The Role of Electoral Saliency in Expat Voter Mobilisation
Oliwia Borek
Introduction
Voter turnout has generated a vast level of scholarly interest, with literature in this area rich in both theoretical and empirical studies which seek to explain why individuals choose to vote in elections. However, very few efforts have been made to distinguish potential differences in the mobilisation of expat and domestic voter groups. This study seeks to address this gap by examining the turnout patterns of both voter groups in response to changing electoral saliency. The study uses data from the official electoral databases of governments in Poland, France, and the Czech Republic to compare the patterns of electoral turnout for expat and domestic voters in these countries. Focusing on presidential and parliamentary elections, held between 2005 and 2022, the study evaluates the effects of electoral saliency on both voter groups based on these findings. The study finds some evidence to suggest that changing electoral saliency has a greater effect on expat voter turnout than on domestic voter turnout, although this effect varies significantly across the three countries considered. Through its findings, the study demonstrates the potential for a wide variety of further research into expat voter behaviours and makes a case for their future inclusion in voter turnout literature.
Literature Review
Electoral Saliency
There exists a wide variety of traditions in explaining mobilisation and voting patterns in the literature. Different voter models emphasise socioeconomic, institutional, and more recently, psychological factors, in determining voter turnout, with a vast number of both theoretical and empirical studies existing in this area (Blais, 2006; Blais and St-Vincent, 2010; Stockemer, 2016). This study assumes the rational voter model in explaining electoral turnout due to its compatibility with the expectation that increased electoral saliency will result in greater voter mobilisation, as will be explained in this section.
Rational voter theories assert that individuals make the decision to vote or not to vote in an election based on a reasoned cost-benefit analysis. An individual is thereby more likely to vote when the costs of voting are low, and the expected benefits of voting are high (Aldrich, 1993; Cox, 1997; Downs, 1957). While a discussion of voting costs will prove significant for our final analysis, the study focuses primarily on the impact of voting benefits on electoral turnout, with electoral saliency increasing the perceived benefits of voting for the individual.
The benefits of voting refer to what the individual is likely to receive if their chosen candidate wins an election. As the likelihood of an individual vote shaping the outcome of an election is miniscule, the voter is unlikely to reap the benefits of a favourable electoral outcome directly as a result of their own vote. This observation has inspired an array of literature on what is known as the ‘rational voter paradox’; as the expected benefits of voting are always negligible on an individual level, voting costs should always exceed benefits for the rational voter. (Downs, 1957; Feddersen, 2004). Individuals consequently require an additional push-factor which will motivate them to vote despite the low likelihood that they will personally alter the course of the election. Electoral saliency can act as such a push-factor. The greater the stakes of an election, the more benefits an individual has to gain from a favourable electoral outcome. The higher expected benefits of voting in a salient election can make the individual view their vote as more significant, even if it remains unlikely that the vote itself will shape the outcome of the election (Ferejohn and Fiorina, 1974; Kanazwa, 1998). The assertion that higher electoral saliency will result in increased voter turnout is therefore compatible with the rational voter model.
While rational voting theories have been critiqued in the literature—with some scholars emphasising that the model cannot solely explain a wide range of voting behaviours—it is worth nothing that the assumption that turnout will increase with electoral saliency has also seen a great level of empirical support (Geys, 2006). Electoral saliency has been positively linked with voter turnout in a vast number of empirical studies, with many scholars identifying it as one of the most significant determinants of voter turnout. A meta-analysis of voter turnout literature between 2004 and 2013 found that electoral saliency was positively linked with voter turnout in 90% of cases (Stockemer, 2016). Another study found that in countries within the European Union, electoral saliency increases voter turnout by 18%, on average (Franklin, 1996). The expectation that voter turnout will increase with electoral saliency should therefore withstand critiques of the rational voter theories in which it is grounded.
Expat Voters
Expat voters constitute a largely under-researched demographic in political science literature. As such, very little is known about expat voting behaviours. Some existing studies have evaluated the efforts of political parties in mobilising their expat voter base, however, these have largely focused on the types of parties likely to engage in such efforts and the mobilisation tactics they are likely to employ (Ostergaard-Nielsen and Ciornei, 2019; Turcu and Urbatsch, 2020). Other studies have compared the voting preferences of expat voters to those of domestic voters, however, these consider only the mobilised section of the expat voter base and make no reference to the reasons for their initial mobilisation (Lawson, 2003; Goldberg, 2021). As such, no long-term mobilisation patterns of expat voters have been identified by the literature.
