STUDY UNEQUAL DEMOCRACIES WHO DOES (NOT) VOTE IN LITHUANIA? Only about a half of Lithuanians vote in national elections. Lithuania has one of the lowest voter turnout rates in Europe and the lowest one in the Baltic states. This report aims to identify who the Lithuanian (non-)voters are and suggest potential solutions to improve the current situation. In line with general European trends, the Lithuanians who are younger, less educated, and come from a lower socioeconomic class are less likely to vote. Non-voters in Lithuania are usually uninterested in politics, but voter turnout trends are not connected to voters’ satisfaction with democracy. Marius Danilevi č ius and Liutauras Gudžinskas September 2024 Systemic non-voting requires the attention of policy makers. We propose several reform tracks for tackling low voter turnout in Lithuania: introducing online voting, opening-up of political parties to the public, lowering the voting age to 16, and considering measures of deliberative democracy.

UNEQUAL DEMOCRACIES WHO DOES (NOT) VOTE IN LITHUANIA?

Content INTRODUCTION 2 LITHUANIAN TURNOUT IN COMPARISON 3 HOW IS VOTER TURNOUT SHAPED BY ­SOCIO-DEMOGRAPHIC CHARACTERISTICS?  5 WHO ARE THE LITHUANIAN 8 HOW MUCH IS VOTER TURNOUT TERRITORIALLY DISTRIBUTED ACROSS THE 13 CONCLUSIONS AND 15 References  17

UNEQUAL DEMOCRACIES – WHO DOES(NOT) VOTE IN LITHUANIA? INTRODUCTION One central element of democracy is the promise to give everyone an equal say in how our societies are run. In representative democracies, elections are the primary method of understanding public opinion. We elect officials who reflect the diverse interests of all society members. These representatives are tasked with forming majorities on various issues and resolving them through deliberation and compromise. They also support and scrutinize governments to ensure the implementation of the enacted legislation. Consequently, governments derive their legitimacy from winning the support of the majority of their citizens. However, if an increasing number of people abstain from exercising their right to vote, what becomes of our democracies and the functionality of this representative system? Set developed for this series. Second, a data set consisting of results from three consecutive Lithuanian post-election surveys(2012, 2016, 2020) is used to gain deeper insights into the demographics and characteristics of non-voters in Lithuania. Third, territorial analysis is conducted to look for places where the voter turnout is lowest. Finally, the report offers general recommendations on how to encourage non-voters to participate in elections. As more and more potential voters decide not to cast their vote at the ballot boxes, effective communication between incumbent government and the electorate breaks down. In addition, trends of voter participation influence the political parties and candidates as well – continuous abstentions from elections and limited resources could deter attempts to mobilize these same voters. Finally, as abstention from voting in elections becomes systematic, political parties might start increasingly adjusting their political programmes and priorities to the interests of only those social groups who are the most vocal about their concerns in the elections. This, in turn, may lead to some members of society having an outsized influence on future policy decisions and others with diminished voice on the matter. In this report, part of the FES Unequal Democracies series, we aim to shed light on these issues by focusing on Lithuania. The Baltic country presents an intriguing case due to its notably high voter abstention rates among European democracies. More than half of the Lithuanian electorate does not participate in voting; in the most recent parliamentary election in 2020, the turnout was just 47.8%. This figure is particularly striking when compared to other European countries, as well as with the turnout in the 1990s when Lithuania regained its independence. This report aims to provide evidence on the trends in voter turnout in Lithuania, identify who the non-voters are, and suggest potential solutions to improve the situation. First, it presents comparative data on voter turnout, with a focus on Lithuania, using the Unequal Democracies Comparative Data 2

Lithuanian turnout in comparison LITHUANIAN TURNOUT IN COMPARISON In Lithuania, similarly to other European democracies, elections have been marked by a downward trend in voter turnout since 1990s when the country regained independence. (Figure 1). The most recent parliamentary election in 2020 drew 47.8% of the electorate to the ballot boxes. Elections in 2008 and 2012 had seen a minimal increase of voters, but the tendency of falling turnout rates is clear. Statistics of the voter participation in legislative elections rank Lithuania among countries with the lowest electoral turnout in Europe and even the democratic post-communist space(during recent elections only Romania(2020: 31.94%), Bulgaria (2024: 33.4%), Albania(2021: 46.29%), and Switzerland (2023: 46.6%) saw lower voter turnout rates). Voter turnout trends in the three Baltic states provide a useful comparison for grasping the current situation(Figure 2). In Estonia, Latvia, and Lithuania, the Supreme Council and founding elections in the early 1990s witnessed an extraordinary surge in turnout as the public enthusiastically embraced their first opportunity to partake in free and fair elections. Nonetheless, this initial enthusiasm waned in the mid-to-late 1990s, marked by a decline in voter participation as the allure of democratic elections diminished. Notably, Lithuania experienced the most significant drop in turnout in the Baltics. As it can be seen from the data, the steepest decrease of the voter turnout in Lithuania is observed during the first decFigure 1 Voter turnout in parliamentary elections in Lithuania 80 75.22 70 71.72 60 58.18 50 52.92 40 46.04 48.59 52.93 50.64 30 20 10 0 1990 1995 2000 2005 2010 2015 47.81 2020 3

