Abstract
This article deals with the problem of the increasing distrust in institutions in European democracies, paying special attention to countries such as Spain and Portugal, which are being severely harmed by the current economic crisis and the austerity measures imposed by supranational organizations such as the European Union and implemented by their respective national governments. Contrary to the purely political-economic paradigm, the present article, based on panel survey data gathered during the years 2011 and 2012 in Spain and Portugal, shows that this deterioration is due principally to the negative perception of the political responsiveness of representative institutions and aggravated by the increasing perception of political corruption. Multilevel governance and the present economic crisis are challenging representative institutions, but their functioning and elite misbehavior during the crisis are the main explanations of increasing citizen distrust in such institutions.
There is a growing perception that political distrust of representative institutions is increasing in contemporary democracies (Klingemann, 1999; Lipset & Schneider, 1983; Norris, 1999a; Nye & Zelikow, 1997; Pharr & Putnam, 2000; Torcal & Montero, 2006). There are two arguments that justify this view. The first is that increasing distrust of national parliaments, political representatives, and political parties is due to the fact that these institutions are perceived as unresponsive to citizens’ demands (Alesina & Wacziarg, 2000). This argument has been especially salient in recent years, because, as has been stated, these institutions appear to be responding more to the demands of supranational or international institutions such as the International Monetary Fund (IMF), the World Bank (WB), or the European Union (EU) than to citizens’ needs (Armingeon & Guthmann, 2013). The second argument is more a purely economic instrumental premise and holds that this attitudinal trend is the result of the deterioration of economic conditions, the dismantling of the welfare state, and the general crisis of the social capitalism model (Offe, 2003). Recently, this viewpoint has appeared to be particularly relevant in many of the European countries severely affected by the current economic-fiscal crisis and the implementation of drastic austerity measures to resolve it (Polavieja, 2013). Thus, does this increasing institutional distrust affect all European democracies or only those for which the economic crisis and the austerity measures have been harsher? Is this increasing institutional distrust due simply to the effect of the economic situation and its consequences? Or, alternatively, is it the consequence of the perceived unresponsiveness of national political institutions to citizens’ problems?
In the present article, and contrary to the literature attributing this recent decline in political trust to the purely economic effects of the current economic crisis and its social consequences (Polavieja, 2013), we shall argue that this increasing institutional distrust is largely due to citizens’ perceptions that political system is not responsive to their demands. Individual economic conditions or personal economic expectations certainly appear to have some effect, but they are not the most relevant explanation of distrust in representative institutions in Europe. As Schlozman and Verba (1979) stated in their classic work, it could be that severe economic conditions and job loss have little direct effect on political life since political attitudes are more a function of general social beliefs than of personal experiences with the economic or social situations. This is why this process does not equally affect all the countries where the economic crisis and its consequent austerity measures appear to be harsher or is not limited exclusively to those countries. The current economic crisis and the increasing scarcity of economic and social resources in European societies are putting democratic institutions under pressure and at the center of scrutiny by citizens. This, in some cases, results in a negative evaluation of their responsiveness and consequently a decrease in the general levels of citizens’ institutional trust.
The empirical analysis in the following pages has theoretical implications that transcend the economic and social consequences of the current crisis for institutional trust in countries expiriencing economic crises. It deals with the factors explaining short-term change in citizens’ attitudes. Thus, the article is not intended to dispute the importance of long-term cultural factors in predicting institutional trust. It is aimed at explaining the relative importance of short-term factors in changing political attitudes in a given cultural and institutional context (Kenworthy & Owens, 2011). Consequently, the intention is to deal with the relative importance of two sets of short-term factors in altering institutional trust. One is the purely political-economic perspective, which holds that this change has been produced by economic conditions and the evaluation of their social and economic consequences (Chanley, Rudolph, & Rahn, 2001; Clarke, Dutt, & Kornberg, 1993; Clarke & Kornberg, 1989; McAllister, 1999). The other concentrates more on the effect on institutional trust of the lack of political responsiveness to citizens’ demands during critical economic and political conditions (Armingeon & Guthmann, 2013; Denters, Gabriel, & Torcal, 2007; Offe, 2006; van der Meer, 2010). The former is an exogenous explanation of institutional trust, whereas the latter is an endogenous justification.
