Political support in Latin America: theoretical and empirical aspects

Institutional trust is vital for democracy (Berg and Hjerm, Reference Berg and Hjerm2010). In fact, as underlined by Marien and Hooghe (Reference Marien and Hooghe2011), trust reduces the costs related to the monitoring of the political process, allows citizens to delegate decision-making, and reduces the complexity of ruling, making it one of the most vital assets of democracies.

From a theoretical point of view, political trust may be defined as ‘a basic evaluative orientation toward the government founded on how well the government is operating according to people’s normative expectations’ (Hetherington, Reference Hetherington1998: 791). Citizens who have trust in political institutions tend to support the political choice of the elite because the latter are working for the collective well-being. Following the institutional explanations of political trust, trust is an extension of generalized trust, assimilated through the process of socialization and later transferred to the political system and its institutions. Since trust in institutions is ‘rationally based’ (Mishler and Rose, Reference Mishler and Rose2001: 31), the quality and performance of the institutions (Rothstein, Reference Rothstein2008, Reference Rothstein2011) are the elements through which the citizens assess their institutions. In this view, high political trust implies that institutions perform well, while low levels of political trust relate to performance failure (see Listhaug and Ringdal, Reference Listhaug and Ringdal2008; Choi and Woo, Reference Choi and Woo2016).

From an empirical point of view, political trust is a multifaceted concept. Although the variety of normative expectations that characterize democratic institutions have prompted some scholars to believe that the basis of trust in democratic institutions seems to be particularly ambiguous (Grönlund and Setälä, Reference Grönlund and Setälä2007: 402), recent studies have analysed it in theoretical and empirical terms. As Denters and colleagues (Reference Denters, Gabriel and Torcal2007) have underlined, in the wake of Gabriel et al. (Reference Gabriel, Kunz, Roßteutscher and van Deth2002), political trust can be conceptualized as a political orientation towards political actors as well as towards political institutions. At the same time, it is possible to distinguish among political institutions, differentiating those that characterize contemporary democracies (the Parliament, the National government), from representatives of a central decision-making agency (political party, politicians), and from institutions that represent the constitutional state (the Police and the Courts). Other scholars prefer to distinguish political institutions in terms of those that are partisan and those that are order/neutral (Rothstein and Stolle, Reference Rothstein and Stolle2008), while other researchers add the international classification to the aforementioned two (André, Reference André2014). However, there are studies in which political trust is represented as a one-dimensional concept (Fischer et al., Reference Fischer, van Heerde and Tucker2010; Hooghe, Reference Hooghe2011) – different levels of trust that citizens have in political institutions are synthesized in an index of political trust using factorial techniques or additive-type aggregation procedures (see Krauss et al., Reference Krauss, Nemoto, Pekkanen and Tanaka2016; Kroknes et al., Reference Kroknes, Jakobsen and Grønning2016; Pitlik and Rode, Reference Pitlik and Rode2016).

The studies that have analysed democratic support, even in light of the current economic crisis (see Armingeon and Guthmann, Reference Armingeon and Guthmann2014), outline a concerning scenario. At the end of the 20th century, the major industrialized democracies appeared to be in crisis (Kaase and Newton, Reference Kaase and Newton1995) and a sense of malaise characterized democratic citizens (Norris, Reference Norris1999); in the new millennium, the scenario does not seem to have changed. From one side of the globe to the other, an erosion of democratic support is evident (Dalton, Reference Dalton2004), and a sense of scepticism is increasing in the public, as well as a moving away from the political arena (Memoli, Reference Memoli2011; Karp and Milazzo, Reference Karp and Milazzo2015; Ezrow and Xezonakis, Reference Ezrow and Xezonakis2016). At the same time, the younger generations appear increasingly inclined not to promote democratic principles and practices, and do not seem to oppose an alternative that is antithetical to democracy (Denmark et al., Reference Denmark, Donovan and Niemi2014).

