Explaining the 'democratic malaise' in unequal societies: Inequality, external efficacy, and political trust

European Journal of Political Research rr :rr –rr , 2023 1 doi: 10.1111/1475-6765.12611 Explaining the ‘democratic malaise’ in unequal societies: Inequality, external efficacy and political trust SIMON BIENSTMAN, SVENJA HENSE & MARKUS GANGL Institute for Sociology, Goethe-University Frankfurt, Germany Abstract. Previous scholarship suggests that rising inequality in democracies suppresses trust in institutions. However, the mechanism behind this has not clearly been identified. This paper investigates the proposition that income inequality leads to increased democratic distrust by depressing perceptions of external efficacy. Based on time-series cross-sectional survey data from the European Social Survey, we find that changes in income inequality have a negative effect on changes in political trust and external efficacy. Causal mediation analysis confirms that inequality affects trust through lower efficacy. Further analyses show that this efficacy-based mechanism does not depend on political orientation. As a direct effect remains among left-wing respondents, our empirical results indicate that inequality affects trust via both a mechanism of substantive output evaluation and a process-based evaluation that measures of external efficacy can capture. These findings highlight the empirical and theoretical relevance of this so far neglected mechanism and provide a potential solution for the puzzle that inequality depresses trust also among those for whom inequality is not politically salient. Keywords: political trust; external efficacy; income inequality; mediation analysis Introduction As most advanced democracies have witnessed significant increases in inequality over the last decades (Atkinson, 2015; Piketty, 2017), the question of whether and how economic inequality affects trust in democratic institutions has become more salient. And indeed, several scholars have documented a negative relationship between economic inequality and political trust for a wide range of countries (e.g., Anderson & Singer, 2008; Lee et al., 2020; Zmerli & Castillo, 2015). If inequality and trust move together (albeit in opposite directions), an overall increase in inequality should entail a decline in trust. Nevertheless, research on inequality’s effects on attitudes towards democracy remains inconclusive. Some scholars either see no overall decrease in trust (or closely related concepts like satisfaction with democracy) or find no clear relation with inequality (Kim et al., 2022; Magalhães, 2014; Martini & Quaranta, 2020; Norris, 1999; van Beek et al., 2019). To help advance the literature and possibly reconcile these contradictory findings, this article provides further empirical evidence and investigates the mechanisms underlying a potential relationship between inequality and citizens’ democratic orientations. In what follows, our focus will be on political trust, and thus on, in Easton’s (1975) famous words, citizens’ ‘reservoir of goodwill’ towards the actors and institutions of democratic governance, and on European democracies. Concerning the study of the political implications of inequality, we aim for a threefold contribution. First, we contribute new empirical estimates for the effect of inequality on political trust. We specifically aim to improve on the existing literature by isolating an over-time effect of rising inequality on changes in citizens’ trust. Most of the existing literature is based on cross-sectional designs and has therefore been unable to identify © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 2SIMON BIENSTMAN, SVENJA HENSE, & MARKUS GANGL whether (recent) changes in economic inequality are contributing to what has been described as the ‘democratic malaise’ (e.g., Foa et al., 2020) of declining support for democracy in the Western world. Drawing on time-series cross-sectional survey data from the European Social Survey (ESS), we provide firm evidence that changes in inequality are indeed negatively correlated with changes in political trust. Given extensive controls for observables and country-specific unobservables as afforded within the context of a fixed effects (FE) regression specification and given an extensive set of further robustness checks, our estimates substantiate a causal role of inequality for political trust. Inevitably, however, any such causal claim carries a degree of uncertainty, and we will therefore discuss the relevant methodological issues and limitations of our research throughout the paper. In addition, we seek to expand on earlier research by testing the full causal chain between inequality and citizens’ trust in democratic institutions when taking external efficacy as an important, but so far also largely overlooked mediator. When reflecting on why inequality (or other aspects of macroeconomic performance) should matter for political trust, the currently dominant theoretical perspective suggests to regard trust in institutions as the consequence of a conscious evaluation process on the citizens’ side. This ‘trust-as-evaluation’ approach (see van der Meer, 2018) sees income inequality as one output of democratic governance, which citizens then evaluate in view of their political preferences, and trust is extended when citizens conclude that the political system is meeting their demands. Yet we argue that this type of substantive and output-based evaluation is not necessarily the only mechanism through which inequality may affect citizens’ trust in democratic institutions, but that a second, more process-based type of evaluation needs to feature more prominently. In making this argument, we build on a tradition in political science which has long argued that inequality in economic resources translates into inequality in political resources and consequently into unequal political representation and responsiveness (Dahl, 1971, 2006; Goodin & Dryzek, 1980; Schattschneider, 1960; see also Polacko, 2022). Rising inequality may be presumed to affect how citizens perceive the distribution of political power and how they see their capabilities to influence the political process. With this type of mechanism in operation, inequality would erode political trust by depressing citizens’ evaluations of the fairness of democratic processes but not because of citizens’ concern about inequality as a political problem. Our study is the first to confirm the existence of this second transmission channel from empirical data. Our analysis shows that