Although there exists a comparative preliminary analysis of the results of the fourth and sixth waves of the European Social Survey (ESS) collected in 2008 and 2012, the present study of institutional trust is based largely on a longitudinal survey collected from a panel online sample of the population in Spain and Portugal during 2011 and 2012. Unable to sustain its public finances, the government of Portugal requested from the IMF and the EU a full bailout package in 2010 and 2011. Spain also requested a special Eurozone bailout to finance its problematic private banking system in 2012. The governments of both countries were therefore impelled to adopt severe austerity measures during those years, producing a decline in the standard of living for many of their citizens. These two countries are also among those with the highest increase in institutional distrust in Europe since 2008. To what extent are these processes linked in these two countries? Survey panel data collected in these two countries during the years 2011 and 2012 permit the appropriate testing of the causal mechanisms behind the deterioration of institutional trust in Spain and Portugal, the presence of cultural and institutional factors remaining constant.
The Theoretical Argument
Trust in institutions is a declaration by citizens that institutions are reliable (Hardin, 1998, 2000; Levi, 1998; Levi & Stoker, 2000). 1 Yet, what should we examine to assess the general levels of political trust? Due to the original operationalization of the concept, 2 political trust has been repeatedly associated with trust in the government, frequently making political trust and trust in government interchangeable concepts (Hibbing & Theiss-Morse, 1995, 2001; Kanji & Nevitte, 2002). This conceptualization and operationalization of political trust is responsible for the traditional confusion between satisfaction with incumbents (political satisfaction) and political trust in the system. 3 This is why this article is simply focused on citizens’ trust in the basic institutions of political representation in contemporary democracies, leaving aside attitudes related to the evaluation of the functioning or performing of the system or support for the government. 4
However, a low level of trust in one institution usually goes together with low levels of trust in other institutions (Denters et al., 2007; Hetherington, 1998); in other words, in most surveys, trust in different institutions is clearly a one-dimensional phenomenon (Marien, 2011). Despite this preceding strong empirical evidence, trust in distinct institutions does not display identical levels in all countries. As Denters et al. (2007) have shown, this depends on the type of institution: trust in political actors (political parties and politicians), trust in the institutions of liberal democracy (parliament and the cabinet), and trust in the courts and the police, which represent the institutions of the Rechtsstaat. These last ones present the highest levels of trust, political actors being those for whom citizens consistently display the lowest levels of trust. In addition, factors explaining individual levels of institutional trust also vary somewhat depending on these three types of institutions (Denters et al., 2007). So, has the current economic and political crisis had the same effect on individual trust for all of these institutions?
Basic attitudes, such as institutional trust, are no longer seen as immutable after early socialization is concluded, as the traditional culturalist model originally argued (Almond & Verba, 1963; Citrin, 1975; Easton, 1965, 1975; Lerner, 1958; Uslaner, 2002). The rational-culturalist paradigm, which defends the endogeneity and greater volatility of political attitudes, currently appears to be the dominant paradigm (Lane, 1992; Letki & Evans, 2005; Mishler & Rose, 2001; Muller & Seligson, 1994; Pateman, 1971; Torcal & Magalhaes, 2010; Whitefield & Evans, 1999). Although the pace of attitudinal change is still typically considered to be slow (Bermeo, 2003; Clarke et al., 1993; Kornberg & Clarke, 1992; Newton, 2007; Torcal, 2006), short-term variations in institutional trust are supported by empirical evidence (Kenworthy & Owens, 2011; Zmerli & Newton, 2011). Consequently, institutional trust could change rapidly in reaction to the perceived performance of political institutions (Offe, 2006), but which areas of their performance most affects citizen trust or distrust of institutions?
There exist two different positions on this subject. The first argues that institutional trust changes according to purely instrumental economic and individual evaluations of economic and social conditions and, thus, ultimately depends on the institutional capacity to meet and represent citizens’ purely economic needs and demands, which are mostly rooted in socioeconomic interest. Levels of trust are widely dispersed across societies, precisely due to their social, economic, and political circumstances (Newton, 2007). Thus, economic stewardship is typically identified as a leading cause of trust: When citizens are dissatisfied with economic performance, distrust of government ensues, whereas the reverse effect is produced when economic prosperity abounds (Citrin & Green, 1986; Citrin & Luks, 2001; Feldman, 1983; Hetherington, 1998; Lawrence, 1997; Listhaug, 2006; Miller & Borrelli, 1991; Offe, 2006).
