A foot in both countries: the effect of origin-country enfranchisement on migrants’ political interest and partisanship

ABSTRACT

How does emigrant enfranchisement affect immigrants’ political interest and partisanship? The ballot access granted by the origin country has both a symbolic and informational value for individuals. It returns political dignity to emigrants and transforms them from passive subjects to political actors, targeted by political campaigns and get-out-the-vote efforts. For the majority of the enfranchised, this empowerment spills over onto their attitude not only as emigrants toward the origin country but also as immigrants toward the country of residence. However, those international migrants who perceive strong rejection from the local environment are pulled away from residence-country politics, as the stark comparison between rights granted extraterritorially and political isolation around them becomes even starker. By combining existing cross-national surveys with data regarding electoral regulations, and an original survey of Latin American immigrants in the United States, this paper offers a unique insight into the political attitudes of individuals who are caught between origin and residence countries.

KEYWORDS:

• enfranchisement
• migrants
• voting abroad
• extraterritorial voting
• political attitudes

Acknowledgements

I wish to thank Hadas Aron, Naor Ben-Yehoyada, Victoria Finn, Peter John Loewen, John Marshall, Johanna Peltoniemi, Carlo Prato, Sebastián Umpierrez de Reguero, as well as the journal’s reviewers for their helpful suggestions and comments on this and earlier versions of the paper. Any errors remain my own.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Role of the international community (Turcu and Urbatsch 2015), remittances (Iskander 2015; Itzigsohn 2000), political competition, ideology, and partisanship (Brand 2010; Joppke, MacGilvray, and Joppke 2005; Mullings 2011; Rhodes and Harutyunyan 2010;) (Wegschaider 2023), role of well-organized lobbies of emigrants (Itzigsohn 2000; Smith 2003) and the perception of the support for the incumbent (Wellman 2021)

2 Gamlen (2015)

3 Less common methods for voting abroad are proxy voting and electronic voting. Estonia has implemented electronic/internet since its Parliamentary election in 2007 (Ellis et al. 2007) and Panama since 2014 (Comisión Voto Especial and Voto en el Extranjero 2021).

4 See endnote 1

5 See endnote 2

6 A working paper by Wellman and Paarlberg (2020) is an exception and analyses the effect of get-out-the-vote messages on turnout.

7 Material collected by the author.

8 See a detailed explanation about coding in Appendix.

9 The survey company Latino Decisions fielded the survey online in late November/December 2019. IRB approval from Columbia University. Columbia University protocol IRB-AAAS3491, final modification approval on 11/12/2019. Funding from Institute for Social and Economic Research and Policy in the Faculty of Arts and Sciences (ISERP), Columbia University ‘Seed Grant’ for 2019.

10 My samples in both analyses consists of those who were born in a different country from where they reside (first-generation migrants) since my focus is their experience as enfranchised emigrants. I want to keep constant this important element of the connection with the origin country: the place of birth. There are important nuances of the interaction between citizenship status, immigration generation, and the regulations in different residence countries that are outside the scope of this paper but should be the subject of a separate study. Nonetheless, robustness checks with more inclusive samples are in the Appendix.

11 Recoded: higher values represent more interest.

12 In the Online Appendix, I explore another version of this variable: Access to Ballot, a six-point scale that describes the easiness of access to the voting process from abroad through the mode of casting the ballots: from no access to internet voting. Moreover, since I am trying to capture respondents’ perception of inclusion, I believe that considering whether the law was implemented when the individual was interviewed or whether individuals in different residence countries could have a reasonable expectation of being included in the process are important nuances that need to be accounted for in the coding. For a similar approach, see Wellman (2021). Other scholarly work that analyses this same phenomenon at an aggregate level (national/institutional levels) has implemented alternative classifications (Burgess and Tyburski 2020; Hartmann 2015; Palop-García and Pedroza 2019). A longer discussion on the rationale of my choice of coding for enfranchisment can be found in the Online Appendix.

13 41 origin-countries/citizenships show a variation of Enfranchisement within country, while 52 show a variation of Access to Ballot (OnlineAppendix). A robustness check dropping the citizenship fixed effects and pooling together observations in 15 sub-regions (e.g. Eastern Europe, North Africa/Middle East, South America, South Eastern Europe, Western Africa and so on) is included in the Online Appendix Table 7A. For details about the grouping see Table 4A. Another robustness check with two smaller sample with only citizenship with 10 or 30 respondents each can be found in the Online Appendix Table 11A.

14 For similar empirical strategies that leverage surveys’ temporal variation see Branton et al. (2015) or Wallace, Zepeda-Millán, and Jones-Correa (2014).

15 Whether the respondent in the last 7 days was ‘In paid work (or away temporarily) (employee, self-employed, working for your family business)’

16 This variable has been found to have important effects on trust in residence-country institutions (Maxwell 2013; Superti and Gidron 2021), and being correlated with higher (Ahmadov and Sasse 2016) or lower (Chaudhary 2018) engagement with origin-country politics.

17 Table 8A in the Online Appendix shows the robustness of the results to alternative clustering of standard errors. It uses a much more conservative clustering at the citizenship level and shows that only in the case of news consumption the standard error of the Enfranchisement coefficient becomes too large, and the corresponding p value is than larger than .10.

18 This variable as well as Not Caring about Origin El are not present in the previous cross-national analysis, thus excluded in some specifications.