UNEQUAL DEMOCRACIES – WHO DOES(NOT) VOTE IN LITHUANIA? Figure 2 Voter turnout trends in parliamentary elections in Estonia, Latvia and Lithuania 100 90 80 70 60 50 40 30 20 10 0 1990 1995 2000 Estonia 2005 Latvia 2010 2015 Lithuania 2020 Out of the three Baltic states Lithuania has seen the sharpest decline in electoral participation since regaining independence. ade of the post-communist transformation. At the time, the radical measures of so-called‘shock therapy’ were applied triggering severe economic hardships for a major part of society. The period was also marked by growing public disillusionment with the political establishment, and a series of corruption scandals. It is, therefore, reasonable to expect that the residents from lower social classes and other vulnerable groups should have been much more affected by this sweeping political alienation. 4

How is voter turnout shaped by ­socio-demographic characteristics? HOW IS VOTER TURNOUT SHAPED BY ­SOCIO-DEMOGRAPHIC CHARACTERISTICS? Looking into who takes part in Lithuanian elections, general patterns can be detected. In the case of the most recent parliamentary elections, several socio-demographic effects are noticeable. As graphs of comparative electoral behaviour data illustrate the differences in age(Figure 3), education(Figure 5), and class(Figure 6) have a strong influence on participation in Lithuanian elections. Firstly, increasing age appears to have a positive relationship with voter turnout as younger generations are continuously more likely to abstain from elections. Additionally, people with lower and medium levels of education also have a higher tendency of non-voting when compared to highly educated group(although, in the case of difference between voters who have obtained low and medium education this tendency seems to be reversed). Last, voter turnout rates of higher socio-economic classes(upper service class, lower grade service class) regularly surpass voting rates in lower socio-economic classes. The differences of voter participation of males and females seems to be an insignificant statistic variation as no clear tendency line can be observed (Figure 4). With an aim to delve even deeper into the election turnout trends in Lithuania, this report utilises a statistical binary logistic regression model of non-voting. For the analysis, a new data set was created after harmonizing and combining the data from three(2012, 2016, 2020) nationally representative Lithuanian post-election surveys accessed from Lithuanian Data Archive(LiDA). The resulting data file includes as many as 4,664 observations. Figure 3 Voter turnout rates in Lithuania by voter age groups 80 70 60 50 40 30 20 10 0 2008 2009 2010 2011 2012 Under 30 2013 2014 30–44 2015 2016 45–60 2017 2018 Over 60 Older generations are more likely to take part in elections than younger voters. Source: UD Comparative Data Set. 2019 2020 5

UNEQUAL DEMOCRACIES – WHO DOES(NOT) VOTE IN LITHUANIA? Figure 4 Voter turnout rates in Lithuania by gender 80 70 60 50 40 2008 2009 2010 2011 2012 2013 Male 2014 2015 Female 2016 2017 No significant gender inequality in voting can be observed. Source: UD Comparative Data Set. 2018 2019 2020 Figure 5 Voter turnout rates in Lithuania by voters’ education attainment level 80 70 60 50 40 2008 2009 2010 2011 2012 Level of education: 2013 2014 Low 2015 2016 Medium 2017 High 2018 2019 2020 Voters with high level of education are more likely to vote than those with medium and lower levels of education. Source: UD Comparative Data Set. Figure 6 Voter turnout rates in Lithuania by voters’ socio-economic class 90 80 70 60 50 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 Routine jobs Semi-skilled jobs Small business owners Lower grade service class Upper service class Voters from higher socio-economic classes(upper service class, lower grade service class) tend to participate in elections more often than members of lower socio-economic classes. Source: UD Comparative Data Set. 6