This trend appears to be even more evident in the countries of Latin America, where even at the beginning of the new millennium, the initial euphoria evoked by democratic change had passed, and in some countries, such as Brazil, Paraguay, Venezuela, and Columbia, democracy appeared to be in dire straits (Lagos Cruz-Coke, Reference Lagos Cruz-Coke2001: 137–138). Many scholars have concluded that in these countries, mass support for democracy is not particularly consistent (Seligson, Reference Seligson2008), and support for democratic performance does not seem to be affected by the economic crisis (Graham and Sukhtankar, Reference Graham and Sukhtankar2004). Although the progress recorded in the process of the assimilation of democratic values and principles (Diniz, Reference Diniz2011) confirms a clear optimism in public opinion, over time, there has been a contraction in the hope for change on the part of public opinion. As pointed out by Smith (Reference Smith2005), the cyclic economic crises, the economic adjustments of the neoliberal matrix, and the issues related to the political system, above all, in terms of cronyism and corruption, have decreased citizens’ expectations and have negatively affected democracy. Citizens are often disappointed with democracy and less inclined to support it (Sarsfield and Echegaray, Reference Sarsfield and Echegaray2006). In addition, episodes of political corruption have made political institutions and political leaders less credible and have lowered the level of support for democracy (Ergun et al., Reference Ergun, Rivas and Rossi2016). Citizens appear to be characterized by more aggressive political participation and less tolerance of abuses of power (Morris and Blake, Reference Morris and Blake2010); these factors, along with the consolidation of abuses from the security forces, have undermined and delegitimized the political system (Cruz, Reference Cruz2015). Observing the democratic process in Latin American from a longitudinal perspective, a clear weakness in the political and institutional conditions emerges, as well as the absence of effective strategies to strengthen them and ensure a stable democratic regime and its consolidation in the long run.

Economy and trust in political institutions

The economy is considered by many scholars as the driving force of democracy (see Lipset, Reference Lipset1959) and its consolidation (Diamond, Reference Diamond1999). According to Huntington (Reference Huntington1991), positive (negative) economic cycles increase (decrease) the chances of survival of democratic regimes, as well as changes in democratic support from the citizens (Dalton, Reference Dalton2004). The numerous theories that adopt a systemic perspective link democracy with the economy. However, not all scholars seem to agree on the causal direction binding them, feeding conflicting visions. Indeed, some scholars believe democracy and its duration are the main factors which contribute to economic growth (Carbone et al., Reference Carbone, Memoli and Quartapelle2016). Others show that economic growth fuels the democratic transition (Boix and Stokes, Reference Boix and Stokes2003) and subsequent democratic maintenance (Epstein et al., Reference Epstein, Bates, Goldstone, Kristensen and O’Halloran2006). This is found in both among the established democracies (Bellucci and Memoli, Reference Bellucci and Memoli2012) and in countries where democracy is still nascent (Quaranta and Memoli, Reference Quaranta and Memoli2016). Over the years, although the number of studies has grown, the empirical results reveal contradictory trends and a weak effect of the economy on democracy (Dalton, Reference Dalton2004).

Since the 1990s, there has been an increase in studies that have focussed on the links between the economy and democratic support (Dalton, Reference Dalton2004). The ‘democratic malaise’ highlighted by several comparative studies (see Norris, Reference Norris1999) summarizes the fear of delegitimization of the political system linked to economic trends, whose oscillations inevitably influence voters’ political decisions (Anderson, Reference Anderson1995), as well as their judgements on the government (Alesina and Wacziarg, Reference Alesina and Wacziarg2000) and on democratic support by citizens (Dalton, Reference Dalton2004).

The connection that combines the economy and the political system is well evidenced in terms of accountability and responsiveness (on this point, see Tavits, Reference Tavits2007), even within the debate on economic voting. In fact, citizens tend to renew their trust in the political institutions (in this case, the incumbent government) if its performance is positive; if it is not, the voters will move their political preferences elsewhere (see Lewis-Beck and Stegmaier, Reference Lewis-Beck and Stegmaier2000: 183).Footnote ⁴ With the economic crisis, citizens have voiced their dissatisfaction, protesting against governments to prompt them to address and resolve the crisis (van Gent et al., Reference van Gent, Mamadouh and van der Wusten2013). This aspect is not secondary, because the economic insecurity that could be generated from the negative institutional performance could affect the sense of trust in institutions (Hacker et al., Reference Hacker, Rehm and Schlesinger2013) and nourish nationalistic sentiments (Kage, Reference Kage2013).

