STUDY Līga Stafecka May 2026 Autor:innen Monat Jahr Who Does(Not) Titel der VPoutbeliiknaLtiaotnvia? Unequal Democracies Untertitel der Publikation ein- oder mehrzeilig , Illustration, Titelfoto arbfläche Icon oder F FES Regional Office for International Cooperation

Imprint Published by Friedrich-Ebert-Stiftung e.V. Godesberger Allee 149 53175 Bonn, Germany info@fes.de Issuing Department FES Regional Office for International Cooperation Democracy of the Future Reichsratsstr. 13/5 A-1010 Vienna Responsibility for content and editing Michael Jennewein michael.jennewein@fes.de Margarete Lengger margarete.lengger@fes.de Contact democracy.vienna@fes.de Design/Layout pertext| corporate publishing 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. May 2026 © Friedrich-Ebert-Stiftung e.V. Further publications of the Friedrich-Ebert-Stiftung can be found here: ↗ www.fes.de/publikationen

Līga Stafecka May 2026 Who Does(Not) Vote in Latvia? Unequal Democracies 

Contents Why Electoral Participation Matters ...................................  3 How Latvia Compares in Electoral Participation .........................  4 Who Votes? Social Patterns of Participation...........................  6 Who Does Not Vote in Latvia?......................................  10 Regional Inequalities in Voter Turnout...............................  12 Latvia’s Unique Participation Dynamics ...............................  14 Conclusions and Recommendations ..................................  16 References ......................................................  18 Quellenverzeichnis .................................................  4

Why Electoral Participation Matters Democratic equality rests on the principle that all citizens should have a genuine opportunity to influence collective decisions. Elections are the primary channel through which this principle is realized: they enable voters to evaluate political elites, articulate preferences, and shape future governance(Verba, 2003). When participation becomes uneven, this communicative link weakens, disproportionately amplifying the voices of socially advantaged groups and reducing the representativeness of elected institutions (Lijphart, 1997). Extensive comparative research shows that unequal turnout is not a marginal flaw but a structural challenge: lower participation systematically widens gaps between high- and low-resource groups and concentrates political influence among those already better positioned (Hicks& Swank, 1992; Mahler, 2008; Mueller& Stratmann, 2003). The persistence of such participation inequalities constitutes the central concern of the Unequal Democracies(UD) research series. As emphasized by the UD comparative report, unequal participation remains an unresolved dilemma of democracy, where systematic non-voting among disadvantaged groups undermines political equality(Elsässer, Schäfer,& Wenker, 2022). Latvia’s experience reflects these broader democratic risks. Following exceptionally high turnout in the early 1990s – driven by the restoration of independence – participation has gradually declined and stabilized at more moderate levels. Since then, turnout has remained relatively stable but socially uneven. Younger voters, citizens with lower income or education, and residents of economically weaker regions consistently participate less than older and more economically secure groups. These patterns place Latvia close to the Eastern European cluster identified in the Unequal Democracies comparative report, where declining ­participation coincides with widening socioeconomic and ­gen erational divides(Elsässer, Schäfer,& Wenker, 2022). At the same time, Latvia exhibits several context‑specific dynamics that warrant closer attention. One is the significant and widening turnout gap between Latvian‑ and Russian‑speaking voters, driven by declining political efficacy, heightened social tensions, and increasingly fragmented information environments. Another concerns the weak societal embeddedness of political parties: despite a large number of formally registered parties, membership levels remain extremely low, internal structures are fragile, and parties often struggle to establish stable links with voters. Together, these factors create a political environment in which participation is not only stratified but also shaped by the country’s distinct historical and institutional context. This report, prepared within the FES Unequal Democracies series, examines how these dynamics manifest in Latvia. It analyses turnout trends over time, identifies the social groups most likely to abstain, and explores the structural, attitudinal, and institutional mechanisms underpinning persistent inequalities in participation. It addresses three key questions: 1. How has voter turnout evolved in Latvia, and how does it compare with similar European countries? 2. Which social groups participate least, and why? 3. What specific barriers – socioeconomic, ethnolinguistic, and institutional – shape non-voting in Latvia today? By answering these questions, the report aims to provide an evidence-based understanding of who does – and does not – vote in Latvia, and to outline practical directions for strengthening democratic inclusion ahead of the 2026 par liamentary elections. 1 1  Voter participation in Latvia has not been extensively studied, and detailed data on voter behaviour beyond standard public opinion surveys remain limited. As a result, this report partly relies on secondary data sources. It is well established that survey respondents tend to overreport electoral participation. For example, turnout in the 2022 elections was 59.41%, while 79% of survey respondents claimed they had voted. Accordingly, the findings presented here should be interpreted as indicative rather than precise measures of voter behaviour. Why Electoral Participation Matters 3

