The Vicious Circle: Does disappointment with political authorities contribute to political passivity in Latvia? Open Access

The recent financial crisis demonstrated that in some countries even small cuts in wages or social benefits can trigger notable civic response, while in others presumably corrupt and unjust behavior of public officials results in seemingly no reaction at all. One of the countries that were hit the hardest by the recent global economic crisis is Latvia. What made the recession there deeper was the failure of the government to stop the overheating of the economy that was driven by high property prices and easily available loans. The highest GDP growth in Europe was replaced by a sharp decline. From late 2007 to late 2009 Latvia lost about 24% of its GDP in what was ‘the worst loss of output in the word, […] enormous by any international or historical comparison’ (Weisbrot and Ray 2011: 1–4). Official unemployment rose from 5.3% in late 2007 to 20.5% in early 2010 and the average disposable income per person decreased by 20%. Instead of currency devaluation, the government opted for a massive internal deflation. Given the magnitude of the budget deficit, this required dramatic cuts in the already low public spending¹ and resulted in a series of unpopular decisions: closing of schools and hospitals, deep cuts in wages, social benefit reductions and tax raises. Economists were initially sceptical about the prospects of successfully implementing the austerity measures, fearing civic unrest (Harrison 2008; Hugh 2008; Roubini 2009). Nevertheless, the reaction of citizens to the tough austerity measures turned out to be surprisingly passive. Now the Latvian ‘success story’ is often brought forward as an example that the other weaker economies could potentially follow. But why did the deep discontent of citizens in Latvia not erupt in a broad civic unrest? How can one explain the passive acceptance of decreasing living standards, especially in the light of strikes and massive demonstrations in Greece, Spain, United Kingdom and other countries? And why are the civil society institutions that could help mobilizing people not developing more successfully (Howard 2003; Letki 2004; Koroleva and Rungule 2006; Makarovič et al. 2007; Morales and Geurts 2007)? This paper is an attempt to shed some light on these questions.

Using the Latvian case I show that political passivity can be partly caused by a vicious circle starting from unsatisfactory performance of political authorities and leading to disenchantment from politics and a weakness of civil society. The loss of political trust reduces: (1) the sense of external political efficacy, (2) perceived internal efficacy, and (3) social trust, thus alienating citizens from politics and hindering participation in both conventional and unconventional political activities.

Securing peace and stability during a process of internal deflation requires that citizens have trust in economic and political authorities to accept temporary economic straits in return for the promise of better conditions in some uncertain future (Catterberg and Moreno 2006). Yet in Latvia, same as in other post-Soviet countries, confidence in government is, and has been for the past 15 years, extremely low (Rose and Shin 2001; Zepa 2001; Howard 2003; Catterberg and Moreno 2006; Koroleva and Rungule 2006; Ijabs 2007; Lagerspetz 2009).² Despite the very high electoral volatility (Birch 2003) and the fact that during 20 years of independence none of the prime ministers managed to keep his post for a full 4-year term, the perceived responsiveness of the government is decreasing rather than increasing (International Social Survey Program, ISSP 1996, 2006). The perceived levels of corruption have constantly been among the highest in Europe (Karklins 2005; Transparency International 2011). Moreover, other factors seem to increase the risk of social unrest: the disproportionately negative impact of the crisis on youth, high levels of inequality, poverty and social exclusion (Eurostat 2011), as well as persisting tensions between the two major ethnic groups: Latvians (59.3%) and Russians (27.8%). The socio-economic context makes the observed political passivity even more surprising.

Although people in Latvia complain about the government, they abstain from active participation in political life (Fuchs and Klingemann 2006; Koroleva and Rungule 2006; Lagerspetz 2009). According to European Values Study in 2008 72% of people aged 18 or over did not belong to any organizations or associations. Moreover, 42–43% indicated that they would never sign a petition or attend a lawful demonstration, while 78% would never join a strike, and 93% would never occupy a building or factory. Thus, Latvia provides an interesting case that can shed new light on the complexity of the formation of political attitudes and on the drivers of civic behavior.

Social scientists usually try to explain political attitudes and behavior from two broad perspectives: the cultural and the institutional perspective (Mishler and Rose 2001, 2005; also Denters et al. 2007). Similarly, social capital researchers explore the sources of social capital (reflected in participation in associations and civic activism in general) using society-centered approach and institution-centered approach (Hooghe and Stolle 2003). Table 1 summarizes these approaches in a theoretical framework that this paper builds on.