How is voter turnout shaped by ­socio-demographic characteristics? Figure 7 Results of statistical model built with the data from 2012, 2016, and 2020 Lithuanian post-election surveys. (Ref. aged 18–24) Aged 25–34 Aged 35–44 Aged 45–55 Aged 55–64 Aged 65+ (Ref. Male) Female (Ref. Rural) Urban (Ref. low level) Education – middle level Education – high level (Ref. Employed) Unemployed Student Retired Homemaker Other (Ref. low income) Income – middle Income – high (Ref. Lithuanian) Russian nationality Polish nationality Other nationality (Ref. Very interested in politics) Quite interested in politics Quite uninterested in politics Very uninterested in politics (Ref. Very satisfied with democracy) Fairly satisfied with democracy Not very satisfied with democracy Not at all satisfied with democracy significant not significant Estimate –2 –1 0 1 2 3 Lines in the graph illustrate the probability of non-voting of different demographics compared to reference groups(“Ref.”). Lines crossing the“0” line on the x-axis(variables on grey background) indicate that no significant differences between that variable group and the reference group could be detected.  If the line is fully on the left(negative side) of the“0” line, that indicates these groups are less likely to not vote than their reference group. On the other hand, if the line is fully on the right(positive side) of the“0” line, this indicates these groups are more likely to not vote than their reference group. Model 1, N=4664, McFadden= 0.5801099 For the purposes of robust predictive capabilities both gender and level of education(three groups – lower, middle, high) were once again included in the model. The effect of age was considered by including narrower age groups(18– 24, 25–34, …, 65+) of respondents as another independent variable. Trying to determine the effect socio-economic standing has for the decision to abstain from elections, indicators of living area(rural – settlements with<2,000 inhabitants, urban – settlements with population of>2,000), status of employment(employed, unemployed, student, retired, homemaker, other), and family income level(low, middle, high) were included in this model. In line with the findings of various studies of divergent political behaviour of ethnic minorities(Fieldhouse, 2008; Bhatti and Kasper, 2016; Vidzbelis, 2020) nationality(Lithuanian, Russian, Polish, other) was also included as a variable. The analysis additionally involves variables on voters’ relationship with the political regime(see satisfaction with democracy(see, for example, Ezrow and Xezonakis, 2016) and political interest(Prior, 2010; for critique – Denny and Doyle, 2008). Finally, the dependent variable of participation in that year parliamentary election remained binary, but for the purposes of easier understanding of non-voting was converted(0 – voted, 1 – did not vote). The resulting model is visualized in Figure 7. 7

UNEQUAL DEMOCRACIES – WHO DOES(NOT) VOTE IN LITHUANIA? WHO ARE THE LITHUANIAN NON-VOTERS? To begin with, the results of the logistic regression indicate that age has a significant effect on non-voting. Adding to the previously briefly discussed trends of turnout rates in various age groups in Lithuania, the data set for this model includes a wider variety of narrower age groups. The results allow to reiterate the noticeable trend of younger voters abstaining from election far more often than voters from older generations(aged 45–54, 55–64, and 65+). These findings seem in agreement with previous Lithuanian studies(for the effect of age on Lithuanian electoral behaviour see Žiliukait ė , 2014; Skirkevi či us, 2022). Figure 8 illustrates, in a simpler way, the probabilities of non-voting for various age groups. While those aged 18–24 have an estimated 31% chance of abstention from participating in elections, such probability decreases with age – model predicts age groups 45–54, 55–64, and 65+ have, accordingly, an 18%, 15%, and 12% chance of not voting – on average almost two times lower than the younger respondents. Additionally, the influence age has on voter turnout rates is shown in Figure 9. The graph with 95% confidence intervals was produced after modifying the original model by replacing age group variable with respondents’ age. Once again, a clear negative relationship between age and non-voting can be observed. Figure 8 Predictive probability of non-voting by voters’ age groups 40 Probability of non-voting in % 30 20 10 18–24 25–34 35–44 45–54 55–64 Older generations of voters(aged 45–65+) are far less likely to not vote in elections in contrast to the younger aged voters(aged 18–44). Confidence intervals are estimated at 95%. 64+ Age group 8

Who are the Lithuanian non-voters? Figure 9 Predictive probability of non-voting by voters’ age 50 Probability of non-voting in % 40 30 20 10 25 50 75 A clear trend line of the relationship between voters’ older age and decreasing likelihood of abstaining from voting in elections can be observed. Model 2, N=4664, McFadden= 0.5801099. 100 Age Noticeably, the gender of voters seems to have a statistically significant effect on voter turnout in Lithuania according to the binary logistic regression model. The model predicts that women, on average, have a 17 percent probability of non-voting, while men seem to be more likely to abstain from elections – model predicts a 23% chance. These results are contrary to some findings on gender electoral participation(see, for example, Kostelka, Blais, and Gidengil, 2019). As recent studies from various researchers indicate, the effect of gender on non-voting seems to be quite dynamic(Stauffer and Fraga, 2022) and dependent on the type of elections studied(Dassonneville and Kostelka, 2021) thus we do not attempt to explain these significant yet small differences in electoral participation based solely on this statistical model. While the scope and limits of electoral surveys should be acknowledged(see Stockemer and Sundstrom, 2023), studies of the gender gap in electoral participation in Lithuania might benefit from a deeper delve into the effects of increasing number of women candidates and political leaders, (changing) cultural attitudes towards women in the society and the socio-economic status of women. As for the size of respondents’ living area in Lithuania, the differences in chances of abstention from participating in elections are estimated to be quite large. The model predicts a 13 percent probability of non-voting if a person lives in a rural area, yet those living in urban settlements are ascribed a 22 percent probability of abstention in elections. These findings on living area influence seem in agreement with previous research(Geys, 2006, Garcia-Rodriguez& Redmond, 2020), yet they should be interpreted carefully. The perceived differences between electorate from urban and rural settlements might be the result of various outside factors – one of the explanations could be the over-reporting of voting(falsely claiming to have voted) in electoral studies. In addition, such disparities in non-voting might be specific to parliamentary elections in Lithuania and the incumbent local parties(see Vidzbelis and Tu č as, 2018) and not systematic. Finally, such predictions seem to be contrary to the actual results of the elections(see for example the mapped trends of electoral participation in the 2020 election). Continuing with the trend noticed in previous analysis(see Figure 5), model also included the measures of different education levels(low – no education or primary, medium – general(and/or vocational), high – BA, MA, PhD). Regression with the data of various post-election surveys allows to grasp, the widely researched and discussed(see Sondheimer and Green, 2010; Dassonneville and Hooghe, 2017) influence that education has on voting. It is worth mentioning, that the effect of different education levels varies – both high (p= 0.158177) and middle(p= 0.349863) levels of education do not seem to be significant predictors of non-voting when compared to low level of education. Such results might come 9