rising income inequality depresses citizens’ perceptions of external efficacy and that lower degrees of efficacy are a determinant of lower trust in institutions. Importantly, these findings are not meant to deny the relevance of output-based performance evaluations. Instead, our results suggest that outputand process-based mechanisms are complementary. We find that efficacy is an important mediator for the relationship between inequality and political trust. However, a direct effect of inequality remains that cannot be explained via efficacy. In addition, and in line with earlier findings (e.g., Anderson & Singer, 2008), we see that the direct effect of inequality is moderated by respondents’ political ideology, as predicted from the assumption that left-leaning citizens react more sensitively to economic inequality because it is a politically salient issue for them. However, the same type of ideological moderation is not visible in the efficacy channel, where leftand right-leaning citizens alike show a declining sense of external efficacy when economic inequality is rising, and where similar declines in efficacy perceptions generate similar responses for political trust. © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License INEQUALITY, EFFICACY AND POLITICAL TRUST 3 We believe that placing a perspective on citizens’ evaluation of political power and fair democratic processes alongside the standard perspective of output-based evaluations enriches the argument that democratic stability may be undermined when economic inequality is translated into political inequality. In addition, drawing attention to the largely non-partisan transmission channel of citizens’ efficacy perceptions may resolve the empirical puzzle of why inequality also depresses political trust among right-leaning citizens, for whom (rising) inequality would presumably not be an important political problem in the first place. Before coming back to the implications of our work, in what follows we first provide further theoretical background to our study and then discuss our data, methodology and empirical findings. Theory and previous research Economic inequality has risen in many Western democracies over the last decades (see Atkinson, 2015; Piketty, 2017, and the online Supporting Information for descriptives for the countries in this study). At the same time, observers and political scientists alike are asserting a ‘democratic malaise’ (e.g., Foa et al., 2020), which is mirrored in gloomy titles such as ‘The People vs. Democracy’ (Mounk, 2018), ‘How Democracies Die’ (Ziblatt & Levitsky, 2018) or ‘How Democracy Ends’ (Runciman, 2018). This malaise is seen as consisting of a longstanding decline in citizens’ support for the processes and institutions of democratic governance (Dalton, 1998; Foa et al., 2020; Hetherington, 1998) and thus threatening the functioning of democracy itself (Lipset, 1963; Easton, 1975). As a central indicator of political support, citizens’ trust in the institutions of representative democracy has repeatedly been shown to be negatively related to inequality (Anderson & Singer, 2008; Goubin & Hooghe, 2020; Lee et al., 2020). Studies on satisfaction with democracy and on other related indicators of political support have likewise reported negative associations with income inequality (Andersen, 2012; Christmann, 2018; Donovan & Karp, 2017; Krieckhaus et al., 2014; Schäfer, 2013). Other scholars remain more sceptical as to whether democratic attitudes may be affected by income inequality in particular, or macroeconomic performance in general (e.g., McAllister, 1999; Kim et al., 2022; see van der Meer, 2018, for a recent review of the literature) As pointed out by van der Meer (2017, 2018), one likely reason for the inconclusive state of evidence is that most studies have been based on cross-sectional data. This evidently makes drawing inferences about how changes in inequality may lead to changes in trust questionable, not the least because of the problem of omitted variables and unobserved heterogeneity between countries. For example, countries may be more or less equal, but they also may exhibit a more or less longstanding cultural tradition of democratic governance. To approach the question on the democratic consequences of rising inequality, longitudinal designs are thus clearly preferable (van de Werfhorst & Salverda, 2012). However, few large surveys exist that include questions about political trust for a broad range of countries with diverging levels and trajectories of inequality over many years. Still, research that has been able to draw on longitudinal data corroborates the idea that inequality affects democratic orientations. In particular, rising inequality was found to lower political trust (Goubin & Hooghe, 2020) and satisfaction with democracy (Christmann, 2018). Closely related studies show that over-time changes in inequality are associated with populist party support (Engler & Weisstanner, 2021), and this has partly been attributed to decreasing levels of political trust (Stoetzer et al., 2021). Other research shows that political trust dipped during the © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4SIMON BIENSTMAN, SVENJA HENSE, & MARKUS GANGL financial crisis in Europe – a time when unemployment and inequality increased in many European countries (Gangl & Giustozzi, 2021; Hooghe & Okolikj, 2020). Why would inequality affect trust in the first place? A prominent argument in the sociological literature holds that inequality creates social dysfunctions because increased social distances come with increased status anxiety and declining social trust (Buttrick & Oishi, 2017; Delhey & Steckermeier, 2020; Wilkinson & Pickett, 2010). Although rarely invoked as an explanation for political effects of inequality, studies on populist voting have recently argued that rising inequality has spurred status fears and thereby the likelihood of voting for populist parties (Gidron & Hall, 2017; Engler & Weisstanner, 2021). Regarding social trust, it could furthermore be argued that if inequality decreases generalized trust, it should indirectly also depress trust in democratic institutions, given that social and political trust have often been argued to be strongly related (Newton et al., 2018; Zmerli & Newton, 2008). In political science, the currently most influential perspective is the ‘trust-as-evaluation’ approach (see van der Meer, 2018, for a review) that regards citizens’ trust in democratic institutions as the outcome of a substantive evaluation of democratic performance. Citizens