How Latvia Compares in Electoral Participation In Latvia, voter turnout declined steadily after peaking in the early 1990s. Participation fell throughout the 2000s, reaching a low point of 55 % in the 2018 parliamentary elections( Figure 1). A particularly sharp drop occurred in the 2006 elections, when turnout decreased by around 11 percentage points. This decline is largely attributed to a series of political corruption scandals that eroded public trust in political elites, combined with the absence of new political actors capable of mobilising protest voters. Over the past two decades, turnout has stabilized within a relatively narrow range of approximately 55–63 %. In 2011, Latvia held extraordinary parliamentary elections following a presidentially initiated referendum to dissolve the Saeima, which had been in office for only one year. The referendum was triggered by a constitutional conflict between the president and the parliamentary majority, after the legislature repeatedly obstructed corruption-re­ lated investigations and refused to authorize a search of a member of parliament’s property. In response, the president called for the dissolution of the parliament. The most recent elections, held in 2022, brought 59 % of eligible vot ers to the polls, marking a modest recovery. The next parliamentary elections are scheduled for 2026. The trajectory of high electoral participation in the early 1990s followed by more moderate levels in subsequent decades is not unique to Latvia but reflects a broader pattern across the Baltic states. Estonia and Lithuania have experienced similar developments, with turnout levels converging over time. Within this shared trajectory, Estonia stands out for maintaining slightly higher and more consistent levels of participation than both Latvia and Lithuania (Figure 1) . Voter turnout in parliamentary elections across the Baltic states 100 Share ( %) 90 80 70 60 50 40 30 1990 1995 2000 2005 2010 2015 Latvia Estonia Lithuania Source: UD dataset and data from the CEC of Latvia, as well as the author‘s compilation of recent turnout data for other Baltic states. 4 Friedrich-Ebert-Stiftung e.V. Figure 1 2020 2025

From a broader European perspective, Latvia’s turnout levels remain below average. Compared with the European median, its turnout is about 10 percentage points lower. Latvia resembles a typical post-­ transition democracy that has stabilised at the lower end of the European turnout spectrum of around 60–70%. At these levels, it falls well behind the high-­turnout countries of Northern and Western Europe – such as the Nordic states, the Benelux countries, and Germany – where partici pation rates often reach 80–90 %. At the same time, it aligns more closely with countries such as Estonia and Portugal(around 60–63 %), while remaining above the low er-turnout group in mostly Southern and Southeastern Europe, including Greece and Bulgaria, but also Lithuania (approximately 40–53 %). 2 Overall, Latvia reflects a broader Eastern European pattern characterised by comparatively low turnout and more pronounced social inequalities in electoral participation. 2  Election turnout data are obtained from national election commissions or national statistical offices in the respective countries. For example, in the Nordic region, recent ­elections show high voter participation: Iceland recorded 80.15 % in 2024, Norway 80.12 % in 2025, and Finland, as well as Germany, reached 82.5 % in their 2025 elections. The Netherlands also reported strong engagement, with 78.4 % turnout in the 2025 elections. In contrast, voter participation in several other European countries is notably lower. Latvian voters remain more active compared to Lithuania(52 % in 2024), Bulgaria(39 % in 2023), and Greece(53 % in 2023). How Latvia Compares in Electoral Participation 5

Who Votes? Social Patterns of Participation Voter turnout in Latvia is not only volatile across elections but also systematically stratified along well-defined social lines. Elections represent the institutional moment when citizens are formally equal; however, the evidence shows that different groups make unequal use of this opportunity. These disparities align with the classic resource model of political participation, which posits that political engagement is largely shaped by individuals’ access to time, money, and civic skills(Brady, Verba& Schlozman, 1995, as cited in the UD study). In Latvia, the most pronounced divides run along age, education, and socioeconomic status(including class, sector, and income). The data present a consistent pattern: older cohorts participate the most, individuals with tertiary education maintain high turnout across electoral cycles, and higher-status groups – such as those in upper service occupations, public sector employees, and higher-­ income brackets – vote far more reliably than routine work ers, the unemployed, and low-income citizens. Gender differences exist but are less pronounced compared to other factors. Taken together, these patterns help explain not only who votes, but also whose preferences are more likely to shape democratic outcomes. Across all observed elections, turnout is highest and most stable among voters aged 45–59 and 60+, while those under 30 consistently remain the least active. Approxi mately 38 % of voters report participating out of a sense of civic duty and hope for a better future, although this motivation has declined by nearly 8 percentage points over the past twelve years(SKDS, 2022). Voter turnout in Latvia by age group 100 Figure 2 Share ( %) 90 80 70 60 50 40 30 2006 2008 2010 Under 30 30–44 2012 45–59 2014 Over 60 2016 2018 2020 2022 Source: UD dataset; for 2018 and 2022, data are supplemented with survey results from the Central Election Commission of Latvia, commissioned from SKDS(Vēlētāju attieksmju pētījumi/ Voter Attitude Studies). 6 Friedrich-Ebert-Stiftung e.V.