Electoral saliency is expected to have a greater impact on expat voter turnout than domestic voter turnout for two reasons. Firstly, expats receive fewer benefits as a result of a favourable electoral outcome than do domestic voters (Blankart and Margraf, 2011). Electoral outcomes have a lesser impact on expat voters who are largely governed by the laws and systems of the country in which they reside. The expat’s decision to vote will therefore be more dependent on factors such as electoral saliency, which increases the stakes of the election for the expat voter and deems their vote more significant. Secondly, expats are likely to experience greater voting costs in comparison to domestic voters, such as difficult registration procedures and greater distances to the polls (Baubock, 2007). Studies have found that such costs are more likely to deter voters when electoral saliency is low, while in elections of high saliency they are largely overlooked (Ortford et al., 2009). As a result of these two factors, expat voter turnout can be expected to show greater fluctuations in response to changing electoral saliency.
Theory
Defining Electoral Saliency
To understand the effects of electoral saliency on expat voter mobilisation, we must first define what it means for an election to be of a high saliency—that is, what are the common characteristics of highly salient elections. For the purpose of this study, electoral saliency will be evaluated based on the circumstances in which any individual election takes place. Elections of the highest level of saliency will be characterised as taking place in exceptional political circumstances.
Such ‘exceptional’ circumstances can be divided into two categories. Firstly, unique electoral circumstances can arise directly as a result of recent political events involving state actors or other electoral candidates. This includes elections which take place following a period of controversial government rule—characterised for example by the breaking of democratic norms by the ruling party or the government’s implementation of highly contested policies. It also includes elections which are significantly impacted by political scandals—such as allegations of corruption—involving prominent electoral candidates (Bågenholm, 2013). Secondly, exceptional political circumstances can arise as a result of unprecedented events occurring outside of the electoral process. Major national or global events will set the wider context in which an election takes place and have a prominent impact on the election campaign. This category includes, for example, elections taking place in times of national health or security crises (Cześnik, 2010).
Exceptional political circumstances raise the electoral stakes for voters and provide them with distinct alternatives from which to choose, in the election campaign. Where a party or its candidate is involved in political controversy, for example, voters will be mobilised to vote either in favour of or in opposition to their way of rule. Where the exceptional circumstances relate to an unprecedented national or global event, voters will be mobilised in favour of a specific solution or way to move forward as a nation (Kriesi, 2008). Party voter bases are consequently mobilised to a greater extent in support of their way of politics, while frequently immobilised parts of the electorate are more likely to recognise the importance of their electoral engagement. Such heightened mobilisation will also increase electoral competition, further contributing to the saliency of the electoral cycle (Cześnik, Miśta and Żerkowska-Balas, 2020).
Evaluating the Saliency of Individual Elections
As per the above definition, we can now distinguish between elections of low saliency and those of high saliency within our election sample. This section will discuss the exceptional circumstances which render a number of these elections as highly salient. All elections which are not included in the following discussion have been identified as being of a lower electoral saliency and will be treated as such in our analysis. These elections of low saliency are as follows: in Poland—the parliamentary elections of 2005, 2011, and 2015 and the presidential elections of 2005 and 2015; in the Czech Republic—the parliamentary elections of 2010 and 2017, and the presidential election of 2013; in France—the parliamentary elections of 2012, 2017, and 2022 and the 2012 presidential election.
Poland
Parliamentary Election 2007: This election was marked by political scandal, taking place two years early following the early dissolution of government in response to a series of corruption allegations, aimed at the leader of a key coalition partner (SRP) to the ruling party (PiS). Additionally, a key voting issue of this election campaign concerned the breaking of a series of democratic norms by the ruling party during its two years in power. This included attacks on the independence of the Central Bank, as well as the questioning of verdicts given by the Constitutional Court. Both issues contributed significantly to the saliency of this election (Markowski and Cześnik, 2011).
Presidential Election 2010: The 2010 presidential election took place in the aftermath of the Smoleńsk disaster—a crash of the Polish Air Force 101 flight which killed all of its 96 passengers. Amongst the deceased was the then current President and candidate for re-election, Lech Kaczyński. In accordance with the Polish Constitution, the speaker of the parliament, Bronislaw Komorowski, became the acting president of Poland following the disaster and became obliged to call an early presidential election to take place within 60 days. The Smoleńsk disaster had a significant impact on the saliency of the election in which the acting president Bronislaw Komorowski and Jarosław Kaczyński, the brother of the late president, emerged as the primary candidates (Cześnik, 2014).
Parliamentary Election 2019: The 2019 election followed an electoral cycle marked by a series of undemocratic actions by the ruling party (PiS) which were widely criticised by opposition parties, democratic watchdogs, and the European Union. These included attacks on the Polish Constitutional Tribunal, the nation’s universities and its public media outlets. The issue of democratic backsliding increased the saliency of the election for voters, who proved divided in their views on the political future of the country (Markowski, 2019).