UNEQUAL DEMOCRACIES – WHO DOES(NOT) VOTE IN LITHUANIA? down to the shortcomings of our statistical model and its scope(wide confidence interval representing mostly the lack of more data in the lower level of education demographic). Yet, the differences between probabilities of non-voting between respondents with medium level of education(25%) and those who achieved a high level of education(13%) are clear-cut(see Figure 10). One could hypothesise that the stark differences between these two groups might be explained by the high percentage of population with a high level of education in Lithuania(46.5 percent of persons aged 25–64 in the year 2022) – as higher education becomes“the new norm” in Lithuanian society, the gap(in their socio-economic standing) between those who have obtained at least a bachelor’s degree and those who have not might become more pronounced. This, in turn, might lead to the differences in electoral behaviour of these groups. Moreover, employment status was also used in the model to account for the socio-economic standing of respondents and its influence on voter turnout. In comparison with employed respondents, unemployed voters are more likely to abstain from elections(probabilities of 18% and 24% accordingly), although one must notice that lower validity of such insights(p= 0.049175). Interestingly, regression model shows that homemakers and those describing their current employment as“other” seem to be significantly more likely to be non-voters when compared with those employed (homemaker probability of non-voting predicted at 33% and“other” at 45%). In addition, factor of family income groups was also included in the analysis. As Figure 11 illustrates, non-voting probabilities of representatives from both middle and high family income groups differ significantly when compared to low family income group. Whereas the model predicts a 26 percent chance of non-voting in lower family income groups, such probability decreases as wealth increases(middle income – 18 percent, high income – 17 percent). While obvious differences can be seen in comparison with lower income groups, no pronounced differences are visible when comparing middle- and high-income groups. It seems that people with disadvantaged socio-economic standing in Lithuania are far more likely to abstain from elections. Such predictions seem to reiterate the importance of social class in voter turnout patterns(see Ehs and Zandonella, 2021; Nadeau, Lewis-Beck, and Foucault, 2019). As Polacko (2023) finds, based on data from 30 countries and 111 elections between 1996 and 2019, inequality significantly reduces turnout. The analysis of the composition of the Lithuanian Parliament elected in 2020 – another investigation conducted under FES Unequal Democracies series – also underlines that non-participation of voters of lower income in the elections correlates with the lack of them within the legislature. Namely, in the current Lithuanian parliament, there are no MPs that could credibly claim their allegiance to the working class(Gudžinskas and Jonutis 2024). Figure 10 Predictive probability of non-voting by voters’ level of education 30 Probability of non-voting in % 25 20 15 low middle high Level of education The model does not detect differences in the likelihood of voting when comparing those with low and medium/high education. Yet those who have attained medium level of education are more likely to not vote in elections when compared to voters with high level of education. Confidence intervals are estimated at 95%. 10

Figure 11 Predictive probability of non-voting by voters’ family income groups 30 Who are the Lithuanian non-voters? Probability of non-voting in % 25 20 15 low middle high Income groups Members of electorate whose family income group is the lowest are more likely to not vote when compared to middle or high family income voters. Such differences between electorate from middle and high income families are not observed. Confidence intervals are estimated at 95%. Figure 12 Predictive probability of non-voting by voters’ level of satisfaction with democracy Probability of non-voting in % 25 20 15 10 very satisfied fairly satisfied not very satisfied not at all satisfied Satisfaction with democracy Results of the statistical model show that none of the positions concerning the respondents’ satisfaction with democracy have a significant effect on the probability of them not voting when compared to each other. Confidence intervals are estimated at 95%. 11