In the 2006 elections, only around 40 % of voters under the age of 30 reported participating. While youth turn out has improved over time – rising by nearly 20 percen­ tage points by 2022(Figure 2) – a substantial gap rela tive to older cohorts persists. This underrepresentation is also reflected in political office: in the 2025 municipal elections, only 2.2 % of elected local politicians were aged 30 or younger, whereas individuals aged 50–60 accounted for 29 % of municipal councils members. 3 Educational attainment is strongly associated with both the level and stability of electoral participation, indicating persistent inequalities along educational lines. People with higher education(completed university degree) demonstrate the highest and most consistent turnout, maintaining participation rates at or above 80 % across election cycles( Figure 3). Those who only completed upper secondary education(middle level) vote relatively consistently, typically in the low- to mid-70% range, with a slight upward trend by 2022. In contrast, voters with no complet ed secondary education(low level) remain the least politically active, with turnout fluctua­ ting between approximately 55% and 65% and declining again after 2018. The gap between highly educated and low-educated voters is substantial – around 36 percentage points. Socioeconomic indicators – such as income, employ ment sector, and occupational class – reveal similarly consistent patterns. Groups with higher socioeconomic status exhibit significantly higher and more stable turnout. High-income individuals and those in upper service occupations maintain participation rates near or above 85 %, whereas low-income voters remain the least en­ gaged, with turnout falling close to 60 % in several elec tion years(Figure 4). Differences across employment sec tors mirror this divide: public-sector employees are consistently more likely to vote than private-sector workers and the unemployed, although this gap has narrowed somewhat since 2018(Figure 5). This pattern is not driven solely by income stability but also by factors such as a stronger sense of civic duty. While turnout has improved in recent years among middle-income groups and private-sector employees, the overall structural gap persists. 3  Central Election Commissions collected statistics about the candidates of the 2025 municipal elections(https://dati.cvk.lv/PV2025/ieveletie-deputati/). Voter turnout in Latvia by education level 100 Figure 3 Share ( %) 90 80 70 60 50 Voters with a low level of education are consistently the the least politically active. 40 30 2006 2008 Low level of education 2010 2012 2014 Medium level of education 2016 2018 2020 High level of education 2022 Source: UD dataset; for 2018 and 2022, data are supplemented with survey results from the Central Election Commission of Latvia, commissioned from SKDS(Vēlētāju attieksmju pētījumi/ Voter Attitude Studies). Who Votes? Social Patterns of Participation 7

Voter turnout in Latvia by income level 100 Share ( %) 90 80 70 60 50 The low-income group of voters remains the least electorally active across all election years. 40 30 2010 Income: Low 2011 Lower middle 2014 Middle Upper middle Note: Average monthly income per household member(after taxes), including all income sources: low – up to€319; lower middle –€320 to€409; middle –€410 to€519; upper middle –€520 to€709; high –€710 and above. Source: CEC, SKDS. 2018 High Figure 4 2022 Election year Voter turnout in Latvia by sector of employment 100 Share ( %) 90 80 70 60 50 State-sector employees are consistently more likely to vote than ­private-sector workers and unemployed individuals. 40 30 2010 State sector Source: CEC, SKDS. 2012 2014 2016 Private sector Unemployed 2018 2020 8 Friedrich-Ebert-Stiftung e.V. Figure 5 2022

Voter turnout in Latvia by gender 100 Figure 6 Share ( %) 90 80 70 60 50 40 30 2006 Men 2008 2010 Women 2012 2014 2016 2018 2020 2022 Source: UD dataset; the results for 2018 and 2022 are supplemented with data from the Central Election Commission’s surveys commissioned from SKDS— Vēlētāju attieksmju pētījumi(Voter Attitude Studies) for 2022 and 2018. Taken together, these findings suggest that socioeconomic inequalities translate into unequal political voice, with higher-­ status groups exercising disproportionately greater influence in electoral politics. From a gender perspective, women in Latvia consistently report higher turnout than men. However, the size of this difference remains modest compared to disparities based on age, education, and socioeconomic status(Figure 6). In this sense, gender functions as a secondary cleavage, aligning with broader international patterns identified in the UD study. Comparative research also points to the“law of dispersion”: in systems with lower overall turnout, parti­ cipation gaps tend to widen more sharply. Put differently, lower overall turnout tends to make existing differences in participation between social groups more pronounced – a pattern that closely fits the Latvian case. Socioeconomic inequalities translate into unequal political voice, with higher-­ status groups exercising disproportionately greater influence in electoral politics. Who Votes? Social Patterns of Participation 9