Presidential Election 2020: The 2020 presidential election took place in a political context similar to that of the 2019 parliamentary election - characterised by several years of democratic backsliding as a result of the actions of the ruling party (PiS). The election furthermore took place in exceptional national circumstances following the outbreak of the COVID-19 pandemic. The nature and timing of the election was widely debated as a result of the crisis, which ultimately took place two months after its originally scheduled date (Vashchanka, 2020).
Czech Republic
Parliamentary Election 2006: The 2006 election followed an electoral cycle of highly unstable rule by the governing party (ČSSD), with the premiership changing three times as a result of declining public support and financial scandal. The electoral campaign itself was characterised by a series of scandals. This included a physical altercation between members of the two frontrunner parties and the ‘Kubice report’—leaked days before the election—which accused key members of the ruling party of involvement in organised crime (Plecita-Vlachova and Stegmaier, 2008).
Parliamentary Election 2013: The 2013 election followed the eruption of a major corruption scandal involving the Prime Minister and his subsequent resignation. A caretaker government called by the President then failed to pass a vote of confidence in the Czech Chamber of Deputies, leading to its dissolution. The election was called a year early in light of these events, with the consequently rushed electoral campaign marked by the instability of its traditional parties and coalitions (Havlik, 2014).
Presidential Election 2018: The saliency of the 2018 election can be attributed largely to the controversial nature of the incumbent Czech president, who faced criticism over the divisive nature of his far-right populist rhetoric as well as his close ties with the Czech Prime Minister, who was at the time involved in a number of corruption scandals. The election campaign itself witnessed a series of controversies, with many nominees facing allegations of financial misconduct and corruption (Rovny, 2018).
Parliamentary Election 2021: The 2021 election took place in a highly salient political environment following four-years of controversial populist rule. The ruling party (ANO) engaged in a series of undemocratic actions during its term, including attempts to increase its control over the courts and the media. The incumbent Prime Minister had also faced a series of corruption allegations throughout the electoral term and was implicated under the Pandora Papers for tax evasion days prior to the election (Rovny, 2021).
France
Presidential Election 2017: The 2017 presidential election marked the first time an incumbent French president did not run for re-election, and the election campaign took a sudden turn following the eruption of a major financial scandal involving favoured-to-win candidate Francois Fillon. Subsequently, for the first time in French history, neither candidate of the two traditional governing parties (PS & LR) progressed to the second round of the election. (Evans and Ivaldi, 2018). The election additionally took place in a national state of emergency, following the 2015 Paris terrorist attacks. This remained a prominent issue throughout the election campaign, with frontrunners Emmanuel Macron (RE) and Marine Le Pen (RN) representing vastly different approaches to dealing with the issue of terrorism in the country.
Presidential Election 2022: The 2022 presidential election took place two months after Russia’s invasion of Ukraine, with questions regarding France’s role in responding to the crisis subsequently a key issue during the electoral campaign. The emerging frontrunners were again Macron and Le Pen, who remained diametrically opposed on key voting issues and presented two very different visions for the future of France (Chopin and Faure, 2021). Le Pen’s previous ties to the Kremlin and her anti-European stance were viewed as particularly controversial in light of Russia’s invasion of Ukraine.
Evaluation of Expat Voter Procedures
Each of the three countries looked at in this study has in place its own set of expat voting rules and procedures. A discussion of these is significant as these procedures will inform which voter figures are used for our analysis and allow us to later evaluate the potential effects of voting costs on our hypotheses.
Our analysis will focus on expat registration figures for Poland, total number of expat votes cast in the Czech Republic, and expat voter turnout in France. These will be compared against the overall voter turnout figures in each country. While the overall voter turnout figures will include both domestic and expat voters, expat voters make up only a very small proportion of the total voting population in each of the three countries. Overall voter turnout will therefore be used to evaluate the behaviours of domestic voters.
Poland
No continuous Register of Voters exists for Polish expats, who must instead register to vote prior to every individual election. Expats can complete their registration in person, in writing, by phone, or electronically. In presidential elections, those who have registered for the first round are automatically registered for both rounds, however, voters can also register for the second round only (Korzec and Pudzianowska, 2021). As a result of this demanding registration procedure, our analysis will focus on the total numbers of votes cast by expats in Polish elections rather than expat voter turnout figures, which are given as a proportion of those registered and do not reflect the total eligible voting population. Fluctuations as a result of varying levels of saliency will be best reflected in the number of expats who choose to register in each electoral cycle, and subsequently choose to cast their vote.