Probability of non-voting in % UNEQUAL DEMOCRACIES – WHO DOES(NOT) VOTE IN LITHUANIA? Figure 13 Predictive probability of non-voting by voters’ level of interest in politics 60 40 20 very interested quite interested quite uninterested very uninterested Interst in politics Compared to those with keen interest in politics voters who claim to be not interested in politics have a high probability of not voting in elections. Confidence intervals are estimated at 95%. The binary logistic regression model included ethnicity(Lithuanian, Polish, Russian, other) as an independent variable too. Expecting a lower turnout rate among ethnic minorities in Lithuania, the model could not support such hypothesis – no ethnic group has a significantly(the smallest p value being 0.095730) higher chance of abstaining from participating in election when compared to each other. Such finding is in line with the spatial analysis by Dovydas Vidzbelis (2020) and could be interpreted by referring to a largely ethnically homogenous population of the country. However, the Vidzbelis’s research also demonstrated that in particular municipalities, such as Šal č ininkai, or Visaginas, where the Polish and Russian minorities comprise the majority of the local population, the non-Lithuanians tend to be more active in the parliamentary elections than the residents belonging to the titular nation. Arguably, such trends reveal relatively strong capacities of the Lithuanian Polish Electoral Action (continuously represented in the parliament since its establishment in 1994) of mobilising their supporters. that actual voting has an impact on satisfaction with democracy, and not other way around(Kostelka and Blais, 2018). Analysis of voters’ interest in politics paints a different picture(see Figure 13). Results are in line with general findings of the influence political interest has on voting turnout (on being informed, see Lassen, 2005). The contrast is quite stark – while those considering themselves‘very interested’ in politics have a predicted 10 percent probability of non-voting,‘very uninterested’ respondents’ chances are as high as 61 percent. Interestingly, there is quite a large gap between ‘quite’ and‘very’ uninterested with a difference in probabilities estimated to around 30%. Lastly, the effects of respondents’ satisfaction with democracy and interest in politics were analysed. As Figure 12 illustrates, the level of satisfaction one has with Lithuanian democracy, has no significant effect on(non-)voting tendencies. Such results contradict the conclusions of the research finding the influence trust in parliament, and satisfaction with democracy have on political participation(Grönlund and Setälä 2007). However, more recent research demonstrated 12

How much is voter turnout territorially distributed across the country? HOW MUCH IS VOTER TURNOUT TERRITORIALLY DISTRIBUTED ACROSS THE COUNTRY? Analysis of non-voting trends would also benefit from a look at how voter turnout is distributed territorially. As numerous Lithuanian constituencies do not match the borders of municipalities, for the purposes of the analysis the official election result data from electoral districts was converted to represent trends in municipalities. Figure 14 illustrates the patterns of voter turnout in the most recent parliamentary elections in 2020 by municipality. At first glance, municipalities of the biggest cities of Lithuania(Vilnius, Kaunas, Šiauliai, Panev ė žys), with the exception of Klaip ė da, are among the leaders in voter participation rates. Yet, one must notice that not as urban and less populous municipalities in the Northeastern and Southeastern parts of Lithuania show high results of voter turnout as well. A noticeable exception might be the municipality of Visaginas, a soviet-era monotown with a predominantly ethnically Russian population. For the purposes of understanding voter turnout diffusion throughout Lithuanian geography and testing the previous non-voting model, another(smaller and having a bit less explained variability – McFadden=0.04731887) statistical model was created. A numeric dependent variable of voter turnout percentage and numeric independent variables of urbanization level(%), unemployment level(%), average income in the municipality, and percentage of younger age group(aged 18–24) representatives were chosen for the multivariate linear regression model. The results show that although urbanization level has a negative influence on voter turnout, its effect is not as pronounced(p= 0.06060). Two factors – unemployment level and percentage of younger population – have a significant impact on electoral participation in municipalities. As Figure 15 illustrates, level of unemployment has a negative effect on voting – as the model Figure 14 Municipality voting turnout in 2020 Lithuanian parliamentary elections  57.12 Darker shade of blue indicates higher voter participation rate in the municipalities. 13

UNEQUAL DEMOCRACIES – WHO DOES(NOT) VOTE IN LITHUANIA? (p = 0.02201) predicts, a 1% increase in unemployment results in a 0.42% decrease in voter turnout. The probabilities of voter turnout trends in relation to younger parts of population are mapped in Figure 16. The direction of this relationship is also downward, with model quite robustly (p = 0.00545) predicting that a 1% increase in the proportion of younger aged population in Lithuanian municipalities have an effect of turnout rates falling by 1.6%. Figure 15 Predicted voter turnout in Lithuanian municipalities by the level of unemployment in the administrative unit 55 Turnout in % 50 45 40 8 12 16 Unemployment in% A higher unemployment level in the Lithuanian municipalities is seen to correlate with a lower electoral participation rate. Confidence intervals are estimated at 95%. Model 3, N=60, McFadden= 0.04731887 Figure 16 Predicted voter turnout in Lithuanian municipalities by the percentage of people aged 18–24 in the administrative unit 55 Turnout in % 50 45 40 5 6 7 8 9 10 % A bigger proportion of younger residents in the Lithuanian municipalities is seen to correlate with a lower electoral participation rate. Confidence intervals are estimated at 95%. Model 3, N=60, McFadden= 0.04731887 14