From the point of view of cultural theories the disenchantment of citizens from politics and their civic passivity is a legacy of the communist regime. It is believed to have undermined interpersonal trust, discouraged any out-systemic initiatives and created a very specific type of political culture characterized by political apathy, scepticism, and distrust towards authorities (Sztompka 1996; Rose and Shin 2001; Howard 2003, among others). This approach has been very popular among researchers of post-communist countries, and has been thoroughly analyzed in the previous literature, therefore it lies outside the scope of this paper.

From the perspective of social capital theories civic incompetence and passivity is a result of inadequate transmission of democratic norms, skills and values trough family, school, workplace and voluntary associations (Almond and Verba 1989; Putnam et al. 1993; Pietrzyk-Reeves 2008, among others). Lack of trust is believed to be hugely responsible for the low citizen involvement in political life.

From the institutional perspective political attitudes are politically endogenous — people build their opinions on the basis of their prior experiences and their interpretation of these experiences. The institutional approach emphasizes the ‘top-down’ development of political attitudes, and holds that institutions play a big role in promoting democratic values and behavior in the society. Lack of confidence in institutions is thought to reflect the dissatisfaction of citizens with the institutional output of political authorities, and their perceived corruption. The perceptions of internal and external efficacy are also believed to be based on individuals previous experience in dealing with institutions. Accordingly, political cynicism and civic passivity are mainly a result of the performance of the contemporary political institutions, experience in dealing with these institutions, and rational expectations of what benefits participation might bring (Madsen 1987; Brehm and Rahn 1997; Stolle 1998; Rothstein and Stolle 2003; Rothstein 2004).

These three approaches do not contradict each other; they are rather complimentary (Mishler and Rose 2001). What they disagree on is how big is the role of institutions. If political attitudes are mostly a ‘bottom-up’ phenomenon there is still something that can be done ‘from above’, yet the opportunities are limited and the mechanisms would be different. Studies focusing on the role of institutions in the development of civil society usually emphasize the importance of direct government interventions in the voluntary sector: financial support for NGO's, tax exemptions, policies supporting the formation of voluntary associations or, on the contrary, suppression of free speech, repressions against dissidents, etc. Yet, in my view an aspect that has been mostly overlooked so far is the socializing role of the government's actions, policies and communication with citizens as well as the psychological mechanisms underlying this kind political learning. Only recently some studies have begun to appear (Stolle 2003; Rothstein and Stolle 2003; Rothstein 2004).

To understand what drives political activism, let us first examine the major factors known to influence political participation. Many studies have revealed a quite consistent and robust correlation between confidence in political authorities and conventional political participation. If confidence in political institutions is low, people are less likely to follow politics, vote or join associations (e.g., Almond and Verba 1989; Brehm and Rahn 1997; Dalton 2004; Mishler and Rose 2005). The democratic theory also holds that low confidence should provoke mobilized unconventional activity (Pateman 1975; Norris 1999; Dalton 2004). At the same time, there are scholars (e.g., Craig and Maggiotto 1981) who argue that for low political support to produce unconventional political activity people also need to have a sense of political efficacy – ‘the feeling that an individual political action does have, or can have an impact upon the political process’ (Campbell et al. 1952). Efficacy is also needed for conventional political participation. If people believe that they are likely to gain some benefits as a result of their actions and that these will improve their living standards, they are more likely to become actively involved in politics – to follow politics, to discuss politics, to be more active partisans and to be more engaged in political activities (Almond and Verba 1989; Wittman 1989; Dalton 2004).³ If, however, they do not believe in the responsiveness of political authorities or their own capability to have a real impact on politics, they will probably choose not to waste their time.