Postal voting is available to expats in presidential elections and was in place for parliamentary elections between 2011 and 2018. However, votes can now only be cast at expat polling stations in parliamentary elections—the number and locations of which vary in each individual election. This imposes additional voting costs on Polish expats (Korzec and Pudzianowska, 2021).
Czech Republic
Czech expats must register to be placed on a Special List of Voters. This process is not automatic and can be done either in person or in writing. The registration does not have to be renewed every election as in the case of Polish expats—once registered, voters remain on the list for a period of ten years (Linek, 2018). The registration is valid for both parliamentary and presidential elections. This is significant for our analysis as Czech voters participated in presidential elections for the first time in 2013 (before this, the president was elected by a vote in the Chamber of Deputies and Senate). As a result, expat turnout figures for parliamentary elections become skewed from 2013 onwards—now also including expats who register to vote in presidential elections. To account for this, our analysis will also focus primarily on the total number of votes cast by expats in Czech elections.
Czech expats face additional voting costs in both parliamentary and presidential elections due to the limited voting methods available to them. Expats can vote at their local embassy or apply for a voter card with which they can vote in any electoral district in the Czech Republic. Expats, therefore, only have the option to vote in person, requiring them to travel to an embassy or their home country for the day of the election (Linek, 2018).
France
Registration for French expats is automatic. Expat voter turnout figures therefore reflect engagement levels across the total expat population and can be used for our analysis. Voting costs are exceptionally low for French expats as a result of both their automatic registration and the variety of voting options available to the demographic. In both parliamentary and presidential elections, expats can vote at embassies, by post, or by proxy. In parliamentary elections, electronic voting is also available. French expats also receive special representation in the lower chamber of the French parliament—electing eleven of its seats. This differs from Poland and the Czech Republic, where expat votes are pooled with those of domestic voters (Arrighi, 2018).
Hypotheses
H1: Expat voters show greater fluctuations in turnout, as a result of changing electoral saliency, than do domestic voter groups.
H1a: Expat voters are mobilised to a greater extent than domestic voters in salient electoral cycles—when turnout increases amongst domestic voters, turnout amongst expats increases by a more significant amount.
H1b: Expat voters are more likely than domestic voters to disengage from the electoral process in less salient electoral cycles—when turnout decreases amongst domestic voters, turnout amongst expats decreases by a more significant amount.
Methodology
All of the data used in this study will be collected from the official electoral databases of each country’s respective government. This includes overall voter turnout figures for each election, expat voter turnout figures for elections held in France and total votes cast by expat voters in Poland and the Czech Republic. Based on these figures, the change in turnout in each election cycle (in comparison to the most recently held election) will be calculated for both domestic and expat voters. In the case of domestic voters across all three countries and expat voters in France, this change will be represented by a percentage point difference in reported turnout. In the case of expat voters in Poland and the Czech Republic, this change will be represented by the percentage change in the number of votes cast by expat voters. For all elections which took place over two rounds, the change in voter turnout will be calculated based on the average of the 1st and 2nd round turnout.
The study will measure the mean voter turnout difference for both voter groups across elections of high saliency and those of low saliency. It is expected that elections of high saliency will have a positive mean change in voter turnout while this mean change will be negative in elections of low saliency. Based on these figures, it will be possible to identify whether these mean changes in voter turnout are greater for expat voter groups and are thereby in line with the study’s hypotheses. A t-test will be further used to validate the conclusions drawn from this process—however, it is important to note that due to the very low sample size of elections considered by this study, the usefulness of this method in identifying significant changes within the data is limited. As a result, the study’s analysis will not disregard identified differences across election cycles on an individual or mean level based on the t-tests conducted alone.
Across electoral cycles of high saliency in Poland, expat voters showed a greater average positive change in turnout (M=196.1, S.D.=111.1) than domestic voters (M=10.8, S.D.=4.3), t=-3.333, p=0.016.
Across electoral cycles of low saliency in Poland, expat voters showed a greater negative change in turnout (M=-17.54, S.D.=7.25) than domestic voters (M=-2, S.D.=3.62), t=3.322, p=0.045.
Across electoral cycles of high saliency in the Czech Republic, expat voters showed a greater average positive change in turnout (M=56.63, S.D.=59.5) than domestic voters (M=1.96, S.D.=4.4), t=1.587, p=0.19.
Across electoral cycles of low saliency in the Czech Republic, expat voters did not show a greater negative change in turnout (M=10.75, S.D.=15.8) than domestic voters (M=-0.435, S.D.=2), t=-0.993, p=0.4.
Across electoral cycles of high saliency in France, expat voters showed a smaller negative change in turnout (M=-1.875, S.D.=7.26) than domestic voters (M=-3.54, S.D.=1.7), t=-0.316, p=0.7.