Conclusions and recommendations CONCLUSIONS AND RECOMMENDATIONS Let us start from the final observation of our logistic regression model that political interest has significant influence on citizens’ inclination to vote or abstain from going to the elections. This finding corroborates the“motivation theory” proposed by André Blais and Jean-François Daoust(2020). They argue that voters foremost go to the elections due to two reasons: either they are genuinely curious about political affairs, or they deeply sense civic duty to voice their opinion who is to govern for the next term of office. In this case it is worth to note diverging trends of electoral participation between parliamentary and presidential elections in the 21 st century in Lithuania. While their average turnout since 2000 has been roughly the same(51% for the former, and 53% for the latter), more recently people have started to show up more frequently in the presidential race. In the 2019 presidential elections, 57.4% voters came, and in the last ones in 2024 60% of them turned out, whereas participation in the legislative elections kept fluctuating around 50%. One can wonder whether this divergence is because of the presidential elections being more appealing to ordinary citizens due to their inherent personalism, presumably more capable of framing the choice, or because of heightened sense of civic duty to elect the chief of armed forces amid the security crisis in the region. We also observe sudden rises of the vote turnout in the legislative elections in various countries in East-Central Europe. By instance, in Croatia in April 2024, 62.3% voters cast their ballot for the new parliament – almost 18 percentage points higher than four years ago. Similar surges of turnout have occurred in Poland, Slovakia, Slovenia, and elsewhere. Increasing polarisation, emergence of powerful anti-establishment force, fears of democratic backsliding arguably have contributed to the voters’ agitation in the region. Against this background, the relative tranquillity of the Lithuanian politics, still without a strong, right-wing populist, and Eurosceptic voice, might be appealing from the first glance. However, mainstream political parties would have been shortsighted, if they were to imply that such electoral dullness or boredom is forever. Therefore, more efforts to engage with disinterest voters are acutely needed. While an idea of compulsory voting most likely would be counterproductive(coercion is a false ally of motivation, after all), there are other, less intrusive ways to boost electoral participation. One of them could be the introduction of online voting successfully realised by Estonia – for the first time in the world in its parliamentary elections in 2007. Despite various concerns, the voters got to trust the system and increasingly use this electronic option to voice their preferences. In the most recent legislative elections in 2023, for the first time, most votes were cast online with the overall turnout being 63.5%. In particular, online voting could help reach younger citizens and urban dwellers who, according to our analysis, are among the least active voters in Lithuania. Moreover, that would also provide entirely new possibilities for an ever-increasing diaspora of the citizens living abroad to re-engage in the country’s politics. Secondly, political parties(foremost those who are in the parliament) need to open up to the society more broadly. They have to expand their members’ and followers’ networks, to develop a more including decision-making system, as well as to nourish constructive ties with the civil society. By investing in their territorial infrastructure and strengthening their organizational capacities(in particular, in major cities, and their youth movements), they would not only boost their electoral chances but also would contribute to building a more engaging and society-oriented party system. In addition, political parties could go directly after non-voters in particular focusing on areas that are notorious in this regard. Although it may cost quite a lot of energy and resources, it may pay off in the long run. Thirdly, it would be helpful to nudge young voters to become more active in politics from the outset by lowering the threshold of minimal age when they are eligible to vote – from 18 to 16 years. One of the major obstacles for young voters to draw attention to the electoral politics is that their life is at transition at that time. However, at 16, their life is likely still less challenging than in a few years’ time, thus there may be indeed better conditions for their initiation to the democratic rituals. Evidence from various European and Southern American democracies, which recently lowered the vote age to 16, tends to support positive impact of the reform on increasing the voter turnout, in particular among the youngest cohorts of voters. The earlier a person starts going to cast their ballot, the more likely they will form the habit in the future. However, one also needs to boost civic education among younger people to make such change durable(Eichhorn and Bergh, 2020). 15

UNEQUAL DEMOCRACIES – WHO DOES(NOT) VOTE IN LITHUANIA? Fourthly, increasingly popular idea of organizing citizens’ assemblies or taking other similar measures of deliberative democracy should also taken into account. Evaluations have found that participants of citizens’ assemblies afterwards also appreciate representative democracy more than before (OECD, 2020). A survey conducted with representative samples of 15 Western European countries also found that the most supportive of citizens’ assemblies are those who are less educated and have a low sense of political competence and an anti-elite sentiment. Such support, however, is conditioned on the expectation of a favourable outcome(Pilet, Bol, Vittori, and Paulis, 2023). The latter insight leads to our final observation that to address the grievances of vulnerable social groups leaning to abstain from voting, political parties have to become more responsive to their needs, in particular, in reducing unemployment rate and providing a better income protection amid the rise of living costs. In particular, left-wing actors should draw lessons from the observed patterns of decreasing voter turnout and growing inequality at the same time. As the author of one study cited above suggests, social democratic parties should mitigate the negative effects of inequality on turnout for low-income individuals by adopting more redistributive welfare state positions(Polacko 2023, p. 553). In general, the Lithuanian citizens are rather critical in how they evaluate the performance of the government and its role to solve most pressing problems. Despite robust economic growth in the country since its entry to the EU in 2004, the Lithuanian voters are eager to punish every incumbent as failing to improve the life in the country satisfactorily. The politicians, thus, need to fight social exclusion more vigorously to reverse this trend and also to send a clear signal to the voters of lower social strata that their voice in the elections matters like of everyone else. 16