Psychologists and political scientists (Bandura 1977; Madsen 1987; Kim 2005; Morales and Geurts 2007) usually distinguish between: (1) external efficacy describing the perceived responsiveness of the government and (2) internal efficacy reflecting an optimistic view of one's own political capabilities. According to Francis Lee (2006), one also needs to consider ‘collective efficacy’ – a citizen's belief in the capabilities of the public as a collective actor to achieve social and political outcomes. Perceived efficacy is itself a product of social learning; it is based on individuals’ previous experience (Bandura 1977). It can decrease as a result of unsuccessful attempts to influence political processes (Madsen 1987; Brehm and Rahm 1997; Stolle 1998). Several scientists have noted that the subjective sense of efficacy also depends on trust in institutions (Zepa 2001; Mishler and Rose 2005). ‘Trust strengthens citizens’ beliefs that government is responsive and encourages citizens to express their demands via participation in activities from voting to joining organizations’ (Mishler and Rose 2005: 1054).

Besides efficacy, another major factor affecting citizen's political activity is generalized trust. In the social capital literature, participation in social networks and trust are considered as being in tight reciprocal relationship, constantly reinforcing each other (Brehm and Rahn 1997; Putnam 2002). Trust is necessary for people to engage in collective action (and most political action is ‘collective’) and join voluntary associations (Uslaner and Brown 2005). However, trust, along other civic virtues and democratic skills, develops in everyday interactions with others in family, school, civic associations (Putnam 1993, 2000; Brehn and Rahn 1997; Herreros 2004; Letki 2004; Zmerli et al. 2007) or even elite-challenging mass action (Welzel et al. 2005).

It is still unclear what the connection between social and institutional trust is. Some scholars have found that social trust is positively related to trust in institutions (Brehm and Rahn 1997; Rothstein 2004; Denters et al. 2007; for post-communist countries Catterberg and Moreno 2006), while others argue that the correlation between social and institutional trust is, in fact, week and/or insignificant (Uslaner 2003; for post-communist countries – Rose and Mishler 2001, 2005; Letki 2004). Moreover, there are disagreements about the direction of linkages. From the social capital perspective, confidence in political institutions is politically exogenous, i.e., it is as an outcome of interactions between individuals – an extension or reflection of social trust which is learned in interactions with people and later projected onto political institutions (Almond and Verba 1989; Putnam et al. 1993; Shyns and Nuus 2007). Scholars representing the institutional view, however, argue that institutions have a very important role in facilitating the creation of social trust, and that institutional trust ‘tickles down’ to social trust (Brehm and Rahn 1997; Rothstein and Stolle 2003; Stolle 2003; Rothstein 2004; Mishler and Rose 2005; Denters et al. 2007). Some of them believe that the impact stretches far outside the pure policy-related issues: (1) by setting a personal example, the government sends signals to citizens about what kind of ‘game’ is being played in the society, what kind of attitudes and behavior is rewarded and expected in this society (Putnam 2002; Rothstein 2004); (2) people make inferences about their system experiences and extend them to everyone else living under the same system (Stolle 2003). If the government is untrustworthy, people may generalize their feelings to their fellow citizens; and (3) the connection might run through the political institutions influencing the behavior of people first, which leads to re-evaluation of interpersonal relations and trust (Rothstein and Stolle 2003).

According to institutional theories, and particularly the ‘performance hypothesis’ that is well rooted in the political science literature, confidence in institutions is based on rational evaluations of the performance of contemporary political authorities. If citizens are satisfied with the institutional output, and see politicians as honest and responsive, the feelings of confidence evolve (Rothstein and Stolle 2003; Rothstein 2004; Catterberg and Moreno 2006; for post-communist countries Mishler and Rose 2001, 2005). If, however, the government does not live up to expectations, trust decreases (Norris 1999). Researchers have found that it is especially important how successful the institutions are in dealing with such matters as promoting growth, governing effectively, and, especially in new or transitional regimes, avoiding corruption. Confidence in authorities can also be affected by the inability of the government to explain their decisions and to guarantee transparency (Russell 2005; Stoker 2008).

Usually researchers analyze the direct relations between particular indicators (social and institutional trust, trust and participation, etc.). However, if we put together what is known from previous studies, a complex model of direct and indirect linkages emerges (Figure 1). If all of them hold, we can conclude that the better the political authorities perform, the more people will trust them, the more trusting and efficacious they will feel and the more likely they will be to take part in democratic processes. Active citizen's engagement in political processes is, in turn, expected to have a positive reverse effect on the governance and, especially, the accountability of politicians (Almond and Verba 1989; Knack 2002; Herreros 2004; Welzel et al. 2005).