extend trust if they perceive political outcomes and processes to meet their expectations, and inequality is regularly seen as one relevant dimension of macroeconomic performance that citizens consider in their evaluation (e.g., Martini & Quaranta, 2020). The study by Anderson and Singer (2008) provides support for this view, as it has shown that inequality depresses trust and satisfaction with democracy primarily among left-wing respondents. This fits with an understanding that citizens with egalitarian preferences will expect inequality to be tackled politically and will, therefore, also evaluate the political system on the basis of their substantive policy preferences. Yet, substantive evaluation may not be the only mechanism to link economic inequality to institutional trust. Anderson and Singer (2008) did not only find striking partisan differences in the democratic relevance of inequality, but they also observed that inequality nevertheless depresses trust among right-leaning respondents – even when economic inequality is not likely to constitute a politically salient outcome to these citizens. This observation clearly is a puzzle to the standard trust-as-evaluation perspective (see also Kumlin et al., 2018), but the puzzle may be resolved when we broaden our perspective and consider an additional mechanism that operates via citizens’ perceptions of efficacy. More specifically, our key argument here is that citizens may evaluate the democratic process just as much as democratic output. If so, then inequality will be relevant, too, because rising economic inequality is likely to translate into political inequality, which in turn is likely to have consequences for how citizens perceive the responsiveness of the political system and their own capabilities to influence the political process. Against the strong democratic norms of political equality and equity, deteriorating perceptions of the fairness of democratic processes clearly have the potential to depress the level of trust that citizens are placing in the actors and institutions of democratic governance. The argument that we are making here is in fact far from being novel but is one with a long tradition in political science. According to Dahl (1971), democracy rests on the premise of intrinsic equality – the idea that every human being is fundamentally equal and that, therefore, the interests of members of a democracy are of equal importance. The democratic norm of political equality does not imply that everyone’s demands must be met all the time, but it does mean that in principle, everyone must be able to participate in the political process to make her or his interests heard. In the words of Pitkin, ‘in a representative government the governed must be capable of action and © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License INEQUALITY, EFFICACY AND POLITICAL TRUST 5 judgment, capable of initiating government activity, so that the government may be conceived as responding to them’ (Pitkin, 1967, p. 232). The degree to which political systems operate in accord with this requirement is captured by the two interrelated concepts of responsiveness and efficacy. Responsiveness captures how well political actors take citizen sentiment into account and adapt policy decisions accordingly, with responsiveness perceptions referring to the concomitant subjective beliefs (Esaiasson et al., 2015). Political efficacy in turn is capturing citizens’ beliefs that (their) individual action can influence the political process (Campbell et al., 1954, p. 187). Whereas internal efficacy refers to the belief in one’s own capacity to understand and participate in politics, external efficacy is the belief in the potential of ordinary citizens to influence the political process, ‘with an emphasis more on the ultimate fairness of both procedures and outcomes’ (Craig et al., 1990, p. 299). Furthermore, efficacy and responsiveness are intricately linked to inequality. Research on external efficacy and responsiveness perceptions points to the fact that some groups are regularly underrepresented in political debates and outcomes (Goodin & Dryzek, 1980; Goubin, 2020; Norris, 2015). Indeed, the unequal responsiveness literature has shown for different contexts that the higher ranks of society find their policy preferences more often mirrored by political decisions than their fellow citizens from modest backgrounds (Elsässer et al., 2021; Gilens, 2012; Schakel, 2021; Oser et al., 2023). One way to explain these findings is that those with more economic resources are able to promote policies in their interest and inhibit those policy proposals that would primarily benefit the middle and lower classes (Hacker & Pierson, 2010). Another is that political participation requires resources and is therefore skewed towards the better-off (Brady et al., 1995; Schäfer & Schwander, 2019). Parties have therefore incentives to design policies that cater to the demands of those segments of society that turn out to vote (Schakel & Burgoon, 2022), and unequal responsiveness then is the result of inequality in economic resources translating into inequality in political power (Dahl 2006; Goodin & Dryzek, 1980). Against this background, it should come as no surprise that rising economic inequality has long been considered to exacerbate the problem of unequal responsiveness. Seminal in this regard was Schattschneider (1960), who first proposed that growing inequality changes the agenda of political conflict and thereby increases the bias in electoral participation, which in turn affects whose interests are represented in political outcomes. Political participation is indeed lower and more skewed in more unequal countries (Schäfer & Schwander, 2019), and other studies have shown that this problem became more severe due to increasing income inequality (Lancee & van de Werfhorst, 2012; Solt, 2008). Also, when rising economic inequality increases the relative power of those with more resources, the repeated experience that certain groups in society get their way more often than others is likely to invoke the perception of a political elite that is disconnected from ordinary citizens, and of a system that is ‘rigged’. That inequality may lead to negative assessments of democratic responsiveness as well as to the diffusion of feelings of powerlessness and cynicism is reflected in empirical research that suggests that both objective economic inequality and subjective perceptions of inequality depress citizens’ perceptions of external efficacy (Loveless 2013; Norris 