The perception of any phenomenon by a citizen is a fundamental element in human decision-making (see Moreira, Reference Moreira2015). Considering that both the perceived and real economy are determinants for democratic support (see McAllister, Reference McAllister1999; Karp et al., Reference Karp, Banducci and Bowler2003; Neundorf, Reference Neundorf2010; Kotzian, Reference Kotzian2011; Quaranta and Martini, Reference Quaranta and Martini2016), it is possible to assume a conjoint effect of the economy (personal and national) on the levels of political trust expressed by citizens. This interaction could be considered using both measures of economic perception (personal and national) and measures that combine these latter with different measures of the real economy.

Following the economic voting theory, it has been shown that voters are influenced by their subjective views of the national economy, even though they are not much swayed by their personal economic standing (Kinder et al., Reference Kinder, Adams and Gronke1989). Thus, voters tend to respond to their beliefs about the state of the overall economy rather than to their personal pocketbooks. Since citizens seem capable of evaluating macroeconomic outcomes (Duch and Stevenson, Reference Duch and Stevenson2011), and considering that the evaluation of government is realized adopting a sociotropic view rather than an egocentric view, it is possible to assume that citizens’ political trust could also be affected by a sociotropic view of the economy. Since egocentric and sociotropic economic assessments are not independent, we consider it appropriate to look at the economy in terms of an egocentric view as well. That choice is supported by the fact that the studies of economic voting realized in Latin American countries show that both egocentric and sociotropic perceptions affect voting for the incumbent government (see Weyland, Reference Weyland2003; Singer and Carlin, Reference Singer and Carlin2013). Therefore, it is possible to hypothesize that:

Political trust in Latin American countries

In the 1960s, the overload of the Latin American governments generated by the growing demand from citizens affected the economic development and state management, which was already fragile, leading to a return to authoritarian regimes (O’Donnell et al., Reference O’Donnell, Schmitter and Whitehead1986), as in the cases of Peru, Brazil, Bolivia, and Argentina. Through the 1980s, the democracy among the countries of Latin America has undoubtedly strengthened: first in Ecuador (1979) and followed by other countries,Footnote ⁵ the transition from authoritarianism produced rapid change on the continent. In the early 1980s, the external debt crisis did not lead to changes in economic institutions, and liberal reforms were implemented only at the end of the decade by rebalancing the relationship between the state and the market. In the 1990s, numerous countries in Latin America could be considered in transition to democracy because, after a long period of dictatorship, most political regimes were transforming into liberal or electoral democracies. However, the optimism of political analysts that was founded on greater democratic stability and a better socio-economic balance did not translate into a real democratic consolidation. Economic and technological dependence, along with socio-economic inequality and poverty, were the main factors that influenced both the problematic evolution of democratic regimes (see Offe, Reference Offe2009) and the fragility of the states in finding a balance between economic performance and social rights.

Following Easton (Reference Easton1965) and other scholars, using information gathered from the Latinbaròmetro, it is possible to trace the support for the political institutions from a longitudinal point of view.Footnote ⁶ To do this, we take into account the citizens’ trust in the following political institutions: the National Congress/Parliament, the Judiciary System, the Police, and Political Parties.Footnote ⁷

Figure 1 shows the percentage values of those who have expressed that they have much or sufficient trust in political institutions. From 1996 to 2013, the trend of trust has been swinging, and it has been significantly affected by the past years. In the early part of the 1990s, political trust continued to be conditioned by a shift between authoritarianism and democracy (Bargsted et al., Reference Bargsted, Castillo and Somma2017), making trust in public institutions uncertain. In the 1990s, both the contraction of poverty and inequality, as well as annual GDP growth (see Cano, Reference Cano1999), fuelled the idea of a clear economic improvement. The effects on citizens’ political trust are evident at least until 1997, when the globalized economy began to decline after the Asian, and later, the Russian, Brazilian, and Argentine crises. In a pricing/punishment logic, the trust that citizens placed in their political institutions, and especially political parties, declined significantly. From 1997 to 2003, although trust had never been particularly high, it declined by 18.2 percentage points. Only trust in the Police grew.

Source: Latinobarómetro (1996–2013).

Figure 1 Trust in political institutions.

Since 2004, the sudden growth of Latin American economies has generated more public trust among citizens in the institutions. Economic growth has contributed to reducing the level of indigence (Economic Commission for Latin American and Carribean, 2012), driving among the citizens an increased propensity to trust in their political institutions. This trend characterizes all political institutions, including Political Parties, which remains at a lower level of trust despite the recordation of a growth in trust by 10 percentage points.