Who Does Not Vote in Latvia? Latvia has one of the highest shares of non-voters in the European Union. Alongside France, it ranks among the countries with the largest proportion of citizens who have either not voted at all in the past five years or have done so only occasionally. Approximately 17 % of Latvian citizens fall into this category of chronic non-voters(Social Cohesion Report, 2023). Non-voters in Latvia form a clearly identifiable socio-­ demographic group, strongly shaped by education and income. Citizens with lower levels of education and lower incomes are significantly more likely to abstain from voting, while higher levels of education and income are associated with more consistent electoral participation. As illustrated in Figure 4, turnout among low-income citizens is around 18 percentage points lower than among high-in come groups. This pattern is also evident from a regional perspective ( Figure 7), where voter turnout closely correlates with income. Latvia exhibits greater inequality in electoral participation than Western and Northern European democracies and aligns more closely with the Eastern European cluster identified in the UD study. Youth participation remains consistently low. Young citizens – particularly those under 30 – demonstrate the weak est electoral engagement: only about 68 % of surveyed individuals in this group reported voting in the 2022 elec tions, which is 15 percentage points lower than among those aged 60 and above. Although research on youth political engagement in Latvia remains limited, available data point to the importance of a supportive environment in fostering participation. Findings from a recent youth survey suggest that many young people do not feel that their views are taken seriously or that they can meaningfully influence decisions that affect their lives. Only 17 % of respondents believe that participation in national-level decision-making leads to outcomes that matter to them. More broadly, youth participation depends heavily on the existence of a supportive environment. Young people feel most comfortable expressing their views within close personal networks, but significantly less so in institutional or public settings. While 55 % report that it is“fully possible” to discuss important issues with friends, only 11 % feel able to do so with teachers or educational staff, 12 % in public discussions, and just 8 % within political organisa tions. Around half report having little or no opportunity to express their views in these environments(Latvian Academy of Culture, 2026). Low electoral participation among youth thus reflects a broader pattern of limited civic engagement and a perception that their voices are neither heard nor influential. Incorporating attitudinal factors provides a more nuanced understanding of non-voting. Non-voters tend to exhibit lower levels of political trust, weaker interest in politics, and lower satisfaction with the functioning of democracy. Many attribute their unfavourable socioeconomic circumstances to political actors or institutions. A key driver of abstention is a lack of belief in the importance and effectiveness of elections, combined with low political efficacy. Only 56 % of Latvians consider parti­ cipation in elections important – the lowest rate in the European Union – and just 42 % believe that their vote makes a difference. Social Cohesion Report, 2023 This contributes to a broader scepticism about the ability to influence political decision-making. Survey data from 2023 further indicate that non-participa tion is closely linked to political dissatisfaction. Approximately 59 % of the population report being dissatisfied with the political situation in Latvia, with particularly high levels of dissatisfaction among low-income groups and Russian-speaking residents(Social Cohesion Radar, 2023). Similar patterns are evident across other forms of civic engagement – such as signing petitions, contacting public officials, or participating in protests – where participation is strongly associated with socioeconomic status, citizenship, and language spoken at home 4 . Structural factors(such as low education, low income, age, and language background) and motivational factors reinforce one another. Socioeconomically disadvantaged citizens are more likely to feel politically alienated, and this sense of alienation further reduces participation, creating a self-reinforcing cycle of inequality. Overall, Latvia closely 4 According to the State Statistical Bureau, in 2025 8,9 % of inhabitants were; of all ethnic Russians approximately 25 % are non-citizens. 10 Friedrich-Ebert-Stiftung e.V.

follows the broader patterns identified in the UD comparative study: voter turnout is strongly stratified by education, income, and occupational class, with socioeconomic disadvantage serving as a key predictor of electoral abstention. The Latvian case also reflects the finding that generational inequalities are particularly pronounced, while gender differences remain comparatively modest. At the same time, Latvia diverges from the typical UD pattern in two important respects. First, inequalities in turnout are more pronounced than the UD average, placing Latvia closer to the Eastern European cluster characterised by both lower overall participation and wider participation gaps – especially among younger voters. Second, Latvia exhibits a significant ethnolinguistic divide, particularly in the lower and declining participation of Russian-speaking voters, a dimension not fully captured by the UD study’s primarily socioeconomic framework. In this sense, Latvia represents both a textbook case of the socioeconomic dynamics identified by the UD project and a distinct context in which additional, country-specific factors contribute to political inequality. Socioeconomic disad­vantage is a key predictor of electoral abstention in Latvia. Who Does Not Vote in Latvia? 11

Regional Inequalities in Voter Turnout Turnout levels in Latvia reveal a pronounced regional divide. The highest participation rates are found in municipalities surrounding the capital, Riga – particularly in the suburban Riga region, the country’s most economically dynamic area. Several of these municipalities regularly exceed 70 % turn out, including Saulkrasti(84 %), as well as Ķekava and Ādaži (both around 75 %). In contrast, the lowest turnout is consis­ tently recorded in Latgale, Latvia’s economically weaker and more remote eastern region, located further from Riga and bordering Russia and Belarus(Figure 7). In Latgale, participation frequently falls ­below 55 %, with Rēzekne munici pality reaching as low as 50 %. Data analysis confirms that income is the strongest predictor of regional variation in electoral turnout. As illustrated in Figure 8 , there is a clear positive association between muni­ cipal turnout and median registered household in­come per household member(Pearson’s r= 0.58, p< 0.001; R²= 0.34), 2022 Latvian Election Turnout by Municipalities Figure 7 13 35 11 20 21 Baltic Sea Estonia 18 12 23 24 17 14 1 3 4 7 16 27 9 25 5 37 39 2 6 36 38 22 28 15 31 33 19 8 43 26 41 10 32 29 30 Lithuania 40 42 Municipalities of Latgale 34 Russia Municipality 1 Saulkrastu novads 2 Ķekavas novads 3 Ādažu novads 4 Siguldas novads 5 Mārupes novads 6 Ogres novads 7 Cēsu novads 8 Varakļānu novads 9 Rīga 10 Rēzekne 11 Kuldīgas novads Turnout Municipality Turnout Municipality 83.90 % 74.89 % 73.99 % 70.96 % 70.36 % 69.53 % 69.52 % 68.21 % 67.55 % 67.23 % 66.85 % 12 Valmieras novads 66.80 % 13 Ventspils 66.70 % 14 Smiltenes novads 66.25 % 15 Jelgava 65.86 % 16 Tukuma novads 65.78 % 17 Talsu novads 65.77 % 18 Limbažu novads 63.32 % 19 Bauskas novads 62.82 % 20 Liepāja 62.79 % 21 Dienvidkurzemes nov. 62.78 % 22 Madonas novads 62.75 % 23 Valkas novads 24 Alūksnes novads 25 Ropažu novads 26 Aizkraukles novads 27 Jūrmala 28 Saldus novads 29 Līvānu novads 30 Preiļu novads 31 Dobeles novads 32 Jēkabpils novads 33 Jelgavas novads Turnout Municipality Turnout 62.18 % 62.07 % 61.90 % 61.85 % 61.25 % 61.09 % 61.04 % 60.58 % 60.51 % 59.96 % 59.37 % 34 Daugavpils 58.96 % 35 Ventspils novads 57.73 % 36 Gulbenes novads 57.13 % 37 Salaspils novads 56.32 % 38 Balvu novads 56.04 % 39 Olaines novads 55.85 % 40 Krāslavas novads 54.11 % 41 Ludzas novads 53.64 % 42 Augšdaugavas novads 52.15 % 43 Rēzeknes novads 50.05 % Source: Central Election Commission. 12 Friedrich-Ebert-Stiftung e.V.