2015). Considering these findings as well as the long-standing theoretical arguments, it seems plausible to assume that there is a negative effect of income inequality on external efficacy and that declines in efficacy then translate into lower levels of trust. Nevertheless, Goubin (2020) is, to the best of our knowledge, the only study that empirically examined the whole chain of inequality, efficacy and trust, but her research question and design differ from ours. More specifically, Goubin © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 6SIMON BIENSTMAN, SVENJA HENSE, & MARKUS GANGL (2020) shows that external efficacy is positively related to trust in a cross section of countries, but she examines the role of economic inequality as a moderator for the relationship between efficacy and trust and documents that the positive effect of perceived responsiveness on citizens’ trust is smaller in countries with higher levels of income inequality. The account we seek to advance here, instead, views economic inequality as a macrostructural cause of citizens’ trust in institutions and sees the relationship between inequality and trust to be partly mediated by citizens’ efficacy perceptions. Our theoretical argument is summarized graphically in the path model of Figure 1 or, alternatively, in the following hypotheses: H1: When income inequality increases over time, political trust declines (total effect =direct effect c +indirect effect ab). H2: When income inequality increases over time, external efficacy declines (path a). H3: External efficacy has a positive effect on citizens’ political trust (path b). H4: By implication of H2 and H3, external efficacy partly mediates the negative effect of income inequality on political trust (path ab, indirect effect). Figure 1. Multilevel path model of political trust. Notably, our emphasis on the relevance of an efficacy-based mechanism does not contradict the trust-as-evaluation perspective. Citizens surely have views on desirable macroeconomic conditions on the one hand, but it is likewise plausible that their evaluations of democratic realities depend on the perceived fairness of the democratic process and on whether, or how strongly, the norm and expectation of political equality are being violated. Existing studies clearly relate political trust to perceptions of responsiveness and efficacy (Goubin, 2020; Torcal, 2014), and several scholars have argued that process evaluations may be more important for political support than outcome evaluations in the longer run (Norris, 2011; van der Meer & Hakhverdian, 2017). The difference between output and process evaluation might be important for theorizing the role of political ideology. The standard trust-as-evaluation perspective implies the expectation © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License INEQUALITY, EFFICACY AND POLITICAL TRUST 7 that political ideology will moderate the relationship between inequality and trust in democratic institutions. When political trust is a function of citizens’ evaluations of the outputs of the political system against their expectations (van der Meer, 2018), left-leaning citizens should be more negatively affected by high levels of inequality because they value equality as a political goal (cf. Alesina et al., 2004). In line with this reasoning, Anderson and Singer (2008) indeed find that the effect is much more pronounced among left-wingers. However, the same partisan stratification is not likely to be relevant when the standard of evaluation is one of democratic process rather than democratic output. When manifest inequality leads to unequal responsiveness and depresses citizens’ perceptions of external efficacy, this may affect leftand right-leaning citizens alike, irrespective of whether they consider questions of income inequality as an important political goal. In line with this view, Norris (2015) finds that inequality is a strong depressor of external efficacy in the general population, not just among left-leaning citizens. Accordingly, we expect a difference between processand performance-based evaluations with respect to the moderating role of political ideology (also see Figure 1): H5: While political ideology moderates the direct effect between inequality and trust, the indirect effect via efficacy perceptions is non-partisan and therefore holds uniformly in the citizenry. Data and methodology To test these hypotheses, we use individual-level survey data from rounds 7 to 9 of the ESS (2014, 2016, 2018) to which we merge macroeconomic data from different sources (see below). The ESS has been fielded every other year with newly selected probability samples of the population in participating countries since 2002. By combining the three rounds of the ESS that contain questions on efficacy, we obtain a repeated cross-sectional dataset for those 22 countries that participated in the ESS at least twice in the period from 2014 to 2018 (see Table A1 in the online Appendix). In conjunction with a model that incorporates the two-way FE estimator at the aggregate level of the data, this allows us to conduct a longitudinal analysis in a panel-of-countries design on the question of how changes in inequality have been affecting changes in political trust in European democracies during the second half of the 2010s. Operationalization Dependent variable. The ESS asks respondents to indicate how much trust they have in several of the institutions of democratic governance and measures each item on an 11-point scale (0 =no trust at all, 10 =complete trust). We construct a political trust index for each respondent based on the sum of their scores for trust in parliament, trust in politicians, and trust in parties, divided by the number of valid responses (Cronbach’s α=0.91). We also perform robustness checks with each item separately. Independent variables. External efficacy is measured by two questions that ask respondents ‘How much would you say the political system in [country] allows people like you to have a say in what the government does?’ and ‘And how much would you say that the political system in [country] allows people like you to have an influence on politics?’ These are very similar to the items Craig © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8SIMON BIENSTMAN, SVENJA HENSE, & MARKUS GANGL et al. (1990) have shown to be valid and distinct measures of external efficacy. In 2014, answers were recorded on an 11-point scale. From 2016 onward, the response categories include 1 =‘Not at all’, 2 =‘Very little’, 3 =‘Some’, 4 =‘A lot’, 5 =‘A great deal’. We harmonize these variables by normalizing them to range from 0 to 10. The respondents’ average score on the two questions provides us with a continuous measure of their level of perceived external efficacy, also ranging from 0 to 10. As for income inequality, we use the S80/S20 income share ratio of household disposable income from Eurostat (2022a). The measure reflects the ratio between the share of total income that is accruing to the top 20 per cent of the income distribution and the income share of the bottom 20 per cent. As a robustness test, however, we also replicate our main analyses with the Gini coefficient of household disposable income as obtained from the Standardized World Inequality Database (Solt, 2020). For the moderator variable of political ideology, we follow Anderson and Singer (2008) and dichotomize the 11-point left-right scale to distinguish left-leaning respondents (0–2) from centre/right-leaning respondents (3–10).1 Control variables As individual-level covariates, we include educational level, household income, social class as well as gender and age. We use a five-category measure distinguishing between respondents who completed less than secondary education, lower secondary, upper secondary, post-secondary non-tertiary education, and tertiary education. We also include a relative measure of household income (country-specific income quintiles +‘Missing’). Social class is measured with the fiveclass scheme by Oesch (2006), distinguishing unskilled workers, skilled workers, small business owners, lower service class and upper service class. We also account for the macro-economic development of the sampled countries by including the economic performance index (EPI; see Khramov & Lee, 2013). The EPI summarizes information on economic growth, inflation, unemployment (WDI, 2022) and the budget deficit (IMF, 2022). Further information on the EPI, as well as a list of survey items used in the analysis can be obtained from the online Supporting Information. Descriptive statistics for all variables are included in Table A2 in the online Appendix. Analytical strategy The pooled ESS data consist of three hierarchical levels, with individual respondents, nested in country years and countries. We account for this setup by estimating a series of FE regression models that permit us to evaluate the microand macro-level correlates of citizens’ political trust while providing adequately clustered standard errors for hypothesis testing. Employing countryand survey-wave FE incorporates the feature of a static panel data model at the aggregate level of countries. Country FEs serve the twin purpose of properly isolating the longitudinal (within-unit) from the cross-sectional (between-unit) effects of any macro-level covariate while simultaneously controlling for any observed or unobserved confounders that are time constant at the country level (see Gangl, 2010; Greene, 2018; Woolridge, 2010). Country FEs afford this by removing all between-country variation from the analysis. As a result, the country FEs control for any © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License INEQUALITY, EFFICACY AND POLITICAL TRUST 9 time-invariant and possibly idiosyncratic country characteristics that create whatever is temporally persistent in a country’s democratic culture throughout our study. In addition, adopting a FE specification implies that longitudinal, within-unit variation over time is the sole informational basis for estimating a regression coefficient to describe the association between a covariate and the outcome variable of interest. FE models effectively demean all the variables in the regression model, thus providing estimates of the correlation between (over time) within-unit changes in the dependent variable and (over-time) within-unit changes in the independent variable. An FE model, therefore, is the equivalent of adopting a longitudinal research design that allows us to estimate the effect of observable changes in inequality on changes in political trust in a before-and-after comparison within units, that is, at the aggregate level of countries in our case. In the context of a multilevel model for repeated cross-sectional data, where controls for individual-level covariates also amount to an implicit control for compositional changes in the citizenry, we in fact specifically estimate the contextual (or sociotropic) effect of income inequality on political trust, net of citizens’ SES or other demographic characteristics and any egocentric political motivations that may come with them. Within this broad approach, our analytical strategy then is twofold: We first show a series of regression models to evaluate each hypothesized path of our theoretical model. In a second step, we test the mediation channel via efficacy perceptions by applying a causal mediation analysis that uses the mediate package in R(Imai et al., 2010; Tingley et al., 2014). Specifically, we simulate the model parameters (the direct effect of inequality on trust and the indirect effect via efficacy perceptions) from the regression models for the outcome and mediator based on their sampling distribution. In total, we obtain a sample of 1000 of these parameters. For each of these draws and each unit in the dataset, we simulate the potential mediator values and the potential outcomes (based on the simulated mediator values) and calculate the average causal mediation effect across the predictions for all respondents. This estimate is one of 1000 Monte Carlo draws so that we obtain a distribution of 1000 mediation effects from which we then calculate point estimates and confidence intervals (the mean and percentiles of that distribution) of the mediation effect (see Imai et al., 2010, p. 328, for details on the algorithm). With this setup, our estimates become a practically useful and informative approximation of the likely range of the causal effect of income inequality on external efficacy and trust. Our two-way FE design and further controls should help isolate our inferences against many obvious confounders, even as some inferential threats inevitably remain. The two most important assumptions we need to retain to give our estimates a causal interpretation are that there are no omitted time-varying confounders and that reverse causation is absent. Concerning the former, concomitant political events would seem the primary candidate factor. We thus provide a range of additional analyses to explore the robustness of our findings when considering other time-varying covariates or