A few studies (Herreros 2004; Zmerli et al. 2007) have recently been exploring a similar, yet simpler, virtuous circle in the context of reciprocal relations between trust, associational membership, and democratic political participation. Yet, if we think about post-communist countries it becomes clear that it is rather short-sighted to only focus on the ‘virtuous circle’ and the positive long-term effects of good governance. Instead, one should ask what happens if the performance of authorities is continuously poor and disappointing? Following the previous logic, there seems to be a risk of getting trapped in a ‘vicious circle’, and reach another kind of equilibrium characterized by poor institutional performance, low trust and efficacy, and week civil society. It would suggest that persistently poor performance and unresponsiveness of the government institutions risks alienating citizens from the state and politics and creating a politically passive society, not able or willing to defend their interests by using democratic means. This study aims to test the plausibility of the described circle in Latvia.

The analysis is based on the ISSP 2006–2007 data, taking advantage of the fact that in Latvia these two survey waves – each containing some information needed for the analysis – were administered together.

The most suitable method for exploring the plausibility of complex models such as the ‘vicious circle’ is Structural Equation Modelling (SEM). The variables in SEM may influence one-another reciprocally, either directly or through other variables as intermediaries. The focus in SEM is not as much on uncovering particular relationships between the variables, as on testing whether the overall web of relationships adequately describes the data (Fox 2002; Kaplan 2009).

The data was gathered in face-to-face interviews, and the survey covered inhabitants of Latvia 18–74 years of age. Original questions included in the analysis, as well as more information about the sampling, data collection, translation of the questionnaire, response and outcome figures and the data, are available on the GESIS ZACAT website. Initial data set contained 1069 interviews, however a number of observations were excluded from the analysis due to missing values.⁴ The final model is based on 578 cases.

To reduce the complexity of the model, it was decided to separate the measurement and structural part of the SEM. Based on 22 variables Principal Components Analysis (PCA) with orthogonal factor rotation model (Varimax criterion) extracted, in six iterations, eight factors accounting for 68% of the variation in the data (factor matrix available on request). Confirmatory factor analysis (maximum likelihood with Promax rotation) was used to calculate the final factor scores. Three factors characterize the institutional output:

• ‘Performance’ [PERF]: perceived performance of the government with respect to controlling crime, dealing with threats to security, fighting unemployment and protecting environment;⁵

• ‘Corruption’ [CORR]: perception that politicians and public officials are not corrupt;

• ‘Trust in political authorities’ [I-TRUST]: trust in MP's efforts to keep promises, and trust in civil servants. We can treat these items as describing confidence in political authorities in general, for in post-communist countries trust in political institutions is essentially one-dimensional, i.e., people do not distinguish among political institutions (Mishler and Rose 2005).

Perceived political efficacy is captured by two factors:

• ‘Internal efficacy’ [INT-EFFI]: how much confidence people have in their own political competence, i.e., how interested and knowing they are about politics.

• ‘External efficacy’ [EXT-EFFI]: confidence that the average citizen can influence politics, and that people like the respondent have a say in what the government does. This factor combines the classical efficacy measures, describing the perceived capability of citizen's to influence politics.

And partly also by:

• ‘Generalized trust’ [G-TRUST]: trust in people in general. Considering the importance of trust for collective action, it can be considered as a proxy for collective efficacy (see Almond and Verba 1989).

Finally, political behavior is captured by two factors:

• ‘Unconventional activities’ [UNCON-ACT]: approval or support for public protest meetings, demonstrations and anti-government strikes;⁶

• ‘Institutionalized participation’ [INST-PART]: frequency of participation in interest groups, non-political voluntary associations such as sports or cultural associations, church or religious organizations, community-service or civic association/group, and political parties or organizations.

In the next step the latent variables were included in a path analytic model relating endogenous variables to exogenous variables and one to another. Relations or lack of relations between them (restrictions on the model) were specified according to the theory, as discussed in the introduction.

