–1992, 1994–1996, 1998–2000, and 2002–2004. In columns (3–4) only those immigrating in 2002–2004 are included. See Section 1 in the Supplemental Appendix for a list of the included birth country groups. The outcome in column (1) takes the value 1 if the immigrant voted in the 2010 municipal election, and the other outcomes in columns (2–4) follow the same logic.

As can be seen from the table’s first two columns, we find little compelling evidence that immigrants who were eligible to vote sooner after arriving were more likely to vote in subsequent elections. Depending on specification, the point estimate of the eligibility cutoff (Before × Real) varies from –0.6 to 1.0 percentage points, and in no case does the coefficient come close to being statistically significant.

To judge from these results, it appears that there is no long-lasting overall effect on turnout from earlier voting eligibility. It could still be the case, however, that early access to voting opportunities can help speed up immigrant political integration by positively impacting voter turnout in the short run, as has been found in Norway (Ferwerda, Finseraas, and Bergh 2020).

In column 3 of Table 4, we, therefore, restrict attention to immigrants arriving between 2002 and 2004. By doing so, we can analyze the short-term effect of being eligible to vote in the 2006 municipal election on the turnout in the 2010 municipal election. Starting with the largest data window in Panel A, we find a positive and statistically significant effect of early access to voting. While there are no differences between pre- and post-cutoff arrivals in the baseline years of 2002 and 2004 (see the “Before” estimate), pre-cutoff arrivals in 2003 (the real cutoff year) are about 3 percentage points more likely to vote in the 2010 municipal election than post-cutoff arrivals.

However, when we reduce the immigration window around the cutoffs in Panels B and C of Table 4, the difference between pre- and post-cutoff arrivals in 2003 is no longer significantly different from the corresponding difference in 2002 and 2004. In fact, the point estimates are substantially reduced, although they remain positive (see the “Before × Real” estimates). Thus, the effect of having the first opportunity to vote after three years, relative to seven years, on subsequent short-run turnout appears to be very sensitive to the choice of specification. This sensitivity, combined with the observation from Table 2 that the full-year data window displayed more imbalances than the other windows, suggests that the estimate in Panel A should be interpreted with caution.

Thus, in contrast to earlier research (Ferwerda, Finseraas, and Bergh 2020), we find no convincing evidence that a shorter residency requirement for voting eligibility in local elections speeds up immigrant political integration by increasing turnout in the short to medium term. Moreover, because the great majority of immigrants included in our main sample came from highly non-democratic countries, the fact that the positive effect found in Norway was solely driven by immigrants originating from dictatorships or weak democracies should not help explain the differences in results between the two studies. For instance, of the 15 country groupings used in our analysis, in only two cases does the average score on democratic culture reach the threshold used by Ferwerda, Finseraas, and Bergh (2018) to distinguish between countries with weak and strong democratic culture.¹⁰ In Table 4‘s fourth column, we show that the short-term results also remain very similar when we exclude immigrants from the two groups with strong democratic culture (United States/Canada and Oceania). That is, we fail to find any effect of shorter residency requirements when also focusing on immigrants from non-democratic countries.

Similar to what we did for the naturalization outcomes, we also performed a number of sensitivity and heterogeneity analyses to check the robustness of this section’s findings. For reasons of space, these results are presented in the Supplemental Appendix, but the findings of these additional analyses are well in line with the ones reported here. For instance, the heterogeneity analyses show no compelling evidence of important treatment heterogeneity. We obtain fairly similar results when we varied the sample by gender, age, origin country, or time of immigration, although the pattern becomes somewhat noisier (Supplemental Table S3). We also re-estimated the main results on a sample of individuals identified as refugees, and the results are very similar to those of the main sample (Supplemental Table S5).¹¹ Further, we obtained virtually identical results when controlling for predetermined individual characteristics (Supplemental Table S10) and when using alternative levels for the clustering of the standard errors (Supplemental Tables S13 and S14).

Conclusions

Voting rights for non-naturalized immigrants are more common today than ever before, and non-naturalized immigrants have some voting rights in at least 65 countries (Immigrant Voting Project Website 2017). One argument in favor of enfranchising non-naturalized immigrants builds on the belief that the right to vote enhances immigrant integration into society. However, systematic studies on the relationship between voting eligibility rules for non-naturalized immigrants and immigrant social and political inclusion are lacking (c.f., Ferwerda, Finseraas, and Bergh 2018). In this article, we focused on one particular aspect of the voting eligibility rules — the importance of the length of the residency requirement for the timing and extent of immigrant formal political integration. Using data from Sweden, we studied whether immigrants who, due to the timing of their arrival vis-à-vis the election cycle, face shorter residency requirements for voting eligibility in local elections are more likely to integrate politically. We find little compelling evidence that such is the case. Our results show that immigrants who became eligible to vote after six to seven years in Sweden were as likely to naturalize and vote in future elections, in both the short and long run, as those who received the right to vote after only three years of residency.