This effect could be based on two different logics or arguments. The first may be the result of citizens’ sociotropic considerations of the situation, such as their retrospective evaluation of the country’s economy (Listhaug & Wiberg, 1995; Miller & Listhaug, 1999; Newton & Norris, 2000). Following the same logic, some authors have also linked the present welfare state retrenchment with declining trends in institutional trust (Alesina & Wacziarg, 2000; Kumlin, 2011). But individual instrumental economic calculations could also be behind these general evaluations of output performance, affecting institutional trust. Individual experiences of economic strain, whether lasting or transitory, can reduce citizens’ degree of trust in institutions (Brooks & Manza, 2007; Chanley et al., 2001; Clarke et al., 1993; Clarke & Kornberg, 1989; McAllister, 1999). Based on the same logic, there is also important and increasing evidence that the experience of unemployment can also increase political distrust (Gallie, 1993; Polavieja, 2013). In addition, awareness of the depth of the economic crisis is likely to create uncertainty about individual economic futures, leading many to feel increasingly unsafe (Mughan & Lacy, 2002), and affecting institutional trust as a result. Thus, not only can sociotropic evaluations affect institutional trust, but such a decline in trust could also be the product of more instrumental egotropic evaluations of the personal consequences of the crisis. Both arguments will be considered in the discussion of the effect of the current economic crisis on institutional trust.
However, other theoretical developments and empirical research indicate that the origins of change and distinct cross-national levels of political confidence may not be as apolitical as the preceding social or economic explanations imply (Kenworthy & Owens, 2011; Schlozman & Verba, 1979). A number of different political explanations have been proposed (Norris, 1999a). Macropolitical explanations such as the institutional setting appear to be significant in explaining cross-national variations in institutional trust and could also help to explain certain variations in institutional trust over time (Anderson & Guillory, 1997; Dalton, 1999; Katzenstein, 2000; Listhaug & Wiberg, 1995; Morlino & Tarchi, 1996; Norris, 1999a, 1999b; Nye & Zelikow, 1997; van der Meer, 2010). However, here, we are more interested in analyzing micropolitical orientations and evaluations resulting from contemporary political experiences (Denters, 2002; Offe, 2006).
Building on earlier work (Crozier, Huntington, & Watanuki, 1975; King, 1975), the current literature has reconsidered the importance of various aspects of citizens’ political orientations (Fiorina, 1999; Norris, 1999a; Pharr & Putnam, 2000). According to Miller’s (1974) classic study, citizens’ lack of confidence (or trust) in the institutions of government may be the result of extended periods of frustrated expectations, resulting from inadequate institutional performance. According to this view, low levels of institutional trust are related less to the government’s objective achievements and more to the gap between actual performance and citizens’ expectations (e.g., Miller & Listhaug, 1999). This generates a widespread feeling that governments are failing to respond to public needs and expectations. Consequently, the perceived responsiveness of political officeholders may be a crucial determinant of citizens’ political trust and especially their trust in democratic actors and institutions (Lipset & Schneider, 1983). Thus, citizens’ negative feelings that institutions are not representative or responsive to the needs and desires of the members of the polity could increase individual institutional distrust (Bellucci, Sanders, Gábor, & Torcal, 2012).
Finally, these two preceding alternative hypotheses (economic instrumental calculations vs. political responsiveness) explaining institutional trust may be not only complementary but also mutually reinforcing. As Catterberg and Moreno (2005) have argued, the negative effect of the lack of political or governmental responsiveness on institutional trust may worsen in situations of economic crisis. This could be the case in many European countries, such as Portugal and Spain, where the severe economic crisis may be exacerbating the negative effects of the perceived lack of political responsiveness on political trust; put differently, increasing institutional distrust is the result not just of the direct effect of the economic crisis but also of the perceived lack of political responsiveness to the severe consequences produced by the current economic and social situation.
Therefore, in this article we intend to test the following five hypotheses:
Hypothesis 1: Citizens’ negative sociotropic evaluations of current economic conditions may be behind the increasing levels of institutional distrust in countries undergoing economic recession.
Hypothesis 2: Citizens’ negative egotropic assessment of their personal consequences of the crisis may be responsible for the increasing levels of institutional distrust.
Hypothesis 3: Citizens’ perceived lack of political responsiveness by political representatives could also be an important factor in explaining political distrust.