Across electoral cycles of low saliency in France, expat voters did not show a greater negative change in turnout (M=1.5, S.D.=6.2) than domestic voters (M=-4.75, S.D.=8.4), t=-0.847, p=0.49.
Discussion
Poland
Expat voters proved more responsive to electoral saliency across all but one of the elections considered. In all four elections of high saliency, expat voters were mobilised to a much greater extent than were voters at home, with three of these elections seeing expat vote increases of over 100%. Expat voters also disengaged from the electoral process in greater numbers than did domestic voters in elections of a low saliency. This was true of two out of three of such elections considered, with expat voter turnout decreasing by up to 23%. The only election which did not follow this pattern was the 2015 parliamentary election, which was identified as a significant outlier and excluded from our analysis. Nevertheless, this is the only election which does not fit the hypotheses and the general fluctuations seen in expat voter turnout in response to changing electoral saliency remain greater than those seen in domestic voters.
In light of these findings, it is worth considering Poland’s expat population changes between these election cycles. Some proportion of the colossal rises in expat voter turnout can be accounted for by the changing number of Polish expats abroad. This is particularly the case with regards to the 2007 parliamentary elections and the 2010 presidential elections, which followed a period of mass emigration following Poland’s accession to the EU (World Bank, 2022). However, the effect of electoral saliency on expat voter turnout cannot be disregarded due to the significant de-mobilisation of expat voters in subsequent elections. Both the succeeding 2011 parliamentary and 2015 presidential elections saw a fall in expat voter turnout of close to 20%, despite continued growth of Poland’s expat population at this time (World Bank, 2022). Additionally, the highly salient elections of 2019 and 2020 saw a mass mobilisation of expat voters—with voter turnout increasing by 81% and 160% respectively—despite following a period of very low emigration from Poland (World Bank, 2022). Our findings therefore support the hypothesis that electoral saliency impacts on expat voters to a greater extent than domestic voters.
Czech Republic
Based on the data on both parliamentary and presidential elections, there is some evidence to suggest that Czech expat voters are mobilised to a greater extent than domestic voters in elections of high saliency. Expat voter turnout increased by a greater amount than did overall voter turnout across all salient elections. Some of these differences in mobilisation were also substantial—in the second round of the 2018 election, the increase in expat votes cast was 120 percentage points higher than the increase seen in overall turnout. The data relating to Czech elections of low saliency is less clear. Czech expat voters were not responsive to the saliency of the 2013 parliamentary election and saw lower levels of de-mobilisation than domestic voters in 2017. Only the first part of our hypothesis is therefore, to some extent, supported by data from the Czech Republic.
Our findings relating to Czech elections of high saliency are not limited by the emigration figures of Czech nationals in the considered timeframe. Both the 2013 and 2018 elections saw significant expat voter turnout increases of between 28% and 124%, despite following periods of very low levels of emigration out of the country (World Bank, 2022).
France
The data on elections in France does not to any extent support any part of our hypotheses. Expat voter turnout was responsive to the saliency of the 2017 election—increasing by a few percent in both rounds of the election. This pattern was not seen, however, with respect to the 2022 presidential election. Despite its high saliency, the election witnessed a fall in both domestic and expat voter turnout, with de-mobilisation proving greater amongst the expat voter base in both rounds of the election. While expat voters therefore may have been more responsive to the electoral saliency of the 2017 election, this trend was reversed in the subsequent 2022 election. It is therefore unlikely that French expat voters are more responsive to electoral saliency in the long-term.
All parliamentary elections considered (2012, 2017 and 2022) were identified as being of a lower saliency. Accordingly, expat voter turnout remained significantly lower than overall voter turnout across all three elections—however, voter turnout in fact saw less fluctuation than did turnout amongst domestic voters. In the 2017 parliamentary election, overall turnout fell by 7.84 percentage points more than did turnout amongst expats. Meanwhile, in 2022, expat voter turnout increased despite the low saliency of the election—a trend not reflected in the turnout of domestic voters. Expat voters therefore in fact appear less responsive to the low saliency of parliamentary elections in this time period.
Comparative Analysis
The degree to which saliency impacted on expat voter behaviours varied vastly between the three countries considered by this study. Our hypotheses were supported by data from Poland and to a lesser extent the Czech Republic. Poland saw the greatest levels of fluctuation in expat voter turnout in response to changing electoral saliency—expat voters were mobilised to a much greater extent than domestic voters in salient elections and disengaged from the electoral process in much greater numbers in elections of low saliency. In the Czech Republic, expat voters saw greater mobilisation in elections of high saliency, but their patterns of de-mobilisation were less strictly responsive to diminishing saliency. Meanwhile, our hypotheses saw very little support from data on French elections. Saliency had a minimal effect on expat voter turnout in France—this effect was seen only in the case of the 2017 presidential election and the identified turnout fluctuations were much lower than in the case of Poland and the Czech Republic. Expat voter turnout in France overall did not appear more responsive to changing electoral saliency than was domestic voter turnout.