Whereas we are the first to study how the length of residency requirements affects immigrants’ naturalization process and long-run turnout, a previous study has used a similar empirical design to study the effect on short-term voter turnout in Norway (Ferwerda, Finseraas, and Bergh 2018). Unlike us, they found a positive effect of shorter wait time on voting in the next local election for immigrants from non-democratic source countries. The most likely explanation for this difference in results is contextual differences between Sweden and Norway. In particular, we believe that the fact that national and local elections are held at the same time in Sweden but at different times in Norway can be of relevance because this difference means that voter turnout in local elections, for both natives and immigrants, is considerably higher in Sweden than in Norway. It is conceivable that differences in wait time mainly matter in contexts with a fairly low baseline turnout, potentially explaining why there appears to be an effect of the waiting time on turnout in the low-turnout context of Norway but not in Sweden, which has higher overall turnout.

Other potential explanations relate to between-country differences in the composition of immigrants, as well as differences in immigrants’ overall social and economic integration in Norway and Sweden. For instance, the fact that immigrants’ employment rate is higher in Norway than in Sweden (Pettersen and Østby 2013) could be hypothesized to ease immigrants’ political integration in Norway and increase the effectiveness of policies aimed at increasing immigrant political inclusion. To investigate whether this explanation actually holds, future research should attempt to investigate if the integrative effect of the voting eligibility rules depends on other factors and policies of relevance for immigrant political and economic outcomes.

Another issue that deserves attention in future studies is how the length of residency requirements for voting eligibility impacts more informal forms of political integration, such as political activism, associational engagement, and attitudes toward the host country (Jimènez 2011). Unfortunately, we were unable to investigate this important topic, due to a lack of adequate data, but earlier results are suggestive of positive effects on informal political integration from shorter wait time to the first opportunity to vote in the host country (Ferwerda, Finseraas, and Bergh 2018). Thus, it is still possible that shorter residency requirements can affect other types of political integration than those studied here.

What, then, can those involved in the debate about immigrant voting rights learn from this article? One important lesson, we believe, is that the conclusion drawn by Ferwerda, Finseraas, and Bergh (2020, 15) that “the early extension of political rights provides a useful policy tool to improve vulnerable groups’ engagement and interaction with the host society” does not necessarily apply to other countries. Although early voting rights may foster certain immigrants’ political integration in certain contexts, our results show that short residency requirements for voting eligibility in local elections are no universal panacea for immigrant political inclusion. We, therefore, believe that the argument that early voting rights for non-naturalized immigrants is desirable since it helps speed up immigrant political integration should be used with some care by those advocating such reforms.

That said, our results do not imply that non-naturalized immigrants’ enfranchisement is not valuable or important. First, our findings contradict one important argument against non-citizen voting rights: that the introduction of this right could be detrimental to political integration by decreasing immigrants’ incentives to naturalize. We found no evidence that such is the case. Second, and even more importantly, the purpose of extending voting rights to non-naturalized residents is clearly about more than merely facilitating individual political integration (Lenard 2015, 130). Thus, when deciding whether (local) voting rights should be extended to non-naturalized immigrants, we should not only consider the effect on individual political integration, but also take other potential benefits into account.

Ultimately, voting rights for non-naturalized immigrants concern their opportunity for political representation. Therefore, even if the extension of voting rights to non-citizen immigrants does not enhance individual political integration, it can nevertheless contribute to the symbolic and substantive representation of the immigrant group (Pitkin 1967). Symbolic representation denotes the feeling of being represented in the political system, and it is easy to imagine that access to early voting rights can increase immigrants’ sense of belonging to the polity. Substantive representation, instead, relates to the extent to which immigrants’ political interests are considered when important political decisions are made. It stands to reason that politicians are more likely to take non-citizen residents’ interests and preferences into account when this group is allowed to vote. Indeed, Vernby (2012) shows that introducing voting rights for non-citizens in Sweden in 1976 substantially raised spending on public services that disproportionately benefited the immigrant population. An important lesson of the current study is, therefore, that the future debate over non-naturalized immigrants’ enfranchisement should focus on the effects of voting rights on immigrant political representation rather than on the alleged effects on individual political integration, since these latter effects appear to be rather weak and context dependent.