Median registered income per household member (EUR) Voter turnout in Latvia by municipal income level 950 850 Mārupe Ādaži Figure 8 750 Saulkrasti 650 550 450 Krāslava 350 45 50 55 60 65 70 75 80 85 90 Election turnout(2022) in municipalities of Latvia(%) Note: Scatter plot shows a moderate positive association between municipal voter turnout and median registered household income per household member in 2022. Pearson r= 0.58(p< 0.001; 95 % CI[0.34, 0.75]), R²= 0.34, n= 42 municipalities; populationweighted r= 0.60. Municipallevel aggregates; not interpretable at the individual level (risk of ecological fallacy). Sources: Central Election Commission(turnout) and Central Statistical Bureau of the Republic of Latvia(median registered household income per household member, 2022; experimental statistics). indicating that higher-income municipalities tend to exhibit higher participation. More affluent areas – particularly in the Riga metropolitan region – consistently report the highest turnout, reflecting the concentration of higher-income house­ holds. Conversely, the lowest participation rates in the 2022 elections – just over half of eligible voters – were recorded in several municipalities in Latgale, the country’s most economically disadvantaged region. This pattern is further reinforced by demographic trends: over the past three years, Latgale has experienced the most pronounced population decline in Latvia, losing approximately 5 % of its population – four times faster than the national average. 5 By contrast, the relationship between turnout and the population’s age structure(specifically the share of residents aged 18–29) is weak and statistically insignificant(r= –0.087), suggesting that age composition alone does not explain regional disparities. A descriptive examination indicates that higher turnout often coincides with higher levels of educational attainment. For example, municipalities such as Mārupe, Ādaži, and Saulkrasti combine relatively high turnout with a larger share of residents holding tertiary education, whereas municipalities such as Rēzekne, Krāslava, and Augšdaugava exhibit both lower turnout and lower educational levels. While this pattern is suggestive, more detailed analysis would be required to determine whether it holds when controlling for other factors. 6 Although socioeconomic factors explain much of the variation, certain cases highlight the importance of local political context. For instance, Rēzekne municipality recorded the lowest turnout in the country, while Rēzekne city – administratively separate and covering only the urban population – experienced significantly higher participation (67.23 %). The presence of a prominent and controversial mayor, along with a more competitive political environment, likely contributed to stronger voter mobilisation in the city compared to the surrounding district. 5 Central Statistical Bureau of the Republic of Latvia Official database“Iedzīvotāju skaits pēc tautības reģionos 2000–2024”, calculated by the author. 6  Central Statistical Bureau of the Republic of Latvia Official database on the level of education of the inhabitants of Latvia by region. Regional Inequalities in Voter Turnout 13