This variation may reflect differences in the voting costs faced by expat voters in the three countries. Accordingly, voting costs are highest in Poland—with voters having to register every election and postal voting only available in presidential elections—and lowest in France, where expat voters are registered automatically and have the choice of embassy, postal, and electronic voting options. It seems plausible that higher voting costs would increase the effect of electoral saliency on expat turnout behaviours, with expats being more likely to pick and choose the elections they feel are worth participating in as a result of these added costs. This could thereby explain why France has a relatively stable expat voter base in contrast to the fluctuating voter base seen in Poland—and to a lesser extent—in the Czech Republic. It is worth noting, however, that despite the relatively very low voting costs faced by French expats, their engagement in French elections is remarkably low—peaking at 43.93% in presidential elections and at 24.77% in legislative elections across the considered time period.
Saliency also had a lower effect on overall voter turnout in France than was seen in Poland and the Czech Republic, suggesting our findings may reflect a more general difference in the turnout patterns of voters in old and new democracies (Mainwaring and Zoco, 2007). Overall voter turnout in French elections has been steadily declining on both a presidential and legislative level, while domestic voters in Poland and the Czech Republic are still seen to respond to varying saliency levels—albeit to a lesser extent than their expat voter base. It is possible then that the effect of electoral saliency on expat voter turnout corresponds to dynamics existing within the country in which the election is taking place. As will be outlined in detail in the following section, such dynamics will be impacted by a number of factors which have not been controlled for in this study—this includes the electoral systems under which elections are taking place and general trends in the political participation of citizens across the three countries. The inclusion of such factors in future studies will help us explain the differences in the mobilisation patterns of French expats in contrast to those of expat voters in Poland and the Czech Republic.
Limitations
This study faces some potential limitations as a result of fluctuating expat populations, the lack of a standard saliency measure and differences in electoral characteristics across the three countries considered.
Expat populations are not fixed and as there are no exact figures on the number of eligible expat voters in each election, it is impossible to entirely isolate the effect of electoral saliency on expat voter turnout from these population changes. In Poland and the Czech Republic—which do not have automatic voter registration in place for expats—this means that some of the turnout fluctuations identified in this study will reflect changes in their expat populations. This study accounted for this to some extent by referring to the general emigration patterns seen in the two countries. In Poland, some correlation between increased expat voter mobilisation and waves of emigration was identified, however, this mobilisation was still largely disproportionate to increases in the expat population. The study also identified significant levels of de-mobilisation in less salient elections which took place during times of high emigration out of the country. Meanwhile, in the Czech Republic, elections with high expat voter mobilisation did not in any way correlate with periods of high emigration out of the country. Therefore, while the lack of exact expat population figures influences the accuracy of the turnout fluctuations identified by this study, the available migration data suggests that our general findings are not discredited by this limitation.
As no standard measure of saliency exists in the literature, the study’s categorisation of salient and non-salient elections may be flawed. For example, despite its categorisation as an election of low saliency, as per our definition, the 2015 parliamentary election in Poland saw increases in both overall and expat voter turnout. It is possible that voter’s perception of electoral saliency differs between countries and is not fixed (Bélanger and Meguid, 2008). A greater understanding into how voters in different countries perceive electoral saliency is needed to determine a more accurate relationship between saliency and voter turnout. Conversely, some inconsistencies in the relationship between voter turnout and electoral saliency may be inevitable no matter what saliency measure is adopted. While there is significant evidence to show that higher electoral saliency corresponds to higher levels of voter turnout, this correlation is not perfect and may not present itself in every election. That said, in most of the elections considered by this study, overall and expat voter turnout moved in the same direction. That is, when overall voter turnout diminished, expat voter turnout also fell—generally by a more significant amount—and vice versa. This suggests that, despite any existing flaws in the saliency measure, expat voters have been shown to be more responsive to the same changing electoral dynamics which affected domestic voters.
Finally, the study’s findings may have been impacted by the differences in electoral characteristics across the three countries considered. As discussed in the methodology section, this includes significant variations in their expat voting procedures. It also includes the electoral system deployed in the elections of each of the three countries. A two-round plurality system is used in both parliamentary and presidential elections in France, while elections in Poland and the Czech Republic take place under a system of proportional representation. Such differences in electoral systems have been shown to have a significant impact on voting behaviours in a variety of studies (Blais, 2006; Cox, 2015). Voter dynamics also naturally differ across countries due to a variety of historical, institutional, and socioeconomic factors. This includes the distribution of power in the political system—which makes it more likely for voters to be mobilised in different types of elections, and country size—influencing the extent to which parties can mobilise their vote bases (Stockemer, 2016). As all of these factors impact on the voting behaviours of expat voters, the cross-country comparisons made in this study are limited. These problems do not influence, however, the study’s analysis of expat behaviours on a country-by-country basis, with no country experiencing any major electoral changes in the time period considered.