Hypothesis 4: The perceived lack of political responsiveness and a negative assessment of the economic situation may have a negative interactive effect on institutional trust.
Hypothesis 5: All the preceding effects will be more significant for the institutions of political representation (actors and liberal institutions).
Institutional Trust in Spain and Portugal During 2008-2012 in Comparative Perspective
A longitudinal analysis between 2008 and 2012 of the ESS data confirms that the decrease in institutional trust is not a general trend among European democracies, despite the current economic crisis (Armingeon & Guthmann, 2013; Bovens & Wille, 2011). This decrease appears to be more evident in those countries where the crisis is deepest and is not uniform in them all. Spain and Portugal are among the countries in Europe with the most significant decline in institutional trust between 2008 and 2012, although their original levels of trust were already low (Magalhães, 2005; Marien, 2011; Torcal, 2006; Torcal & Magalhaes, 2010). In addition, the recent deterioration of institutional trust is not uniform for all institutions; it is especially visible for the most significant representative institutions included in this survey: national parliaments, the legal system, politicians, and political parties.
The five graphs included in Figure 1 display the different country percentages for citizens located at the lowest levels of institutional trust (positions 0–3 on a 0–10 scale). These figures show that there has been an increase in institutional distrust in many European countries, but it is not uniform across all European countries, the Eurozone countries, or the European countries most severely punished by the economic crisis and by the austerity measures dictated by the EU and other supranational institutions. For instance, this increasing distrust is very low in Ireland. In addition, although this varies according to the institution, in Europe or even among Eurozone countries institutional distrust has decreased in the same time period in Belgium, Norway, Hungary, Israel, and Switzerland and has remained stable or almost stable in Finland, Estonia, France, Germany, Israel, the Netherlands, Sweden, and the United Kingdom. The countries with the most significant increase in levels of distrust are Cyprus (with the highest levels) and, at a considerable distance, Greece, Portugal, Slovakia, Slovenia, and Spain.

Percentage of people with low trust in five basic institutions in Europe, 2010-2012.
The two Iberian countries are among the countries with the largest increase in institutional distrust in Europe between 2008 and 2012. The increase of distrust among citizens in Spain rose from 22.9% to 51.1% for the national parliament, from 53.8% to 77.7% for political parties, from 52.8% to 76.4% for politicians, and from 37.1% to 48.4% for the legal system. Similar figures can be observed for Portugal, in which the percentage of distrustful citizens increased from 49.7% to 67.5% for the national parliament, from 69.1% to 80.2% for political parties, and from 70.1% to 80.7% for politicians. The only exception in the pattern of declining trust in Portugal is the legal system, for which the level of distrust is apparently stable (45.2–49.8%), although it displays high levels overall. The conclusion is that the crisis is not affecting all institutions equally, even in these two countries. For instance, the percentage of citizens distrusting the police in Spain was at a low level of 12.8% and remained at a low 16.6% in 2012. Although the levels of distrust in this institution are higher in Portugal, they also appear to have been unaffected in these last 4 years (20.3% and 21.7%).
The graphs in Figure 1 also show that institutional trust seems to be decreasing differently depending on the institutions: It appears to be especially drastic among the institutions of representation and less so for the institutions of the Rechtsstaat (the courts and the police). This evidence highlights two important facts: It appears to contradict the unidimensionality of institutional trust by showing its distinctive evolution depending on the institution, and second, this is apparently a process confined to the main institutions and actors of democratic representation (political parties, parliaments, and politicians).
This brief description can be concluded by underlining that this decreasing trend in institutional trust is important in Spain and Portugal but is not uniform across Europe (see also Armingeon & Guthmann, 2013; Bovens & Wille, 2011; Marien, 2011), not even across countries suffering the most severe consequences of the economic crisis and the austerity measures implemented by their national governments under the supervision of supranational and international institutions. In addition, it does not affect all institutions equally, despite the important correlation that normally exists in the trust awarded to different institutions.