There are some differences between the final SEM (Figure 2) and the theoretical model (Figure 1). Initially the analysis was meant to cover four types of participation: (1) mobilized conventional participation; (2) unconventional participation; (3) participation in political groups or associations; and (4) participation in non-political voluntary associations. Unfortunately, there were no reliable measures of mobilized conventional participation available in the Latvian data. Voting in the last elections could not be used due to the high number of non-citizens in the country that are not eligible to vote. Due to a high risk of systematic error, this question was not included in the analysis. Secondly, participation in political and non-political associations was so highly correlated that treating them as separate factors was not justified. Based on the results of factor analysis participation in political and non-political associations was merged. This move is also theoretically justified, for individual's potential for political involvement and activity may increase with membership in both types of organizations (Almond and Verba 1989; Lelieveldt and Caiani 2007). Finally, in the SEM the feedback loops from participation to performance and corruption were not specified. Individual level survey data cannot uncover the existence or non-existence of such link, therefore for the argument about the positive feedback effect of civic participation on the governance we shall rely on previous surveys. The model also includes measurement errors for endogenous and exogenous variables and allows for covariance among measurement errors ε₃ and ε₁, and covariance between ‘Performance’ and ‘Corruption’.

The model is estimated by the ADF (asymptotically distribution-free) method (see Fox 2002), and it was calculated using SPSS Amos 7.0.

The model fit indices of the initial model (not shown here) were not satisfactory, so the model was respecified by eliminating the links that were not significantly different from ‘0’ even at the 90% level (p>.1). Among them were the hypothesized links leading from participation in associations to generalized trust and internal efficacy. The fact that our model does not provide support for the classical assumptions of the social capital theory is actually not surprising. A growing number of studies reveal that the role of civic participation and associational membership in developing trust and democratic attitudes is most likely overstated. Despite the huge amount of studies trying to find evidence for the so called ‘Putnam's hypothesis’, researchers have failed to find a strong, systematic connection between participation and trust at the micro level (Hooghe and Stolle 2003; Morales and Geurts 2007; Zmerli et al. 2007; for post-communist countries Letki 2004; Mishler and Rose 2005, but for critique and positive findings – Hererros 2004). According to Badescu and Neller (2007) in post-communist countries the connection between trust and participation is even weaker than elsewhere.

The results of the SEM confirm that distrust in political authorities discourages institutionalized political participation, even though the effect is small (α=.13). However, the hypothesized link between the confidence in institutions and unconventional political action was not statistically significant. Finally, the model fit was significantly improved by adding a direct link from performance of political authorities to generalized trust.

The final respecified model is shown in Figure 2. Chi-square statistics for the model is very satisfactory (chi-square=12.7 with 13 degrees of freedom), and the p-value of .468 suggests that the model cannot be rejected. The goodness-of-fit index (GFI) and the adjusted goodness-of-fit index (AGFI) that measure the descriptive adequacy of the model allow to conclude that the model fits the data very well. The same conclusion can be drawn from RMSEA (root mean-squared error approximation), which is an estimate of fit of the model relative to a saturated model in the population. The fact that it is almost 0, indicates a very good fit of the model. And, finally, the normal fit index (NFI) of 0.94 means that the overall fit of the tested model is 94% better than that of an independence model (where the variables are assumed to be uncorrelated with each other), based on the sample data.

In order to test the robustness of results, a recursive model without the two feedback loops was also calculated (available on request). The model fit indices suggested that this model is slightly worse than the non-recursive model, but it also could not be rejected (chi-square 26.5, df=15, p=.033, RMSEA=0.037, GFI=0.987, NFI=877). Overall, the results suggest that the ‘vicious circle’ – starting from the poor performance and unresponsiveness of political authorities, and leading to disenchantment from politics and a weakness of civil society – is indeed a real possibility.

The data strongly supports the ‘performance hypothesis’: low institutional trust is largely a result of high perceived corruption (α=.45) and dissatisfaction with the performance of the authorities (α=.21). Covariance between these two indicators is also significant.

Second, as expected, I find very clear evidence of the importance of political efficacy for political participation. External efficacy has a moderately strong effect (α=.25) on institutionalized participation: if people believe that the average citizen does not have any influence on politics they will be less likely to engage in any social or political groups. Perceived internal efficacy has a significant yet not very strong impact on both types of political participation analyzed in this study: if people are not into politics and/or feel that they do not understand it, the participation rates both in voluntary associations and unconventional political activities will be low.