when conducting jackknife estimation to test against the presence of influential single-country cases. For reverse causality, we must acknowledge the principal possibility that changes in trust may (also) precede changes in inequality, for example, when high levels of trust enable democratic governments to implement far-reaching political interventions to combat inequality. While theoretically possible, reverse causation seems not likely to be a dominant factor in our study. The short (4-year) observation window of our data should render it much more likely that we see associations because citizens react to changes in inequality (which they can do without much delay) rather than the other way. That principle optimism notwithstanding, we do not wish to claim a similar level of confidence about strictly causal conclusions from our mediation © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 16 SIMON BIENSTMAN, SVENJA HENSE, & MARKUS GANGL of subjective perceptions over objective inequality (Weisstanner & Armingeon, 2022). By contrast, our results show that neither a correct perception of the degree of economic inequality nor a political preference for economic equality are strictly necessary for objective inequality to affect democratic attitudes. As objective economic inequality affects external efficacy, it indirectly affects political trust even among those for whom inequality is, for ideological reasons, not a major concern. While we believe these contributions to be important, our study faces certain limitations. First, unlike for the effect of inequality on trust, we cannot draw strictly causal conclusions from the mediation analysis as data limitations prevent us from employing a full set of candidate mediators. Relatedly, contrary to our assumptions, trust may precede efficacy. For example, when a change in government happens and the party a respondent voted for takes the lead after having been in opposition before, citizens may adapt their feelings of external efficacy to their increased trust in political institutions. We consider this causal direction less likely, yet we cannot rule it out entirely. Finally, our data consist of a limited number of European countries and a short period of time. Levels of income inequality in Europe are comparatively low and the more drastic changes in inequality occurred prior to the period under investigation. Yet, even in this setting, which can be considered a least-likely design, our findings are in line with our theoretical propositions and highly robust. Nevertheless, future studies should test whether the proposed mechanism holds in different contexts and periods. Despite these limitations, we believe that this paper makes an important contribution by providing new evidence showing that objective economic inequality has the potential to erode democratic support and that external efficacy plays a key role therein. This is practically relevant, as the impact of rising income inequality may be cushioned through interventions that raise perceptions of efficacy. Recognizing this efficacy-based transmission channel can also contribute to political sociology more generally. Placing a perspective on citizens’ evaluation of political power and fair democratic processes alongside the perspective of output-based evaluations may enrich the argument that democratic stability may be undermined when economic inequality is getting translated into political power and inequality of representation. Acknowledgements Earlier versions of this research have been presented at the AS Conference ‘Cohesive Societies?’ in 2021 and the ‘Workshop on Perceptions and Policy Preferences’ at the University of Hamburg in 2021. We thank all participants, Carlotta Giustozzi, Claudia Traini, and the three anonymous reviewers for their helpful comments. We also thank Elisa Ebertz, Clara Eul and Emir Zecovic for valuable research assistance in the project. Open access funding enabled and organized by Projekt DEAL. Conflict of interest statement The authors declare no conflict of interest. Data availability statement The data and syntax are available on OSF (DOI 10.17605/OSF.IO/F4NUA). © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License INEQUALITY, EFFICACY AND POLITICAL TRUST 17 Funding Information This study is part of the project POLAR: Polarization and its discontents: does rising economic inequality undermine the foundations of liberal societies? that has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Programme (Grant agreement No. 833196). Ethics approval statement Ethical approval is not applicable to this article. Online Appendix Additional supporting information may be found in the Online Appendix section at the end of the article: Supplementary information Note 1. In a robustness check, we confirm that our substantive results are being replicated when varying the cut–off point to compare left-leaning and centre-left citizens (0–4) to center/right-leaning respondents (5–10). References Alesina, A., Di Tella, R., & MacCulloch, R. (2004). Inequality and happiness: Are Europeans and Americans different? Journal of Public Economics,88, 2009–2042. Andersen, R. (2012). Support for democracy in cross-national perspective: The detrimental effect of economic inequality. Research in Social Stratification and Mobility,30(4), 389–402. Anderson, C. J., & Singer, M. M. (2008). The sensitive left and the impervious right. Multilevel models and the politics of inequality, ideology, and legitimacy in Europe. Comparative Political Studies,41(4), 564–599. Atkinson, A. B. (2015). Inequality: What can be done? Harvard University Press. Brady, H. E., Verba, S., & Schlozman, K. L. (1995). Beyond SES: A resource model of political participation. American Political Science Review,89(2), 271–294. Buttrick, N. R., & Oishi, S. (2017). The psychological consequences of income inequality. Social and Personality Psychology Compass,11(3), e12304. Campbell, A., Gurin, G., & Miller, W. E. (1954). The voter decides. Row, Peterson, and Co. Christmann, P. (2018). Economic performance, quality of democracy and satisfaction with democracy. Electoral Studies,53, 79–89. Coppedge, M., Gerring, J., Knutsen, C. H., Lindberg, S. I., Teorell, J., Altman, D., Bernhard, M., Cornell, A., Fish, M. S., Gastaldi, L., Gjerløw, H., Glynn, A., Grahn, S., Hicken, A., Kinzelbach, K., Marquardt, K. L., McMann, K., Mechkova, V., Neundorf, A., Paxton, P., Pemstein, D., Rydén, O., vonRömer, J., Seim, B., Sigman, R., Skaaning, S.