But can this declining trend also be confirmed at the individual level in Portugal and Spain? To respond to this question, data collected from an online panel sample of the Spanish and Portuguese population have been used (further explanations about this panel design are given below). The individual variation in institutional trust over time during 2012 and 2013 is surprisingly high, given the stability that these attitudes are reported as having, at least at the aggregate level. In Spain, the percentage of individuals who have changed their institutional trust on a 1 to 7 scale during this short period varies from 48% to 60%, depending on the institution, meaning that a majority of citizens have changed their level of trust in this period. The result of this individual change is an important general decrease in institutional trust: for the national parliament from an average of 2.76 to 2.20; political parties from 2.14 to 1.89; and finally, for regional parliaments from 2.59 to 2.32. Only trust in the courts or legal system has not decreased (2.50 to 2.65), showing once more that this trend does not affect all institutions equally. In Portugal, the same pattern is also observable. The individual (within) variation over time in the levels of trust among the Portuguese varies from 50% to a high of 61%, resulting in a general decrease in trust in the national parliament (2.80–2.42), political parties (2.22–2.08), and municipal assemblies (3.54–3.39). Once more, trust in the legal system or courts appears to be low but stable (2.55 and 2.59), confirming the preceding pattern observed at the aggregate level with the ESS data. So what could explain this individual (within) time variation in the level of institutional trust in the two countries?
Data, Research Design, and Models
To assess the relative effect of instrumental (sociotropic and egotropic) evaluations versus the effect of political responsiveness on the recent decline in institutional trust, it is essential to employ panel survey data. Using this type of data has many advantages. The first is the possibility to test causality with longitudinal-linear models. This design allows two methodological problems to be tackled. The first is the problem of endogeneity, which this type of attitudinal study tends to possess. Measurement of independent and dependent variables among the same “i” individuals and the same variables in “t-1” and “t1” allows a clearer estimation of causality among these variables to be established. In addition, it helps to maintain constant contextual variables that might influence institutional trust, such as the institutional setting of each country and additional macrocultural factors.
In this study, we use data collected from an online panel sample of the Spanish and Portuguese populations, two of the countries that, as indicated, have suffered the most severe decline in institutional trust and, at the same time, are exposed to dramatic economic crises and austerity measures imposed by supranational organizations and implemented by their respective national governments.
The survey panel data used here are samples selected from a commercial online pool of respondents in both countries. Respondents were recruited by active invitation among the registered users of the principal commercial websites. Self-registering is not allowed, so the problems of self-selection are thereby reduced. The sample was designed using gender, age, education, and size of habitat quotas. Although this is not a probability sample of the population, this sampling procedure provides a variety in key demographics that minimizes the risk of the results being driven by heterogeneous effects of the stimuli on a specific population group. In addition, recent studies using the sample from this online panel have shown that the resulting estimations produce results very similar to those obtained using the probabilistic sample of the ESS data (Revilla, Saris, Loewe, & Ochoa, 2013).
The data are a panel survey conducted in three different waves in Spain (February-March 2012, November-December 2012, and May-June 2013). The present article uses data from only the first two waves (the third wave deals with a different topic), comprising 1,275 and 1,700 respondents, respectively. The data analysis in Spain will focus only on the 946 respondents who took part in both waves. The survey panel was also conducted in Portugal but only in two waves (May 11-24, 2012, and January-March 2013). The first wave interviewed 1,005 citizens, although only 782 participated in the second wave.
We shall use as dependent variables, on a 1 to 7 scale, individual’s trust in some of the most important institutions of democratic representation already discussed: political parties, regional (municipal in Portugal) governments, national parliaments, and finally, the legal system.
The most significant independent variables included in the model to test the five hypotheses are political responsiveness and economic evaluations (sociotropic and egotropic). Following the existing literature on the topic (Catterberg & Moreno, 2005; Denters et al., 2007; Easton, 1965; Miller, 1974), political responsiveness is measured by the following indicator: “I don’t think that public officials care much what I think.” It is to be expected that disagreement with this statement indicates a favorable perception of responsiveness by government and other authorities. For the sociotropic economic evaluations, we used respondents’ evaluation of the current economic situation in t-1 and t1. Finally, we added to the analysis two indices for the egotropic evaluation of the economic situation: an index of respondents’ personal economic situation and an index of personal future economic uncertainty. These last two indices are individual-level scales created from different items that measure the effects of the economic crisis at the individual level and the level of individual economic uncertainty, respectively. 5 Both are intended to draw on egotropic economic concerns to estimate their different effects on institutional trust.