Using different information about trust in political institutions (National Congress/Parliament, Judiciary System, Police, and Political Parties) and applying a factor analysis,Footnote ⁸ we obtained factor scores as a synthesis of the analysed information. Through this index, it is possible to differentiate the countries of Latin America from the perspective of time (1996–2013; Table 1). Our results reveal that all four indicators are interconnected to one factorial dimension obtained by the method maximum likelihood factor and have values greater than or equal to 0.615. This value confirms that attitudes towards political institutions tend to be generalized to different levels of measurement (Bartlett’s test of sphericity = 0.000; Kaiser–Meyer–Olkin = 0.774).Footnote ⁹ All items positively contribute to the reliability of the factor (Cronbach’s α = 0.779).Footnote ¹⁰

Table 1 Factor analysis

Source: Latinobarómetro (1996–2013).

The substantial drop in political trust over time reveals a certain dissatisfaction that characterizes some countries (Table 2). This is especially evident in Honduras, Guatemala, Paraguay, Nicaragua, and Nicaragua, where democracy has slowly tried to take off, but also in Chile, where democracy is consolidated. On the other hand, in Venezuela, political trust had a positive peak with the coming to power of Chávez (1999–2013), and his redistributive social policies contributed to fostering greater trust among the citizens in the institutions until 2013, when his successor, Maduro, replaced the late President. The same trend is noted in Ecuador, where the increase in public spending and the resulting investments made Ecuador’s economy more and more dynamic at the end of the past decade, with obvious reflections on the collective welfare. Moreover, the establishment of a safety net for the neediest and unemployed, and contributions in favour of families reduced the rate of poverty, making the institutions more credible and worthy of support in public opinion.

Table 2 Trust in political institutions

The values are factor scores obtained from the factor analysis. Positive values express citizen’s positive evaluation of political institutions. By contrast, negative values synthesize an unfavourable assessment in the institutions.

Source: Latinobarómetro (1996–2013).

The explanatory model

We estimated the effects of citizens’ economic perceptions on political trust among Latin American countries through different multilevel regression models. The dependent variable is represented, as described above, by an index (factor scores), which summarizes the levels of trust that citizens have in political institutions.Footnote ¹¹ The main explanatory independent variables at the individual level are the perceptions of the current state of the national economyFootnote ¹² and the status of the personal economic situation.Footnote ¹³ At the aggregate level, GDP per capita PPP and GDP growth were used.Footnote ¹⁴ Because the information that characterizes our data set has been collected over an extensive period (1996–2013) and considering that there is a temporal gap between changes in the economic situation and people’s perceptions, the information related to the real economy (GDP per capita PPP and GDP growth) was collected in the year preceding each survey.

A variety of information related to the dependent variable, collected both individually (gender,Footnote ¹⁵ age,Footnote ¹⁶ education,Footnote ¹⁷ income,Footnote ¹⁸ trust in people,Footnote ¹⁹ and satisfaction with democracyFootnote ²⁰) and at aggregate level (government approval,Footnote ²¹ Gini index,Footnote ²² economic crisis,Footnote ²³ and level of corruptionFootnote ²⁴), were used as control variables. Finally, to overcome any problems of heteroscedasticity, we used the cluster standard errors.

The effects the economy has on trust in political institutions account for the differences among the countries of Latin America. In the first model (Table 3), whose variance explained is equal to 14.1%, we shed light on citizens’ personal and national economic perceptions. What emerges appears to confirm that a public perception of economic wealth fosters trust in political institutions. In fact, when citizens perceive a sense of economic well-being on a personal level, it appears more likely that they will trust their political establishment, when the perception of the national economy increases (b = 0.025). The combined effect of the two indicators underlines how the economy plays an important role in explaining institutional trust, namely the pattern of trust between governed and governors.

Table 3 Perception economic and political trust in Latin America

*P<0.1; **P<0.05; ***P<0.01; ****P<0.001.

Source: Latinobarómetro (1996–2013); World Bank (1995–2012); The Executive Approval Project (1996–2013).