References REFERENCES Bhatti, Yosef, and Kasper M. Hansen(2016).“The effect of residential concentration on voter turnout among ethnic minorities.” International Migration Review, 50.4: 977–1004. Polacko, Matthew(2023).“Turning off the base: Social democracy’s neoliberal turn, income inequality, and turnout”. Politics& Policy 51: 538–568. Blais, André, and Jean-François Daoust(2020). The Motivation to Vote: Explaining electoral participation. UBC Press. Prior, Markus(2010).“You’ve either got it or you don’t? The stability of political interest over the life cycle.” The Journal of Politics, 72.3: 747–766. Dassonneville, Ruth, and Filip Kostelka(2021).“The cultural sources of the gender gap in voter turnout.“ British Journal of Political Science, 51.3: 1040–106. Dassonneville, Ruth, and Marc Hooghe(2017).“Voter turnout decline and stratification: Quasi-experimental and comparative evidence of a growing educational gap.” Politics, 37.2: 184–200. Denny, Kevin, and Orla Doyle(2008).“Political interest, cognitive ability and personality: Determinants of voter turnout in Britain.” British journal of political science, 38.2: 291–310. Ramonait ė, Ainė; Jastramskis, Mažvydas; Kavaliauskaitė, Jūratė; Morkevičius, Vaidas; Petronytė , Ieva(2021).“Post-Election Survey, November–December 2016”, https://hdl.handle.net/21.12137/CGJGCI, Lithuanian Data Archive for SSH(LiDA), V3, UNF:6:PYwLXCqTLarCilezeOB9jg==. Ramonait ė, Ainė; Jastramskis, Mažvydas; Kavaliauskaitė, Jūratė; Morkevičius, Vaidas; Pukelis, Lukas; Ulinskaitė, Jogilė; Vaiginytė, Adelė ; Vijeikis, Paulius(2022).“Post-Election Survey, January–February 2021”, https://hdl.handle.net/21.12137/KVY7LE, Lithuanian Data Archive for SSH(LiDA), V3, UNF:6:CUnvK66WAYwcGV/Ur74ivA==. Ehs, Tamara, and Martina Zandonella(2021).“Different Class Citizens: Understanding the Relationship between Socio-economic Inequality and Voting Abstention.” Politics in Central Europe, 17.3: 525–540. Eichhorn, Jan, and Johannes Bergh(2020). Lowering the Voting Age to 16: Learning from Real Experiences Worldwide. Palgrave Macmillan Cham. Ezrow, Lawrence, and Georgios Xezonakis(2016).“Satisfaction with democracy and voter turnout: A temporal perspective.” Party Politics, 22.1: 3–14. Fieldhouse, Edward, and David Cutts(2008).“Diversity, density and turnout: The effect of neighbourhood ethno-religious composition on voter turnout in Britain.” Political Geography, 27.5: 530–548. Garcia-Rodriguez, Abian, and Paul Redmond(2020).“Rainfall, population density and voter turnout.” Electoral Studies, 64: 102128. Geys, Benny(2006).“Explaining voter turnout: A review of aggregate-level research.” Electoral studies, 25.4: 637–663. Grönlund, Kimmo, and Maija Setälä(2007).“Political trust, satisfaction and voter turnout.” Comparative European Politics, 5: 400–422. Gudžinskas, Liutauras, and Karolis Jonutis(2024).“Unequal Democracies: Who does(not) have a seat in the Lithuanian parliament?” Vienna: FES Regional Office for International Cooperation. Kostelka, Filip, and André Blais(2018).“The chicken and egg question: Satisfaction with democracy and voter turnout.” PS: Political Science& Politics, 51.2: 370–376. Ramonait ė, Ainė; Žiliukaitė, Rūta; Kavaliauskaitė, Jūratė; Jastramskis, Mažvydas; Nevinskaitė , Laima(2021).“Post-Election Survey, November–December 2012”, https://hdl.handle.net/21.12137/ C6YH6D, Lithuanian Data Archive for SSH(LiDA), V3, UNF:6:W1zaXwEKqEZPDvk2kEJU+A==. Skirkevi č ius, Paulius(2022).“Po pulistinės nuostatos ir balsavimas: ar populistines nuostatas turintys rinkėjai balsuoja už naująsias partijas?(2016 metų Seimo rinkimų atvejis).” Politologija 3(107): 51–89. Sondheimer, Rachel Milstein, and Donald P. Green(2010).“Using experiments to estimate the effects of education on voter turnout.” American Journal of Political Science, 54.1: 174–189. Stauffer, Katelyn E., and Bernard L. Fraga(2022).“Contextualizing the gender gap in voter turnout.” Politics, Groups, and Identities, 10.2: 334–341. Stockemer, Daniel, and Aksel Sundstrom(2023).“The gender gap in voter turnout: An artefact of men’s over-reporting in survey research?.” The British Journal of Politics and International Relations, 25.1: 21–41. Vidzbelis, Dovydas(2020).“Lietuvos tautini ų mažum ų rinkiminio aktyvumo teritorin ė raiška.” Geografijos metraštis: 45–62. Vidzbelis, Dovydas, and Rolandas Tu č as(2021).“Lietuvos gyventoj ų rinkiminio aktyvumo teritorin ė diferenciacija 2014–2016 metais.” Geologija. Geografija, 4.3. Žiliukait ė, Rū ta(2014).“Balsavimas rinkimuose ir rink ė j ų socialini ų tinkl ų charakteristikos.” Politologija, 1(73): 98–128. Kostelka, Filip, André Blais, and Elisabeth Gidengil(2019).“Has the gender gap in voter turnout really disappeared?.” West European Politics, 42.3: 437–463. Lassen, David Dreyer(2005).“The effect of information on voter turnout: Evidence from a natural experiment.” American Journal of political science, 49.1: 103–118. Nadeau, Richard, Michael S. Lewis-Beck, and Martial Foucault (2019).“Wealth and voter turnout: investigating twenty-eight democracies.” Polity, 51.2: 261–287. OECD(2020). Innovative Citizen Participation and New Democratic Institutions: Catching the Deliberative Wave, OECD Publishing, Paris. Official Statistics Portal(2023). https://osp.stat.gov.lt/en/lietuvos-svietimas-ir-kultura-2023/svietimas/bendrieji-svietimo-rodikliai. Pilet, Jean-Benoit, Damien Bol, Davide Vittori, and Emilien Paulis (2023).“Public support for deliberative citizens’ assemblies selected through sortition: Evidence from 15 countries”. European Journal of Political Research, 62: 873–902. https://doi.org/10.1111/1475-6765.12541. 17