Supplemental Material

Online_appendix,_final_version - The Role of Local Voting Rights for Non-Naturalized Immigrants: A Catalyst for Integration?

Online_appendix,_final_version for The Role of Local Voting Rights for Non-Naturalized Immigrants: A Catalyst for Integration? by Mattias Engdahl, Karl-Oskar Lindgren and Olof Rosenqvist in International Migration Review

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Mattias Engdahl

Supplemental Material

Supplemental material for this article is available online at

Notes

References

Adman

Strömblad

. 2000. “Resurser för Politisk Integration.” Integrationsverkets rapportserie 2000:16. Integrationsverket, Norrköping.

Åslund

Rooth

D-O.

. 2007. “Do When and Where Matter? Initial Labour Market Conditions and Immigrant Earnings.” The Economic Journal 117(518):422–48.

Bernitz

2013. “Access to Electoral Rights Sweden.” EUDO Citizenship Observatory, European University Institute.

Bevelander

2015. “Voting Participation of Immigrants in Sweden—A Cohort Analysis of the 2002, 2006 and 2010 Elections.” Journal of International Migration and Integration 16:61–80.

Bäck

2003. “Explaining and Predicting Coalition Outcomes.” European Journal of Political Research 42(4):441–72.

Cameron

A. C.

Miller

D. L.

. 2015. “A Practitioner’s Guide to Cluster-Robust Inference.” Journal of Human Resources 50(2):317–72.

Cancela

Geys

. 2016. “Explaining Voter Turnout: A Meta-Analysis of National and Subnational Elections.” Electoral Studies 42:264–75.

Coppock

Green

D. P.

. 2016. “Is Voting Habit Forming? New Evidence from Experiments and Regression Discontinuities.” American Journal of Political Science 60(4):1044–62.

Dancygier

R. M.

Lindgren

K. O.

Oskarsson

Vernby

. 2015. “Why Are Immigrants Underrepresented in Politics? Evidence from Sweden.” American Political Science Review 109(4):703–24.

10.

de Rooij

2012. “Patterns of Immigrant Political Participation: Explaining Differences in Types of Political Participation Between Immigrants and the Majority Population in Western Europe.” European Sociological Review 28(4):455–81.

11.

Dinas

2012. “The Formation of Voting Habits.” Journal of Elections, Public Opinion & Parties 22(4):431–56.

12.

Duleep

2015. “The Adjustment of Immigrants in the Labor Market.” In Handbook of the Economics of International Migration (Volume 1A), edited by Chiswick

B. R.

Miller

P. W.

, 105–82. Amsterdam: North Holland, Elsevier.

13.

Earnest

2003. “Noncitizen Voting Rights: A Survey of an Emerging Democratic Norm.” Paper prepared for the 2003 annual convention of the American Political Science Association, Philadelphia, August 28–31, 2003.

14.

Ferris

Hayduk

Richards

Schubert

Acri

. 2020. “Noncitizen Voting Rights in the Global Era: A Literature Review and Analysis.” Journal of International Migration and Integration 21:949–971.

15.

Ferwerda

Finseraas

Bergh

. 2020. “Voting Rights and Immigrant Incorporation: Evidence from Norway.” British Journal of Political Science 50:713–730.

16.

Garrman

2016. “Concurrent Elections and Turnout: Causal Estimates from a German Quasi-Experiment.” Journal of Economic Behavior and Organization 126: 167–78.

17.

Goodman

S. W.

2015. ”Conceptualizing and Measuring Citizenship and Integration Policy: Past Lessons and New Approaches.” Comparative Political Studies 48(14):1905–1941.

18.

Grembi

Nannicini

Troiano

. 2016. “Do Fiscal Rules Matter?” American Economic Journal: Applied Economics 8(3):1–30.

19.

Hainmueller

Hangartner

Lawrence

. 2016. “When Lives are Put on Hold: Lengthy Asylum Processes Decrease Employment Among Refugees.” Science Advances 2(8):e1600432.

20.

Hainmueller

Hangartner

Pietrantuono

. 2015. “Naturalization Fosters the Long-Term Political Integration of Immigrants.” Proceedings of the National Academy of Sciences 112(41):12651–56.

21.

Hammar

1990. Democracy and the Nation State. Aliens, Denizens, and Citizens in a World of International Migration. Aldershot: Avebury.

22.