The sense of external political efficacy, however, is affected by the confidence in political authorities (α=.22). If people do not trust political authorities they tend to see them as unresponsive, and do not expect to be listened to even if they expressed their dissatisfaction. In addition (and partly as a result of the perceived unresponsiveness of the authorities), internal political efficacy decreases too⁷ : people become disenchanted from politics and lose confidence in their own political competence. These effects lead to them abstaining from institutionalized conventional political activities. Whether people will show support for unconventional political activities depends on their sense of internal political efficacy and their trust in each other. Unfortunately, these attitudes are themselves negatively influenced by the lack of confidence in political institutions, largely resulting from their poor performance and corruption.

Performance of political authorities has both a direct and indirect impact (mediated by trust in political authorities) on how much trust people will have in their fellow citizens in general. This result is in line with some previous research (e.g., Muller and Seligson 1994) showing that week institutional performance can spread general distrust throughout society. If an individual believes that those who do not play by the rules will be punished, it increases his trust in other people (Gambetta 2000; Rothstein and Stolle 2003; Herreros 2004). It also provides further support for the argument that trust in institutions ‘tickles down’ and facilitates social trust among people in general. At the same time, just like Mishler and Rose (2005), we do not find that generalized trust (or distrust) ‘spills up’ to institutions (p=.307). It means that trust in political institutions is essentially endogenous.

Even though generalized trust is not related to institutionalized participation in Latvia, trust seems to have a small (α=.13) effect on support for unconventional activities. This conclusion is easy to understand from the perspective of collective action and the collective efficacy thesis. People need to have trust in others, and have confidence that they are also going to contribute to the common good, otherwise they are less likely to contribute anything themselves.

One can conclude that distrust in political authorities discourages institutionalized political participation both directly (α=.13), and indirectly, via trust and efficacy. At the same time, neither the trust in government, nor their perceived responsiveness allows to form clear expectations regarding the likelihood of unconventional mobilized activities.

Although I do not find a direct link between institutional trust and unconventional activities, there is an indirect effect which leads through internal efficacy and generalized trust. It means that the reaction of people in the case of institutional distrust is mediated by their perceived efficacy and generalized trust. However, this indirect effect is weak (smaller than .10). This is probably why the relations between distrust in institutions and unconventional activities are sometimes found to be inconsistent or insignificant. If for some reasons people develop high internal efficacy, it can ‘outweight’ the lack of generalized trust and still initiate unconventional activities.

This paper represents an attempt to better understand political apathy and civic passivity in the face of huge disappointment with the decisions of political authorities, such as was the case of Latvia during the recent economic crisis. Studies on post-communist countries, such as Latvia, typically advance cultural explanations to political passivity, yet the results of this study suggest that institutional explanations deserve far more attention than they usually get with regards to these countries. We need to critically analyze current processes more, besides looking for explanations in the past. It is not just the path-dependency of political culture that is to blame for the political apathy of the society but also the continuously disappointing performance and low responsiveness of political authorities. Whether institutions promote growth, avoid corruption, are effective in enforcing the law, and prove to be responsive and trustworthy affects the norms and values that prevail in the society: the dispositions, expectations and perceptions of people regarding other people, themselves and their role as citizens.

In Latvia, the poor performance of institutions has led to disenchantment of people from politics and loss of both internal and external political efficacy, thus decreasing participation in conventional political activities. It has also spread general distrust throughout the society, creating an additional obstacle to unconventional activism. As civic participation has direct consequences for the quality of governance this creates a vicious circle.

From the point of view of the rational choice theory there are two kinds of Nash equilibriums of political participation that can be reached: the high or ‘civic’ and the low or the ‘uncivil’ equilibrium. Both of them are, in essence, rational expectation traps (Wittman 1989). As noted by David Wittman (1989), people's behavior is guided by rational expectations regarding the potential outcomes of their actions and the actions of other political actors. Both the political leaders and the masses have certain expectations about the behavior of others and about what can be gained or lost if a certain course of action is taken. If the masses do not feel that they are listened to and do not expect any benefits from participation, they will choose to abstain. If politicians see the masses as incompetent, passive and unable to hold them accountable, poor decisions and corruption will tend to persist. Surveys suggest that this is exactly what is happening in Latvia, where neither the majority of the public nor the representatives of the state or local governments believe in the efficiency of active forms of public influence such as NGOs and initiative groups (Menshikov 2001; Jansone and Vilka 2007). This negative equilibrium can be considered a sign of democratic failure. Because there is certain inertia in people's perceptions and expectations creating a ‘learning lag’ this Nash trap might be hard to get out of. Apart from fighting corruption and restoring confidence in political authorities, reducing poverty and social exclusion and raising the education level of the population might be one of the best available solutions to increase the political activity of citizens.