Latvia’s Unique Participation Dynamics Latvia’s electoral landscape is shaped not only by socio-­ economic divides but also by distinctive contextual factors. One of the most significant is the country’s ethnolinguistic composition, where a sizeable Russian-speaking population plays a key role in understanding participation dynamics. Another important factor is the weakly institutionalised party system, characterised by low membership, fragile organisational structures, and limited societal embeddedness – all of which undermine citizens’ trust and sense of political representation. Approximately 23 % of Latvia’s population is of Russian ethnicity, and about one quarter of this group holds non-­ citizen status. 7 Prior to the 2014 elections, Russian­-speaking voters demonstrated relatively high electoral participation. However, over the past decade, their turnout has declined significantly(Figure 9). By 2022, the gap in participation bet ween Latvian- and Russian-speaking voters had widened to 12 percentage points. This decline coincided with a broader decrease in overall voter turnout, but it was particularly pronounced among Russian-speaking voters. A key contributing factor was the stagnation of Saskaņa(Harmony) – historically the most influential, though not the only, party appealing to this electorate. In the 2022 elections, Saskaņa lost eight parliamentary seats compared to the previous cycle, and its earlier decision to publicly criticise Russia’s 2014 aggression in Ukraine had already caused internal divisions and alienated parts of its traditional voter base. These dynamics further eroded voter confidence and contributed to the party’s near collapse. Nearly half of Russian speakers(48.8 %) feel they have no chance of influencing politics, while another 23.5 % believe their opportunities are very limited(Kaprāns et al., 2025). As a result, this group often perceives that people like them cannot meaningfully affect political outcomes, which reduces motivation to vote and contributes to low levels of perceived electoral efficacy. Compared with Latvians, Russian-­ speaking residents have historically shown lower levels of political participation, but Russia’s full-scale war against Ukraine has significantly changed the context and deepened this gap(Kaprāns et al., 2025). Since 2022, political disen gagement among parts of the Russian-speaking population has increasingly taken the form not only of low participation, but also of cynicism, alienation, apathy, self-censorship, fear of expressing views openly, and withdrawal from political issues, including news consumption(Kaprāns et al., 2025). This suggests that the war has made this group less reachable through ordinary democratic participation channels and has turned political disengagement into a broader issue of social cohesion and democratic inclusion in Latvia. Media consumption patterns also differ between Latvianand Russian-speaking audiences. This fragmentation matters because Russian-speaking voters are exposed to partly different political narratives, which may shape their perceptions of parties, state institutions, and the war in Ukraine in distinct ways. Nevertheless, a political offer for Russian-­ speaking citizens does exist: several parties actively target this electorate, increasing competition for these voters. It is important to note, however, that Russian speakers are not a homogeneous group: they differ in terms of ethnic origin (Russians, Ukrainians, Belarusians) as well as in their political identities, levels of attachment to Latvia, and attitudes toward Russia’s war. Thus, the existence of a political offer for Russian-speaking citizens does not necessarily translate into clear or stable political representation. A 2025 study indicates that Russian-speaking residents often feel apprehensive about expressing their political opinions due to persistent social tensions surrounding language and ethnicity. These tensions have intensified further following Russia’s full-scale invasion of Ukraine, which has mobilised sup­ porters of nationally oriented political parties. This dynamic was evident, for example, in the 2025 municipal elections in the ethnically diverse capital, Riga. Another Latvia-specific factor influencing voter turnout is the weak institutionalisation of the party system, which has struggled to embed itself in society, partly due to historical legacies(Saarts, 2011). The linkage between political parties and voters remains weak: parties are fragmented, ideological positions are often un­clear, and connections between social cleavages and party support are blurred. Organisational capacity is limited, and public trust in political parties is low. At the same time, parties are generously funded by the state and are among the wealthiest in the Baltic region(KNAB, 2025). 7  Central Statistics Bureau data, describing the demographic situation in 2025. 14 Friedrich-Ebert-Stiftung e.V.

Voter turnout in Latvia by family language 85 Figure 9 Share ( %) 80 75 70 2010 2011 Use Latvian language in family 2014 Use Russian language in family Source: Central Election Commission of the Republic of Latvia–commissioned surveys conducted by SKDS(2010–2022). Although Latvia formally has a large number of political parties – 81 registered in 2021 – most have very small mem bership bases. By comparison, Estonia has 12 parties and Lithuania 27. The legal threshold for establishing a party in Latvia is relatively low: only 500 members are required, rep resenting approximately 0.032 % of the electorate. Political activism within parties is also underdeveloped: Just 1.4 % of Latvian residents are party members, 8 – the lowest proportion in the Baltic region – compared to around 3 % in Estonia and 5 % in Lithuania. ERR, 2023; Lithuanian Ministry of Justice, 2023 Even the largest party,“Unity,” has just over 2,000 members. Together, these factors weaken voters’ sense of representation and political efficacy, ultimately depressing electoral participation. Research by UD highlights a reinforcing feedback loop: parties with limited resources tend to focus mobilisation efforts on groups that already vote, thereby deepening disengagement among low-participation groups – a pat tern that is particularly evident in Latvia. 2018 2022 Election year 8  Based on the author’s calculation, there were approximately 26,000 party members out of a population of 1,857,000 at the beginning of 2025, as reported by political parties. Latvia’s Unique Participation Dynamics 15