The economy of Latin American countries in the last decades has undoubtedly been fluctuating, alternating positive trends with negative trends. The economic experience in this continent has certainly influenced the citizens’ economic perceptions. Nevertheless, the economic evaluation of citizens does not appear to be different from what emerges from the real economy. In fact, as shown in model 2, when citizens appear satisfied with their economic well-being, they support their political institutions more when the real economy, estimated in terms of GDP per capita PPP, increases (b=0.033). Although adverse economic experiences often are negatively correlated to economic knowledge (Kalogeropoulos et al., Reference Kalogeropoulos, de Vreese, Albæk and Van Dalen2015), Latin American citizens undoubtedly appear sophisticated, perhaps even more than the data in our possession allow us to estimate. This is confirmed when we analyse the empirical findings reported in model 3, whose variance explained is 13.9%. In fact, the combined effect of citizens’ national economic perceptions and economic growth increases the willingness of citizens to trust their political institutions (b=0.006). From this perspective, even when the public evaluates the economy from a sociotropic view, relying on its knowledge, it tends to positively evaluate political institutions.

In order to clarify the primary factors that nurture political trust, in the fourth model, whose variance explained is 14.2%, we considered all the independent variables analysed previously. The perceptions that citizens have of the personal and national economy have a significant impact on political trust (b=0.022 – H1 confirmed). Specifically, when citizens perceive a sense of personal economic well-being and, at the same time, they positively evaluate the national economy, they tend to trust more in their political institutions (see also Figure 2).

Source: Latinobaròmetro (1996–2013).

Figure 2 Average marginal effects of personal economic perception with 90% CIs.

While GDP per capita PPP (log) shows a negative effect on trust in political institutions (b=−0.059), unlike what is expected, the combined effect of individual economic perceptions and GDP per capita PPP on the dependent variable is negative, and it is not statistically significant (b = −0.014, H2 not confirmed). As people have more available resources, they should be more likely to have trust in political institutions and eventually appreciate political outcomes. However, at least in the Latin American countries, this is not the case, probably because the fluctuating trends of the economy and political instability induce citizens to fear political and economic upheavals.

When citizens express a positive view regarding the national economy, even in the presence of a negative economic performance, they tend to assess their institutions positively (b = 0.006, H3 confirmed, see also Figure 3). This may be due to the fact that the evaluations of economic performance expressed by citizens may be affected, and then reflect, their political predispositions in terms of support for the government, as well as their personal financial experience (Duch et al., Reference Duch, Palmer and Anderson2000). However, as the economy improves, marking a positive trend, it strengthens the trust of the citizens in the institutions. The results appear to attenuate the idea that random variation associated with the data survey can make the association between the information collected at the aggregate level and the individual level inconsistent (on this point, see Page and Shapiro, Reference Page and Shapiro1992).

Source: Latinobaròmetro (1996–2013); World Bank (1995–2012).

Figure 3 Average marginal effects of National economic perception with 95% CIs.

The empirical findings obtained using individual and structural variables confirm our results. Even if the economic crisis continues to affect the nexus between the citizens and the political institutions (b = −0.010), those with a satisfactory income to cope with their individual and family needs support and trust the political institutions (b = 0.059). In general, education presents a negative relationship with the political trust index. However, those who have a higher education level appear to be less critical towards their institutions (b = −0.014) than those with a lower level of education; in contrast, those who approve the choices of the government, appear satisfied with democracy, trust in others seem to be more inclined to support political institutions. Finally, although the economic crisis, despite the passing of time, continues to represent a burden for political trust, when corruption decreases, the level of trust that citizens place in their political institutions increases (b = 0.116).Footnote ²⁵

These results undoubtedly require consideration, because the choices that governments have taken in recent years among Latin American countries have generated clear dissatisfaction, fuelling a reactionary attitude in citizens. The radical opposition of the right and the loss of public trust in progressive governments in Bolivia, Argentina, and especially in Venezuela and Brazil, confirm this trend;Footnote ²⁶ in Brazil, it was manifested openly with the last election and the impeachment of Dilma Rousseff. Undoubtedly, this scenario is affected by the economic crisis among the countries of Latin America, which having undermined the perception that citizens have of the economy, has eroded public support for democracy (see Córdova and Seligson, Reference Còrdova and Seligson2010). Nowadays, the citizens continue to swing between Western-style democracy and the personal and authoritarian role of charismatic leaders. Nevertheless, when the economy is perceived positively, citizens, in line with the theory of economic voting, tend to support their political representatives and political institutions.