One other theoretical framework that could be used to try to better understand the formation and functioning of the vicious circle is the learned helplessness theory that is well established in social psychology. The basic idea of the learned helplessness theory is that, if individuals learn that events (in our case, politics) are uncontrollable, that is, independent of their own behavior, it has a debilitating effect on their future performance, and results in an ‘inappropriate passivity’ in situations when effective coping is possible (Abramson et al. 1978; Peterson et al. 1995; Overmier 2002). Learned helplessness theory has been successfully applied when analyzing depression, academic achievements, and problems related to racial minorities. So far, it has never been tested at the group level in the field of politics even though such tests have been suggested (Peterson et al. 1995). Analysis of learned political helplessness would ideally require longitudinal data. Still, the main assumptions of the learned helplessness theory fit extremely well with the results of this study. In line with the mechanisms described in this theory, in Latvia the disappointing performance and unresponsiveness of political authorities has negatively affected the sense of political efficacy, having an additional detrimental effect on interpersonal trust and resulting in political apathy and decreased participation in all kinds of political activities. Thus, it seems that the concept of ‘learned helplessness’ at the group level could indeed be successfully applied to analyzing political behavior.

Finally, the very high rate of emigration Latvia has been experiencing, especially during the recent economic crisis,⁸ suggests that citizens’ activity might be exogenous. Instead of trying to influence the situation in their country ‘from within’, many are choosing an ‘exit’ strategy as an alternative to the ‘voice’ option (see Hirschman 1970). Studies on emigrants show that this preference is not just based on economic factors; it is also a result of people having lost their belief in citizens’ political efficacy with respect to political authorities and hope that the situation at home can be improved (Hazans 2011).

The analysis is based on the case of Latvia. However, as noted by Rose and Shin (2001), many of the post-communist countries have became incomplete ‘broken-back’ democracies, falling into the low-level equilibrium trap, where poor performance and low trustworthiness of elites are matched by mass scepticism, disenchantment from politics, low civic efficacy and low expectations. Considering the communist past, similarities in political and economic context after the collapse of communist regimes, the cultural closeness (Fuchs and Klingemann 2006; Inglehart 2006), and similarities in political attitudes and behavior among the citizens of post-communist countries (Makarovič et al. 2007),⁹ it seems reasonable to assume that similar mechanisms might be at work causing political apathy and leading to low rates of civic participation in other post-communist countries too; especially those with semi-democratic governments, such as Russia. Moreover, institutional distrust in Russia is negatively affecting social trust (Mishler and Rose 2005), further hindering the ability of people to organize for collective action. As shown on the example of ‘vicious circle’, rampant corruption, rigged elections, controlled media and empty rhetoric from partisan politicians clearly contribute to alienation of citizens from politics. Effective functioning of democracy is hardly possible under the conditions of continuously poor performance and unresponsiveness of political authorities.

Participation is driven by a number of factors, and the ones described here cannot possibly tackle the whole complexity of political behavior. Yet, my hope is that this study will provoke further debates and contribute to a better understanding of the psychological and social mechanisms that lie behind the political behavior of citizens.

I would like to thank the two anonymous reviewers for their valuable comments and suggestions. I would also like to thank the AABS foundation for making possible my research visit to the University of Illinois at Chicago with the support of Mudītes I. Zīlītes Saltups scholarship, during which most of the paper was written. I am indebted to Martin Paldam, Natalia Letki, and my colleagues at the University of Latvia for inspiring my work and suggesting important improvements.

Dr Inta Mieriņa is an assistant professor at the Institute of Sociology, University of Warsaw, where she is a researcher in the ERC Starting Grant ‘Public Goods through Private Eyes: Exploring Citizens’ Attitudes towards Public Goods and the State in East-Central Europe'. She is also a researcher at the Institute of Philosophy and Sociology, University of Latvia. Her main research interests are related to post-communist democratization and the formation of political attitudes. Her recent publications explore the relations between the citizens and state in Central and Eastern Europe, as well as the sources of the weakness of civil society in post-communist countries.