Conclusions and Recommendations This report demonstrates that non-voting in Latvia is neither random nor socially neutral. Electoral participation is shaped by a combination of socioeconomic inequalities, ethnolinguistic dynamics, and institutional features of the party system. Several key patterns stand out: First , age, education, and income remain the strongest predictors of electoral participation. Older, better-educated, and higher-income citizens vote at substantially higher rates than younger people, low-income earners, and those with lower levels of education. These divides have persisted across electoral cycles and mirror patterns identified in the Unequal Democracies comparative analysis. The result is a structurally unequal distribution of political voice, in which those facing greater socioeconomic vulnerability are also the least likely to participate. Second , ethnolinguistic divisions have deepened, producing a widening gap between Latvian- and Russian-speaking voters. Russian-speaking residents increasingly report lower political efficacy, limited trust, and reluctance to express political views openly, particularly in the context of heightened social tensions following Russia’s full-scale invasion of Ukraine. This environment not only demobilises parts of the electorate but can also indirectly increase mobilization among nationally oriented parties, reinforcing political polarisation. Improving channels of representation for Russian-speaking voters – while maintaining dem ocratic values and national security considerations – will be complicated but essential to prevent long-term political alienation. Third , the weak societal embeddedness of political parties contributes to lower turnout. Although Latvia has a large number of formally registered parties, their membership bases remain very small. Many parties lack the organisational capacity, ideological clarity, and local presence needed to sustain durable links with voters. This weak embeddedness limits opportunities for civic engagement, undermines political trust, and reinforces perceptions that politics is distant and unresponsive. Addressing these challenges requires a targeted set of measures that tackle both structural barriers and motivational factors shaping political participation. Boosting youth turnout requires approaches that reflect how young people engage with politics and where they feel their voices matter. Strengthening civic education, expanding municipal-level youth initiatives, and creating structured co-creation opportunities in policy areas that directly affect young people – such as education, housing, and mobility – can help reduce feelings of political distance. Evidence shows that supportive environments are crucial: young people are more likely to engage when they feel heard not only within families but also in schools and universities. A strong sense of belonging in educational settings is closely linked to higher levels of civic participation. Latvia already provides several participation tools for young people aged 16 and above, including participatory budgeting and collective petitions via the platform manabalss.lv. These instruments should be more actively promoted. A promising recent innovation is Signālvēlēšanas (“signal elections”), first piloted during the 2025 municipal elections and planned for repetition in the 2026 parliamen tary elections. These simulated elections allow 15–17-yearolds to cast symbolic votes, learn electoral procedures, and compare their choices with actual results. Such early, lowstakes exposure can reduce informational and psychological barriers to participation. Although lowering the voting age to 16 has been debated, public support for this meas ure remains limited(27 % in 2020; TV3, 2020). Increasing participation among low-income citizens requires addressing underlying socioeconomic inequalities that constrain political engagement. Latvia remains among the more unequal countries in the European Union, with a Gini coefficient of around 34(Eurostat, 2023). Reducing income dis parities is therefore central to narrowing participation gaps. Political actors should prioritise clear and credible socio­ economic policies – particularly on wages, housing afforda bility, and public services – and communicate them through accessible and targeted outreach. Independent media also play an important role in ensuring that these issues remain visible and relevant. A persistent challenge is the self-reinforcing cycle of political disengagement: individuals who feel unrepresented are less likely to vote, which in turn further reduces their representation. The concerns of low-income groups are more often brought onto the political agenda by populist actors, who mostly operate from opposition. As a result, they have had limited opportunities to demonstrate whether they can actually deliver on their promises to these voters. At the same time, they tend to reinforce distrust: while their political stance may resonate with disadvantaged voters, ambiguity on other important issues such as Russia’s aggression 16 Friedrich-Ebert-Stiftung e.V.

against Ukraine undermine their credibility and make their political offer appear risky. Ensuring credible, programmatic (rather than purely identity-based) political representation is essential for fostering inclusion without reinforcing divisions. Tailored, issue-based communication that addresses everyday concerns(such as cost of living and social security) can help reconnect low-income citizens to the political process. Addressing the ethnolinguistic participation gap is particularly complex in the current geopolitical context. Political engagement among minority communities is shaped by heightened sensitivities linked to security concerns and regional tensions. Creating safe spaces for political expression and supporting cross-community initiatives can help foster trust and shared civic identity. The introduction of mandatory participatory budgeting in all municipalities since 2025 offers an opportunity to shift attention toward common local concerns rather than divisive identity issues. Finally, rebuilding the link between political parties and society is crucial. This requires stronger incentives for parties to expand membership and actively engage supporters – for example, by linking part of public funding not only to electoral performance but also to active membership. Parties should also strengthen internal democracy through open primaries, transparent candidate selection, and regular public engagement. Expanding local party structures, particularly in regions with persistently low turnout such as Latgale and rural municipalities, is equally important. Increasing transparency in the use of public funding can further contribute to rebuilding public trust. Conclusions and Recommendations 17