Conclusion
This study aimed to identify differences in the mobilisation patterns of expat and domestic voter groups in response to changing electoral saliency. Such differences were identified in Poland and the Czech Republic. In both countries, expat voters were mobilised to a greater extent than domestic voters in elections of high saliency. In Poland, expat voters also witnessed greater levels of de-mobilisation in elections of low saliency. Expat voter turnout saw significantly greater fluctuations across election cycles than did domestic voter turnout as a result. The same pattern was not identified in France, with expat and domestic voter groups responding similarly to the changing saliency of French elections.
The study’s findings demonstrate the potential for a wide variety of future research. Firstly, further studies can seek to explain the disparities in our findings across the three countries. Future studies investigating expat voter mobilisation should control for factors such as voting costs, turnout trends between old and new democracies and differences in the electoral system, as well as including a greater sample of countries in their analysis. Secondly, the study adopted a unique saliency measure which proved effective in evaluating the saliency of a wide range of parliamentary and presidential elections. Further research could test the viability of this measure across a greater sample of elections, as well as investigating whether voters’ perception of electoral saliency varies between countries. Finally, by identifying substantial differences in the effect of electoral saliency on expat and domestic voter turnout in Poland and the Czech Republic, the study showed the potential benefits of considering expat voters in future voter turnout research. Future studies should consider other institutional or socioeconomic factors which have been linked with higher levels of voter turnout in the literature and evaluate their effect on expat voters in isolation. The inclusion of expat voters in future research can thereby significantly contribute to our understanding of voting behaviours.
Bibliography
Aldrich, J. H. (1993). ‘Rational Choice and Turnout’. American Journal of Political Science, 37(1), pp. 246–278. Available at: https://doi.org/10.2307/2111531.
Arrighi, J. (2018). ‘Political participation of mobile EU citizens: France’. RSCAS/GLOBALCIT-PP 2018/6.
Bågenholm, A. (2013). ‘Throwing the rascals out? The electoral effects of corruption allegations and corruption scandals in Europe 1981–2011’. Crime, Law and Social Change, 60(5), pp. 595-609.
Baubock, R. (2006). ‘Stakeholder citizenship and transnational political participation: a normative evaluation of external voting’. Fordham L. Rev., 75, p. 2393.
Bélanger, É. and Meguid, B. M. (2008). ‘Issue salience, issue ownership, and issue-based vote choice’. Electoral studies, 27(3), pp. 477-491. Blais, A. (2006). ‘What affects voter turnout?’ Annual Review of Political Science, 9(1), pp. 111–125. Available at: https://doi.org/10.1146/annurev.polisci.9.070204.105121.
Blais, A. and St‐Vincent, S. L. (2011). ‘Personality traits, political attitudes and the propensity to vote’. European Journal of Political Research, 50(3), pp. 395-417.
Blankart, C. B. and Margraf, S. (2011). ‘Taxing expats—instrumental versus expressive voting compared’. Swiss Journal of Economics and Statistics, 147(4), pp. 461-478.
Chopin, T. and Faure, S. B. (2021). ‘Presidential election 2022: A euroclash between a “liberal” and a “neo-nationalist” France is coming’. Intereconomics, 56, pp. 75-81.
Cox, G. W. (1997). Making votes count: Strategic coordination in the world's electoral systems. Cambridge: Cambridge University Press.
Cox, G. W. (2015). ‘Electoral rules, mobilization, and turnout’. Annual Review of Political Science, 18, pp. 49-68.
Cześnik, M. (2014). ‘In the shadow of the Smolensk catastrophe—The 2010 presidential election in Poland’. East European Politics and Societies, 28(3), pp. 518-539.
Cześnik, M., Miśta, R. and Żerkowska-Balas, M. (2020). ‘Uczestnictwo i mobilizacja w wyborach parlamentarnych 2019 roku’. Studia socjologiczne, (4 (239)), pp. 91-121.
Downs, A. (1957). ‘An economic theory of political action in a democracy,’ Journal of Political Economy, 65(2), pp. 135–150. Available at: https://doi.org/10.1086/257897.
Dupoirier, E. and Sauger, N. (2010). ‘Four rounds in a row: The impact of presidential election outcomes on legislative elections in France’. French Politics, 8, pp. 21-41.