Three additional time-variant variables with measurements in t-1 and t1 are included to control for the effects of potential rival hypotheses: ideological preferences, social trust, and political interest. First, citizens’ ideological preferences (left-wing or right-wing) may provide further political clues regarding their attitudes toward the current government and governmental policies. Institutional trust might be affected, for instance, by partisan preferences or support for incumbents (Hetherington, 1998; Holmberg, 1999) or by the “winner and loser effects” of the elections (Anderson & LoTempio, 2002; Clarke & Acock, 1989; Kornberg & Clarke, 1992). Second, we cannot discard the possibility of change in social trust over short time periods (Letki & Evans, 2005; Torcal & Magalhaes, 2010), consequently producing change in the individual levels of institutional trust over time (Brehm & Rahn, 1997; Keele, 2007; Newton, 2007; Newton & Norris, 2000; Putnam, 1993; Zmerli & Newton, 2008, 2011). This might be especially relevant in countries exposed to the effect of a sharp economic crisis that severely affects the welfare state and social cohesion (Polavieja, 2013) and, consequently, individual social trust (Kumlin & Rothstein, 2005). Finally, political cognitive mobilization (Inglehart, 1977, 1990), measured by political interest, could also vary as a consequence of more difficult economic circumstances, affecting institutional trust as well.
The typical sociodemographic controlling variables are not needed, as they appear to remain constant in all individuals between t-1 and t1, and their effects are also present in institutional trust at t-1.
A longitudinal-linear regression model with random effects will be estimated, initially including seven time-variant variables with measures in t-1 and t1 as a baseline model:
A second model (2) with random effects will also be estimated, adding only the interactive term between political responsiveness and the sociotropic evaluation of the economic situation to test for the possibility that institutional trust decreases regarding lack of responsiveness, especially in situations where the individual economic evaluation is negative.
Fixed effects linear regression models are those normally used when it is necessary to omit individual-specific variables that are treated as fixed or constant over time. However, fixed effects models are not recommended when there are only two measures in a short period (Hsiao, 1986; Verbeek, 2000). Therefore, although our two initial models will include time-variant variables, we shall estimate both using two longitudinal-linear regression models with random effects.
Results
The results of the estimation of models 1 and 2 for each dependent variable (trust in the national parliament, trust in political parties, trust in regional/municipal authorities, and trust in the legal system) are given in Tables 1, 2, 3, and 4 for Spain and Portugal.
Individual Trust in the National Parliament in Portugal and Spain (Longitudinal-Linear Model With Random Effects).
Note. b/se = coefficient (standard error); Obs. = number of observations.
p < .10. **p < .05. ***p < .01.
Individual Trust in Political Parties in Portugal and Spain (Longitudinal-Linear Model With Random Effects).
Note. b/se = coefficient (standard error); Obs. = number of observations.
p < .10. **p < .05. ***p < .01.
Individual Trust in Regional Parliaments in Portugal and Spain (Longitudinal-Linear Model With Random Effects).
Note. b/se = coefficient (standard error); Obs. = number of observations.
p < .10. ** p < .05. *** p < .01.
Individual Trust in Courts in Portugal and Spain (Longitudinal-Linear Model With Random Effects).
Note. b/se = coefficient (standard error); Obs. = number of observations.
p < .10. ** p < .05. *** p < .01.
As can be observed, there are robust and consistent findings across both countries that should be underlined. Political responsiveness is the most important and robust predictor of institutional trust in both countries and for all institutions, confirming the importance of hypothesis 3. The increasing perception that political authorities are not responsive to citizens’ demands is negatively affecting individual institutional trust. This is true for all institutions, even after including other relevant controlling variables. Instrumental economic variables are also significant in explaining institutional trust in both countries, although to a much lower degree than political responsiveness, showing that the current decline of institutional trust is mostly endogenous to the political process, since it depends largely on citizens’ perceptions of the responsiveness of the political system.
Among the economic instrumental variables, it appears that the sociotropic evaluation of the economy is the most prevalent in both countries and for most institutions. Sociotropic considerations, measured by the effect of the current economic situation, have also been responsible for some of the deterioration of institutional trust in both countries, confirming hypothesis 1. This finding is also consistent for all institutions, although it does not appear to be significant when the interaction is included in model 2 (which means that it is not significant only when the value of the original variable is zero). However, the effect of individual egotropic evaluations of the economic situation is much less significant and even completely insignificant when it concerns personal economic uncertainty regarding the future. Leaving aside the important effect of political responsiveness, only the index of personal economic situation, measured by an additive scale based on the occurrence of three negative events in the past year (losing a job, a salary freeze, or salary cuts), seems to be significant in explaining individual time variation in institutional trust for both countries and institutions (except the legal system in Spain) but appears to be a substantively worse predictor than the sociotropic evaluation of the economy. The other egotropic indicator included in the models, the index of future personal economic uncertainty, proves to be nonsignificant for most of the models. Therefore, hypothesis 2, regarding the effect of egotropic considerations on institutional trust, appears to be largely residual, although not completely groundless.