–E., Staton, J., Sundström, A., Tzelgov, E., Uberti, L., Wang, Y., Wig, T., & Ziblatt, D. (2022). V-dem country-year dataset v12. Varieties of Democracy (V-Dem) Project. https://www.v-dem.net/data/the-v-dem-dataset/ Craig, S. C., Niemi, R. G., & Silver, G. E. (1990). Political efficacy and trust: A report on the NES pilot study items. Political Behavior,12(3), 289–314. Cunningham, S. (2021). Causal inference: The mixtape. Yale University Press. Dahl, R. A. (1971). Polyarchy. Participation and opposition. Yale University Press. © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 18 SIMON BIENSTMAN, SVENJA HENSE, & MARKUS GANGL Dahl, R. A. (2006). On political equality. Yale University Press. Dalton, R. J. (1998). Political support in advanced industrial democracies. UC Irvine: Center for the Study of Democracy. Retrieved from https://escholarship.org/uc/item/8281d6wt Delhey, J., & Steckermeier, L. C. (2020). Social ills in rich countries: New evidence on levels, causes, and mediators. Social Indicators Research,149(1), 87–125. Donovan, T., & Karp, J. (2017). Electoral rules, corruption, inequality and evaluations of democracy. European Journal of Political Research,56(3), 469–486. Easton, D. (1975) A re-assessment of the concept of political support. British Journal of Political Science,5(4), 435–457. Elsässer, L., Hense, S., & Schäfer, A. (2021). Not just money: Unequal responsiveness in egalitarian democracies. Journal of European Public Policy,28(12), 1890–1908. Engler, S., & Weisstanner, D. (2021). The threat of social decline: Income inequality and radical right support. Journal of European Public Policy,28(2), 153–173. Esaiasson, P., Kölln, A.-K., & Turper, S. (2015). External efficacy and perceived responsiveness-similar but distinct concepts. International Journal of Public Opinion Research,27(3), 432–445. ESS Round 7: European Social Survey Round 7 Data (2014). Data file edition 2.2. Sikt - Norwegian Agency for Shared Services in Education and Research, Norway – Data Archive and distributor of ESS data for ESS ERIC. doi: 10.21338/NSD-ESS7-2014. ESS Round 8: European Social Survey Round 8 Data (2016). Data file edition 2.2. Sikt - Norwegian Agency for Shared Services in Education and Research, Norway – Data Archive and distributor of ESS data for ESS ERIC. doi: 10.21338/NSD-ESS8-2016. ESS Round 9: European Social Survey Round 9 Data (2018). Data file edition 3.1. Sikt - Norwegian Agency for Shared Services in Education and Research, Norway – Data Archive and distributor of ESS data for ESS ERIC. doi: 10.21338/NSD-ESS9-2018. Eurostat (2022a). Inequality of income distribution S80/S20 income quintile share ratio - EU-SILC and ECHP surveys.https://ec.europa.eu/eurostat/web/products-datasets/-/ilc_pns4 Eurostat (2022b). Distribution of income by quantiles—EU-SILC and ECHP surveys [ilc_di01]. https://ec.europa.eu/eurostat/web/products-datasets/-/ilc_di01 Fairbrother, M. (2014). Two multilevel modeling techniques for analyzing comparative longitudinal survey datasets. Political Science Research and Methods,2(1), 119–140. Fairchild, A. J., & MacKinnon, D. P. (2009). A general model for testing mediation and moderation effects. Prevention Science,10, 87–99. Foa, R. S., Klassen, A., Slade, M., Rand, A., & Collins, R. (2020). The global satisfaction with democracy report 2020. Centre for the Future of Democracy. Gangl, M. (2010). Causal inference in sociological research. Annual Review of Sociology,36(1), 21–47. Gangl, M., & Giustozzi, C. (2021). Unemployment and political trust across 24 Western democracies: Evidence on a welfare state paradox. Acta Sociologica,64(3), 255–273. Gidron, N., & Hall, P. A. (2017). The politics of social status: Economic and cultural roots of the populist right. The British Journal of Sociology,68, S57-S84. Gilens, M. (2012). Affluence and influence: Economic inequality and political power in America. Princeton University Press. Goodin, R., & Dryzek, J. (1980). Rational participation: The politics of relative power. British Journal of Political Science,10, 273–292. Goubin, S. (2020). Economic inequality, perceived responsiveness and political trust. Acta Politica,55, 267–304. Goubin, S., & Hooghe, M. (2020). The effect of inequality on the relation between socioeconomic stratification and political trust in Europe. Social Justice Research,33(2), 219–247. Greene, W. H. (2018). Econometric analysis (8th ed). Pearson. Hacker, J. S., & Pierson, P. (2010). Winner-take-all politics: Public policy, political organization, and the precipitous rise of top incomes in the United States. Politics and Society,38(2), 152–204. Hetherington, M. J. (1998). The political relevance of political trust. The American Political Science Review,92(4), 791–808. © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License INEQUALITY, EFFICACY AND POLITICAL TRUST 19 Hooghe, M., & Okolikj, M. (2020). The long–term effects of the economic crisis on political trust in Europe: Is there a negativity bias in the relation between economic performance and political support? Comparative European Politics,18, 879–898. Imai, K., Keele, L., & Tingley, D. (2010). A general approach to causal mediation analysis. Psychological Methods, 15(4), 309–334. IMF (2022). World economic outlook database.https://www.imf.org/en/Publications/WEO/weo-database/2022/April. Khramov, V., & Lee, J. R. (2013). The Economic Performance Index (EPI): An Intuitive Indicator for Assessing a Country’s Economic Performance Dynamics in an Historical Perspective (IMF Working Paper 13/214). IMF. Kim, Y., Sommet, N., Na, J., & Spini, D. (2022). Social class—not income inequality—predicts social and institutional trust. Social Psychological and Personality Science,13(1), 186–198. Kline, R. B. (2015). The mediation myth. Basic and Applied Social Psychology,37(44), 202–213. Krieckhaus, J., Son, B., Bellinger, N. M., & Wells, J. M. (2014). Economic inequality and democratic support. The Journal of Politics,76(1), 139–151. Kumlin, S., Stadelmann-Steffen, I., & Haugsgjerd, A. (2018). Trust and the welfare state. In E. M. Uslaner (Ed.), The Oxford handbook of social and political trust (pp 385–408). Oxford University Press. Lancee, B., & Van de Werfhorst, H. G. (2012). Income inequality and participation: A comparison of 24 European countries. Social Science Research,41(5), 1166–1178. Lee, D., Chang, C. Y., & Hur, H. (2020). Economic performance, income inequality and political trust: New evidence from a cross-national study of 14 Asian countries. Asia Pacific Journal of Public Administration,42(2), 66–88. Lipset, S. M. (1963). Political man. The social bases of politics. Doubleday. Loveless, M. (2013). The