Evans, J., Ivaldi, G. (2018). ‘The 2017 presidential election: Continuity and change’. In The 2017 French Presidential Elections: A Political Reformation?, pp. 17-14.
Franklin, M. N. (1996). ‘Electoral participation’. In Comparing democracies: Elections and voting in global perspective, pp. 216-235.
Feddersen, T. J. (2004). ‘Rational choice theory and the paradox of not voting’. Journal of Economic perspectives, 18(1), pp. 99-112.
Ferejohn, J. A. and Fiorina, M. P. (1974). ‘The Paradox of Not Voting: A Decision Theoretic Analysis’. American Political Science Review. 68(2), pp. 525–536. Available at: https://doi.org/10.2307/1959502.
Geys, B. (2006). ‘Explaining voter turnout: A review of aggregate-level research’. Electoral Studies, 25(4), pp. 637–663. Available at: https://doi.org/10.1016/j.electstud.2005.09.002.
Goldberg, A.C. and Lanz, S. (2021). ‘Living abroad, voting as if at home?’. Electoral motivations of expatriates. Migration Studies, 9(2), pp. 279-310.
Havlík, V. (2014). ‘The 2013 parliamentary election in the Czech Republic’. Evropská volební studia, 9, pp. 43-49.
Kanazawa, S. (1998). ‘A possible solution to the paradox of voter turnout’. The Journal of Politics, 60(4), pp. 974-995.
Korzec, P. and Pudzianowska, D. (2021). ‘Report on political participation of mobile EU citizens: Poland’. RSC/GLOBALCIT-PP 2021/07.
Kriesi, H. (2008). ‘Political mobilisation, political participation and the power of the vote’. West European Politics, 31(1-2), pp. 147-168.
Lawson, C. (2003). ‘Voting preference and political socialization among Mexican Americans and Mexicans living in the United States’. Mexican Studies, 19(1), pp. 65-79.
Linek, L. (2018). ‘Report on political participation of mobile EU citizens: Czech Republic’. RSCAS/GLOBALCIT-PP 2018/16.
Mainwaring, S. and Zoco, E. (2007). ‘Political sequences and the stabilization of interparty competition: electoral volatility in old and new democracies’. Party politics, 13(2), pp. 155-178.
Markowski, R. and Cześnik, M. (2011). ‘Wybory parlamentarne w 2007 roku–ciągły zamęt czy utrwalenie nowego rozłamu?’Studia Polityczne, 27, pp. 109-127.
Markowski, R. (2020). ‘Plurality support for democratic decay: the 2019 Polish parliamentary election’. West European Politics, 43(7), pp. 1513-1525.
Orford, S., Rallings, C., Thrasher, M. and Borisyuk, G. (2009). ‘Electoral salience and the costs of voting at national, sub‐national and supra‐national elections in the UK: a case study of Brent, UK’. Transactions of the Institute of British Geographers, 34(2), pp. 195-214.
Østergaard-Nielsen, E. and Ciornei, I. (2019). ‘Political parties and the transnational mobilisation of the emigrant vote’. West European Politics, 42(3), pp. 618-644.
Plecitá-Vlachová, K. and Stegmaier, M. (2008). ‘The parliamentary election in the Czech Republic, June 2006’. Electoral Studies, 1(27), pp. 179-184.
Rovny, J. (2018). ‘Three lessons from the Czech presidential election’. LSE European Politics and Policy (EUROPP) Blog.
Rovny, J. (2021). ‘Populism punished: the 2021 Czech parliamentary election’. LSE European Politics and Policy (EUROPP) blog.
Stockemer, D. (2016). ‘What affects voter turnout? A review article/meta-analysis of Aggregate Research’. Government and Opposition, 52(4), pp. 698–722. Available at: https://doi.org/10.1017/gov.2016.30.
Turcu, A. and Urbatsch, R. (2020). ‘European ruling parties’ electoral strategies and overseas enfranchisement policies’. European Journal of Political Research, 59(2), pp. 269-289.
Vashchanka, V. (2020). ‘Political manoeuvres and legal conundrums amid the COVID-19 pandemic: the 2020 presidential election in Poland’. International Institute for Democracy and Electoral Assistance.
World Bank. (2022). ‘Net migration – Czechia’. World Population Prospects: 2022 Revision. Available at: https://data.worldbank.org/indicator/SM.POP.NETM?end=2017&locations=CZ&start=1999.
World Bank. (2022). ‘Net migration – Poland’. World Population Prospects: 2022 Revision. Available at: https://data.worldbank.org/indicator/SM.POP.NETM?end=2017&locations=PL&start=1999.