Finally, the interactive term, which measures the reinforcing or conditional effect of sociotropic economic considerations according to the levels of political responsiveness, is significant for both countries, but only for the national and regional/municipal parliaments, and for the legal system in Spain. This hypothesis (H4) should be rejected for political parties, as it can be observed that the effect of this interactive term generally explains trust in the institutions of liberal democracy (parliaments) and less so for representative actors. For the remaining institutions, the effect of political responsiveness and the sociotropic economic evaluation on institutional trust are complementary but not reinforcing.
Finally, the consistency of these findings in the first three tables (Tables 1, 2, and 3) appears to change when we estimate this model for trust in the courts. As expected, distrust of the courts not only has increased substantially less but has been scarcely affected by the current economic crisis and its consequences. The relationship between trust in the courts and economic evaluations is largely absent, especially in Spain, and is far less closely related to political responsiveness. Consequently, as hypothesis 5 predicts, the institutions of political representation are those under critical scrutiny, due to the current economic and political situation in these two countries.
Other consistent findings are related to the controlling variables included in models 1 and 2. Social trust appears to be a strong and consistent predictor of institutional trust for both countries. This finding could also have negative connotations for the consequences of present events on institutional trust in these two countries. The economic crisis and the dismantling of the welfare state might be having a negative effect on social cohesion (measured by individual social trust) and consequently on institutional trust. The left–right scale also has some effect across both countries, although in Portugal it is significant only for trust in the national parliament. The coefficients show that right-wing citizens tend to show higher levels of institutional trust. This is because parliaments in both countries have a majority of representatives from the conservative parties (the Popular Party in Spain and the Social Democratic Party in Portugal), demonstrating the presence of a certain performance evaluation of incumbents, which mitigates the deterioration of institutional trust.
The Role of Corruption and Multilevel Governance
Is the deterioration of institutional trust in representative institutions mostly attributable to the perceived lack of responsiveness of political authorities? There are two other complementary factors in these countries that, although related to the general perception of political responsiveness, might also be contributing to the decreasing levels of institutional trust: political corruption and the perception that national authorities are responding solely to the demands and interests of supranational organizations and economic actors.
According to the preceding literature on the topic, the presence of specific political scandals in some countries might also have contributed to the deterioration of political trust (Lipset & Schneider, 1983; Nye, 1997; Orren, 1997). The actions of political representatives and the president of the government could be tainted by scandals, and the media focus on such scandals is seen as another contributor to the decline in institutional trust (Hakhverdian & Mayne, 2012; Pharr, 2000; Uslaner, 2011; van der Meer & Dekker, 2011). In addition, this appreciation is more important under specific circumstances, such as a period of economic recession, when individual trust in political institutions is needed more than ever. In such cases, trust is understood as the expectations that people have regarding the promises that institutions make about an uncertain future (Catterberg & Moreno, 2005).
The public perception of corruption in the two Iberian countries is high and has increased substantially in recent years. According to the Transparency International Global Corruption Barometer 2013, 73% of Spaniards and 84% of the Portuguese consider that political corruption has increased in their respective countries between 2009 and 2011. 6 In addition, the three institutions with the highest levels of perceived corruption are political parties (77% and 84%, respectively), national parliaments (61% and 53%, respectively), and the judiciary (48% in both cases). 7 The only institutions that reach figures closer to these are business organizations (51% and 53%, respectively).
European Union membership and multilevel government could be another potential source of institutional distrust in Spain and Portugal. The current economic and fiscal situations in those two countries are impelling national governments to implement austerity measures supervised and even imposed by the EU. This could also increase the sensation that national institutions no longer matter when it comes to representing citizens’ interests and priorities; they are perceived as unresponsive to citizens’ demands (Alesina & Wacziarg, 2000; Armingeon & Guthmann, 2013).