References Elsässer, L., Schäfer, A.,& Wenker, J.(2022). Unequal democracies: Who does(not) vote? Voter turnout trends in the OSCE region since 1970. Friedrich-Ebert-Stiftung, https://collections.fes.de/publikationen/ident/fes/19356. ERR News(2023). Parties see no problem with membership declining, https://news.err. ee/1609362446/parties-see-no-problem-with-membership-declining. Eurostat(2025). Living conditions in Europe – income distribution and income inequality. European Commission, https://ec.europa.eu/eurostat/statistics-explained/index.php? title=Living_conditions_in_Europe_-_income_distribution_and_income_inequality. Hicks, A. M.,& Swank, D. H.(1992). Politics, institutions, and welfare spending in ­industrialized democracies. American Political Science Review, 86(3). Kaprāns, M., Žabko, O., Juzefovičs, J., Tetarenko-Supe, A.,& Mieriņa, I.(2025). Latvijas rusofono iedzīvotāju ģeopolitiskie priekšstati, pilsoniskā līdzdalība un mediju lietošana 2025: aptaujas rezultātu prezentācija. Latvijas Universitāte/ Rīgas Stradiņa universitāte/ Norstat. Kažoka, I.,& Bērziņa, L.(2023). Sabiedrības saliedētības radars 2023. Sabiedrības in tegrācijas fonds, https://www.sif.gov.lv/lv/media/5598/download?attachment. KNAB(2025). Informatīvais ziņojums“Par valsts budžeta finansējuma piešķiršanas modeļa politiskajām organizācijām(partijām) izvērtējumu Latvijā no 2018. gada līdz 2023. gadam”. Latvian Academy of Culture(2026). J Preconditions for authentic youth participation in formal and non-formal education settings(UNFRAMED), National Research Programme“Education” project No. VPP-IZM-Izglītība-2023/6-0002, survey data presenta tion; https://www.lps.lv/uploads/docs_module/4. %20A.La %C4 %B7e %20 LPS_26012026_UNFRAMED.pdf Lijphart, A.(1997). Unequal participation: Democracy’s unresolved dilemma. American Political Science Review, 91(1), 1–14. Lithuanian Ministry of Justice(2023). Data on political party membership in Lithuania. Mahler, V. A.(2008). Electoral turnout and income redistribution. European Journal of Political Research, 47. Mueller, D. C.,& Stratmann, T.(2003). The economic effects of democratic participa tion. Journal of Public Economics, 87. Saarts, T.(2011). Comparative party system analysis in the Baltic states: The case of Estonia. Studies of Transition States and Societies, 3(3), 102–127, https://www.ssoar. info/ssoar/handle/document/36380. SKDS(2010). Vēlētāju attieksmju pētījums 2010: Latvijas iedzīvotāju aptauja, https://www.cvk.lv/lv/media/129/download?attachment. SKDS(2011). Vēlētāju attieksmju pētījums 2011: Latvijas iedzīvotāju aptauja, https://www.cvk.lv/lv/media/128/download?attachment. SKDS(2014). Vēlētāju attieksmju pētījums 2014: Latvijas iedzīvotāju aptauja, https://www.cvk.lv/lv/media/122/download?attachment. SKDS(2018). Vēlētāju attieksmju pētījums 2018: LR pilsoņu aptauja, https://www.cvk.lv/lv/media/117/download?attachment. SKDS(2022). Vēlētāju attieksmju pētījums 2022: pēcvēlēšanu aptauja, LR pilsoņu ap tauja, https://www.cvk.lv/lv/media/110/download?attachment. TV3 News(2020, 25. jūnijs). Vai balsstiesības būtu jāpiešķir no 16 gadu vecuma? Rai dījums“900 sekundes”, https://tv3.lv/zinas/latvija/sabiedriba/aptauja-vai-balsstiesibas-butu-japieskir-no-16-gadu-vecuma/. University of Latvia(2022). Vēlētāju aktivitātes samazināšanās cēloņu un to ietekmē jošo faktoru izvērtējums. LU Sociālo zinātņu fakultāte, https://ppdb.mk.gov.lv/wp-content/uploads/2022/02/Veletaju-lidzdalibas-petijums-LU_PETIJUMS.pdf. Verba, S.(2003). Would the dream of political equality turn out to be a nightmare? Perspectives on Politics, 1(4), 663–679. 18 Friedrich-Ebert-Stiftung e.V.

About the author Līga Stafecka is Senior Policy Analyst at the Centre for Public Policy PROVIDUS. Her work focuses on good governance, public administration, anti-corruption and participatory democracy, with a particular interest in how institutions can become more open, accountable and responsive to society. She has long-standing experience in policy research, advocacy, and has authored or co-authored numerous analytical reports and policy recommendations on governance, public participation and integrity systems in Latvia and internationally. She is currently pursuing a PhD in Public Administration at the University of Latvia.

Who Does(Not) Vote in Latvia? → Why should Latvians care about turnout? Latvia’s voter turnout has stabilized at moderate levels but remains marked by deep and persistent inequalities. Participation gaps by age, education, income and region mean that political decisions are disproportionately shaped by older, better-off, and better-educated citizens. This unequal political voice undermines democratic representation and reinforces broader social and territorial divides. → Who are the Latvian nonvoters? In line with European patterns, younger people, citizens with lower income and education, and those living in economically weaker regions participate least. Additionally, Latvia shows two context-specific dynamics: a decline in turnout among Russian-­ speaking residents, driven by low political efficacy and heightened social tensions especially since Russia’s invasion of Ukraine, and very weak societal embeddedness of political parties, reflected in extremely low membership levels and fragmented linkages between voters and parties. → What should be done? Reducing these participation gaps requires ­targeted action: strengthening civic education and youth engagement, improving programmatic outreach to low-income voters, providing safe spaces for the largest ethnic communities, and rebuilding political parties’ social roots. Municipal-level tools such as participatory budgeting and new initiatives like pilot elections for youth under 18 can also help foster early civic habits and broaden democratic inclusion. ­You can find further information on this topic here: ↗ democracy.fes.de/topics/inequality-democracy.html , Illustration, Titelfoto arbfläche Icon oder F