deterioration of democratic political culture: Consequences of the perception of inequality. Social Justice Research,26(4), 471–491. Magalhães, P. C. (2014). Government effectiveness and support for democracy. European Journal of Political Research,53(1), 77–97. Martini, S., & Quaranta, M. (2020). Citizens and democracy in Europe: Contexts, changes and political support. Springer International Publishing. McAllister, I. (1999). The economic performance of governments. In P. Norris (Ed.), Critical citizens: Global support for democratic governance (pp. 188–203). Oxford University Press. Mounk, Y. (2018). The people vs. democracy: Why our freedom is in danger and how to save it. Harvard University Press. Newton, K., Stolle, D., & Zmerli, S. (2018). Social and political trust. In E. M. Uslaner (Ed.), The Oxford handbook of social and political trust (pp. 38–56). Oxford University Press. Norris, M. (2015). The economic roots of external efficacy: Assessing the relationship between external political efficacy and income inequality. Canadian Journal of Political Science/Revue Canadienne de Science Politique, 48(4), 791–813. Norris, P. (1999). Introduction: The growth of critical citizens? In P. Norris (Ed.), Critical citizens: Global support for democratic government (pp. 1–27). Oxford University Press. Norris, P. (2011). Democratic deficit: Critical citizens revisited. Cambridge University Press. OECD (2011). Divided we stand: Why inequality keeps rising. OECD Publishing. Oesch, D. (2006). Redrawing the class map. Stratification and institutions in Britain, Germany, Sweden and Switzerland. Palgrave Macmillan. Oser, J., Feitosa, F., & Dassonneville, R. (2023). Who Feels They Can Understand and Have an Impact on Political Processes? Socio–demographic Correlates of Political Efficacy in 46 Countries, 1996–2016. International Journal of Public Opinion Research,35(2). Paskov, M., Gërxhani, K., & van de Werfhorst, H. G. (2017). Giving up on the Joneses? The relationship between income inequality and status-seeking. European Sociological Review,33(1), 112–123. Piketty, T. (2017). Capital in the twenty-first century. The Belknap Press of Harvard University Press. Pitkin, H. F. (1967). The concept of representation. University of California Press. Polacko, M. (2022). The Politics of Income Inequality: Redistribution, Turnout and Responsiveness. Statistics, Politics and Policy,13(2), 201–218. Runciman, D. (2018). How democracy ends. Basic Books. Schäfer, A. (2013). Affluence, inequality and satisfaction with democracy. In S. Keil & O. W. Gabriel (Ed.), Society and democracy in Europe (pp. 139–161). Routledge. © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 20 SIMON BIENSTMAN, SVENJA HENSE, & MARKUS GANGL Schäfer, A., & Schwander, H. (2019). Don’t play if you can’t win’: Does economic inequality undermine political equality? European Political Science Review,11(3), 395–413. Schakel, W. (2021). Unequal policy responsiveness in the Netherlands, Socioeconomic Review,19(1), 37–57. Schakel, W., & Burgoon, B. (2022). The party road to representation: Unequal responsiveness in party platforms. European Journal of Political Research,61(2), 304–325. Schattschneider, E. E. (1960). The semisovereign people. A realist’s view of democracy in America. Holt, Rinehart and Winston. Solt, F. (2008). Economic inequality and democratic political engagement. American Journal of Political Science, 52(1), 48–60. Solt, F. (2020). Measuring income inequality across countries and over time: The standardized world income inequality database. Social Science Quarterly,101(3), 1183–1199. Stoetzer, L. F., Giesecke, J., and Klüver, H. (2021). How does income inequality affect the support for populist parties? Journal of European Public Policy,30(1), 1–20. Tingley, D., Yamamoto, T., Hirose, K., Keele, L., & Imai, K. (2014). mediation: R Package for Causal Mediation Analysis. Journal of Statistical Software,59(5). Torcal, M. (2014). The decline of political trust in Spain and Portugal: economic performance or political responsiveness? American Behavioral Scientist,58(12), 1542–1567. van Beek, U., Fuchs, D., & Klingemann, H.-D. (2019). The question of legitimacy in contemporary democracies. In U. van Beek (Ed.), Democracy under threat: A crisis of legitimacy? (pp. 321–335), Palgrave Macmillan. van der Meer, T. (2010). In what we trust? A multi-level study into trust in parliament as an evaluation of state characteristics. International Review of Administrative Sciences,76(3), 517–536. van der Meer, T. (2017). Democratic input, macroeconomic output and political trust. In S. Zmerli & T. Van der Meer (Eds.), Handbook on political trust (pp. 271–284). Edward Elgar Publishing. van der Meer, T. (2018). Economic performance and political trust, In E. Uslaner (Ed.), The Oxford handbook of social and political trust (pp. 599–616). Oxford University Press. van der Meer, T., & Hakhverdian, A. (2017). Political trust as the evaluation of process and performance: A crossnational study of forty-two European democracies. Political Studies,65(1), 81–102. Van de Werfhorst, H. G., & Salverda, W. (2012). Consequences of economic inequality: Introduction to a special issue. Research in Social Stratification and Mobility,30(4), 377–387. Weisstanner, D., & Armingeon, K. (2022). Redistributive preferences: Why actual income is ultimately more important than perceived income. Journal of European Social Policy,32(2), 135–147. Wilkinson, R. G., & Pickett, K. (2010). The spirit level: Why equality is better for everyone. Penguin Books. Woolridge, J. M. (2010). Econometric analysis of cross section and panel data (2nd ed.). MIT Press. WDI (2022). World development indicators. http://databank.worldbank.org/ Ziblatt, D., & Levitsky, S. (2018). How democracies die. Crown. Zmerli, S., & Castillo, J. C. (2015). Income inequality, distributive fairness and political trust in Latin America. Social Science Research,52, 179–192. Zmerli, S., & Newton, K. (2008). Social trust and attitudes toward democracy. Public Opinion Quarterly,72(4), 706–724. Address for Correspondence: Institute for Sociology, Goethe-University Frankfurt, Theodor-W.-Adorno-Platz 6, 60323 Frankfurt am Main, Germany. Email: [email protected] © 2023 The Authors. European Journal of Political Research published by John Wiley & Sons Ltd on behalf of European Consortium for Political Research. 14756765, 0, Downloaded from https://ejpr.onlinelibrary.wiley.com/doi/10.1111/1475-6765.12611 by Cochrane Germany, Wiley Online Library on [03/07/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License