An additional model 3 (shown also in Tables 1, 2, 3, and 4) has been estimated, using two variables to test the effect of these potential sources of institutional distrust. The variables added to our original models 1 and 2 are the perception of corruption and the classic EU member evaluation indicator. It is unfortunate that these variables were included only in t-1, meaning that they are two individual-specific time-invariant variables. When numerous individual units are observed over time, it could be assumed that some of the variables omitted in fixed models will represent factors specific to both the individual units and time periods for which the observations are obtained. It could even be the case that the variables omitted will reflect individual differences that affect the observations for a given individual in more or less the same fashion over time. The estimated model 3 contains this type of variable, estimating a model using the preceding assumptions. So, instead of a fixed model, we shall use once more a random effects model, which treats these individual variables as random. 8
In addition, and for the correct specification of the model, we have also included other political controlling variables. The variables to be included in this final model are, therefore, whether the respondent is a member of a political party, whether he or she is a member of a voluntary organization, the perception of the fairness of the administration, the perception of corruption, EU member evaluation, and finally, political knowledge.
The results of the time-invariant variables in model 3 are strongly significant. The effect of the perception of corruption for all institutions and both countries is marked and significant, supporting those academics who argue that corruption and the perception of political scandals are very important in affecting institutional trust (Catterberg & Moreno, 2005; della Porta, 2000; Hakhverdian & Mayne, 2012; Orren, 1997; Pharr, 2000; Uslaner, 2011; van der Meer & Dekker, 2011). However, its effect is stronger for trust in the courts and constitutes the best predictor for the low decline in trust for this particular state institution. Political responsiveness is still strong and significant for the institutions of representation, although its predictive capacity is reduced when including the corruption variable. However, the effect of EU membership evaluation is essentially residual, although its effects are a little stronger in Spain.
In addition, from among the controlling variables, only political knowledge has some effect in both countries, showing that institutional trust tends to be higher among those with more political information. Finally, the only finding worth mentioning for the remaining variables is that party membership apparently continues to have a positive effect on the level of trust in political parties. Members of parties tend to trust political parties more.
Final Discussion
The growing concern regarding the increasing distrust in political institutions in contemporary democracies appears to be concentrated in a group of democracies, at least in Europe. It is far from being a uniform process among European democracies and even among democracies in that region of the world that are undergoing a dramatic economic and social crisis. In fact, as we have seen in Spain and Portugal, two of the countries suffering the most severe effects of this crisis, sociotropic concerns about the economy are by no means the best predictor of the deterioration of trust in the institutions of representation, having almost no effect on the individual trust in other state institutions. Even egotropic economic concerns appear to be residual in explaining this trend in trust in the institutions of representation.
The main problem in these democracies is the perception that representative institutions are not responsive to their demands and concerns. Contemporary democracies are being challenged as a consequence of globalization, supranational integration, and other issues that are placing these institutions under further scrutiny by citizens. However, the effect of such challenges on citizens’ evaluations of representative institutions is not uniform but depends on the outcome, which is largely dependent on the institutions, the way they work, and how responsive they are to citizens’ demands. It also depends on elite behavior; elite misbehavior has very negative consequences for institutional trust (della Porta, 2000; Pharr, 2000; Offe, 2006). Spain and Portugal present high levels of political corruption (this is the general perception of the public, at least) and this perception is also contributing to the decline in institutional trust. Citizens’ conclusions are clear: Representative institutions and their members are not responsive and, furthermore, are corrupt.
Citizens are responding accordingly, increasing their distrust in representative institutions and in other state institutions also affected by corruption, such as the courts. Citizens are also able to discern and distrust those institutions that are to blame for a lack of responsiveness and/or corruption. There is not, therefore, a general trend of institutional distrust, even in those countries where institutional distrust is clearly growing; this depends on the institutions and how they function.
The current economic crisis is not fully responsible for the decline in institutional trust. Economic and social crises are placing democratic institutions under pressure because scarcity produces more social conflicts. Democratic institutions are supposed to respond to such conflicts fairly and inclusively. According to citizens’ perceptions, this is not the case in Spain and Portugal, and thus their trust in these institutions is affected dramatically. This process is exacerbated by the general perception of corruption. Consequently, economic recovery in the years to come will not restore acceptable levels of institutional trust by itself but will need to be accompanied by serious institutional reforms and measures to substantially reduce political corruption.
Footnotes
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Spanish Ministry of Innovation and Science, research project reference: CSO2009-14434.