Imprint ABOUT THE AUTHORS IMPRINT Marius Danilevi č ius is a Political Science bachelor’s student at the Institute of Political Science and International Relations of Vilnius University. His research interests involve political attitudes, voting behaviour, diaspora politics and studies of post-communist societies. Publisher: FES Regional Office for International Cooperation Democracy of the Future Reichsratsstr. 13/5 A-1010 Vienna Liutauras Gudžinskas is an associate professor of comparative politics at the Institute of Political Science and International Relations of Vilnius University, and director of the Institute for Solidarity, analytical centre of the Lithuanian Social Democratic Party. His main research interests are quality of democracy and governance in East-Central Europe, left-wing parties in the region. Responsibility for content: Johanna Lutz| Director, Democracy of the Future Phone:+43 1 890 3811 301 X:@FES_Democracy democracy.fes.de Contact / Orders: democracy.vienna@fes.de Design: pertext, Berlin| www.pertext.de The views expressed in this publication are not necessarily those of the Friedrich-Ebert-Stiftung(FES) or of the organization for which the author works. Commercial use of media published by the FES is not permitted without the written consent of the FES. Publications by the FES may not be used for electioneering purposes. ISBN 978-3-98628-578-4 © 2024 ABOUT UNEQUAL DEMOCRACIES Unequal Democracies is a project by FES Democracy of the Future. The main goal is to promote comparative understanding of why inequality in voting, political representation and other democratic processes hurt our democracies. In the series Who does(not) have a seat in Parliament? we analyse the social representation of European parliaments. In the series Who does(not) vote? we investigate election turnout levels across the parameters gender, age, social class and education in European democracies. Both series contain comparative studies and selective country reports. The comparative studies lay out general trends while the country reports provide country-specific analyses about the state of particular national contexts with the aim to develop and discuss political recommendations for decision-makers. More information at: https://democracy.fes.de/topics/inequality-democracy 19

WHO DOES(NOT) VOTE IN LITHUANIA? Why should Lithuanians Who are the Lithuanian What should be done? care about turnout? non-voters? Only about a half of Lithuanians vote in national elections. Lithuania has one of the lowest voter turnout rates in Europe and the lowest one in the Baltic states. This report aims to identify who the Lithuanian(non-)voters are and suggest potential solutions to improve the current situation. In line with general European trends, the Lithuanians who are younger, less educated, and come from a lower socioeconomic class are less likely to vote. Non-voters in Lithuania are usually uninterested in politics, but voter turnout trends are not connected to voters’ satisfaction with democracy. Systemic non-voting requires the attention of policy makers. We propose several reform tracks for tackling low voter turnout in Lithuania: introducing online voting, opening-up of political parties to the public, lowering the voting age to 16, and considering measures of deliberative democracy.