Harper-Ho

2000. “Noncitizen Voting Rights: The History, the Law and Current Prospects for Change.” Law & Inequality: A Journal of Theory and Practice 18(2):271–322.

23.

Immigrant Voting Project Website. 2017. “Countries That Grant Some Voting Rights to Foreign Residents.” www.immigrantvoting.org.

24.

Jiménez

T. R.

2011 Migrants in the United States: How Well Are They Integrating in Society? Washington, DC: Migration Policy Institute.

25.

Kahanec

Tosun

. 2009. “Political Economy of Immigration in Germany: Attitudes and Citizenship Aspiration.” International Migration Review 43(2):263–91.

26.

Karlsdottìr

Norlén

Rispling

Randall

, eds. 2018. State of the Nordic Region 2018: Immigration and Integration Edition. Copenhagen: Nordic Council of Ministers. https://doi.org/10.6027/ANP2018-742.

27.

Kleven

Ø.

2015. “Lower Voter Turnout than Sweden and Denmark.” Samfunnspeilet 2.

28.

Konle-Seidl

Bolits

. 2016. Labour Market Integration of Refugees: Strategies and Good Practices. Brussels: European Parliament.

29.

Lenard

P. T.

2015. “Residence and the Right to Vote.” Journal of International Migration and Integration 16(1):119–32.

30.

Lindgren

K.-O.

Oskarsson

Persson

. 2019. “Enhancing Electoral Equality: Can Education Compensate for Family Background Differences in Voting Participation.” American Political Science Review 113(1):108–22.

31.

Meredith

2009. “Persistence in Political Participation.” Quarterly Journal of Political Science 4(3):187–209.

32.

Morales

2011. “Conceptualizing and Measuring Migrant’s Political Inclusion.” In Social Capital, Political Participation and Migration in Europe: Making Multicultural Democracy Work?, edited by Morales

Giugni

, 19–42. Basingstoke: Palgrave Macmillan.

33.

Munro

2008. “Integration through Participation: Noncitizen Resident Voting Rights in an Era of Globalization.” Journal of International Migration and Integration 9(1):63–80.

34.

Oscarsson

2001. Skilda valdagar och vårval? Regeringskansliet, SOU 2001:65 Proposition 1975/76: 23 Om kommunal rösträtt för invandrare.

35.

Pettersen

S. V.

Østby

. 2013. “Immigrants in Norway, Sweden, and Denmark.” Samfunnspeilet 5. https://www.ssb.no/en/befolkning/artikler-og-publikasjoner/_attachment/204333?_ts=1497ab86428

36.

Pickus

1998. Immigration and Citizenship in the Twenty-First Century. Lanham, MD: Rowman & Littlefield.

37.

Pitkin

, 1967. The Concept of Representation. Berkeley: University of California Press.

38.

Ruedin

2018. “Participation in Local Elections: Why Don’t Immigrants Vote More?” Parliamentary Affairs 71(2):243–62.

39.

SCB. 2019. “Valdeltagande i kommunfullmäktigeval 1976–2014 bland utländska medborgare efter kön.” https://www.scb.se/hitta-statistik/statistik-efter-amne/demokrati/allmanna-val/allmanna-val-valdeltagandeundersokningen/pong/tabell-och-diagram/historisk-statistik/valdeltagande-i-kommunfullmaktigval-19762014-bland-utlandska-medborgare-efter-kon/

40.

Schuck

1989. “Membership in the Liberal Polity: The Devaluation of American Citizenship.” Georgetown Immigration Law Journal 3(1):1–18.

41.

Seidle

P. T.

2015. “Local Voting Rights for Non-nationals: Experience in Sweden, the Netherlands, and Belgium.” Journal of International Migration and Integration 16(1):27–42.

42.

Slotwinski

Stutzer

Gorinas

. 2017. “Democratic Involvement and Immigrants’ Compliance with the Law.” IZA DP No. 10550.

43.

Street

2017. “The Political Effects of Immigrant Naturalization.” International Migration Review 51(2):323–43.

44.

Vernby

2012. “Inclusion and Public Policy: Evidence from Sweden’s Introduction of Noncitizen Suffrage.” American Journal of Political Science 57(1):15–29.

45.

Wass

Blais

Morin-Chassé

Weide

. 2015. “Engaging Immigrants? Examining the Correlates of Electoral Participation among Voters with Migration Backgrounds.” Journal of Elections, Public Opinion and Parties 25(4):407–24.

46.

Wolbrecht

Hero

. 2005. The Politics of Democratic Inclusion. Philadelphia, PA: Temple University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB