Clientelistic parties and satisfaction with democracy

Abstract

Using both a cross-national data set on parties’ accountability strategies and public opinion survey data, this article provides a systematic analysis of how parties’ reliance on clientelistic strategies correlates with citizen evaluations of regime performance. Our analyses suggest that in systems where parties in general rely more heavily on clientelism, principles of democratic equality are undermined, and citizens on average express a lower level of satisfaction. In addition to democratic norms, material benefits are also closely related to citizens’ regime evaluations. Within countries, supporters of parties that make substantial clientelistic efforts are more satisfied than other citizens, because they are likely to be beneficiaries of this accountability mode. This within-country gap between clientelistic and non-clientelistic parties’ supporters widens where the clientelistic parties deliver benefits more effectively and in countries with lower overall democratic quality.

Keywords

accountability strategies clientelism democratic satisfaction party politics

Introduction

Determinants of citizens’ evaluations of democracy have been extensively studied in political science. Existing studies have provided ample evidence showing that citizens’ perceptions of system outputs and opportunities for political inputs have impact on their satisfaction with democracy. With regard to system outputs, citizens express higher levels of satisfaction if the government performs well in delivering economic and political goods. As for the opportunities of political inputs and the quality of political representation, electoral winners show significantly greater satisfaction. Moreover, the gap between winners and losers is reduced with greater opportunities for minorities to voice in decision-making processes.

The existing research has mainly focused on the programmatic mode of political exchange between politicians and voters. That is, how the provision of public goods and representation of ideological/broad policy positions affect citizens’ evaluations of democracy. However, programmatic politics, in which politicians provide broad policy packages in exchange for voters’ support, is not the only linkage mode between politicians and voters. Scholars have pointed to clientelism as an alternative accountability strategy (Kitschelt and Wilkinson, 2007). In contrast to political competition characterized by parties offering different programmatic platforms of policy bundles, clientelism involves direct and contingent exchange between votes and targeted goods delivered only to party supporters.

How does clientelistic exchange affect citizens’ evaluations of democracy? Clientelism is portrayed as a linkage form that undermines democratic equality principles (Fox, 1994; Hicken, 2011; Stokes, 2007). When politicians heavily rely on clientelism to gain support, the democratic principles, including citizens’ equal access to political process and the fair application of laws and policies, are weakened. Since the distribution of benefits is contingent on voters’ support for the winning party in a clientelistic relationship, only preferences of clientelistic party supporters are taken care of. Moreover, voters’ ability to hold politicians accountable is undermined. When a clientelistic ruling party is able to punish voters/regions for supporting others, voters instead become the ones held accountable for their behavior (Stokes, 2005).

The literature has documented how clientelism undermines democratic principles and how undermined democratic principles may affect citizens’ regime support. However, whether parties’ reliance on clientelism has influence on citizens’ democratic evaluations has not been systematically examined. This article aims to bridge this gap. We hypothesize a negative relation between clientelistic politics and citizens’ satisfaction with democracy. In countries where parties in general rely more heavily on clientelism, citizens on average express a lower level of satisfaction. In addition to democratic equality, government output of material benefits is also related to citizens’ regime evaluations. Therefore, within countries, voters who support parties that make substantial clientelistic efforts should evaluate the political system more positively than supporters of less clientelistic parties, because they are likely to be beneficiaries of this linkage mode. Furthermore, this within-country gap between clientelistic and non-clientelistic parties’ supporters is enlarged if the clientelistic parties are able to deliver benefits more effectively and in countries with lower democratic quality. We test these arguments by combining data from several public opinion surveys and data from the Democratic Accountability and Linkage Project (DALP), the most comprehensive cross-sectional data set of parties’ clientelistic strategies.

Satisfaction with democracy and clientelism

Satisfaction with democracy

Studies of citizens’ satisfaction with democracy are part of the larger literature on regime legitimacy. As theorized by Easton (1975) and Norris (1999), political legitimacy is a multidimensional phenomenon. Norris’ five-fold model includes support for the political community, general regime principles, regime performance, specific institutions, and specific actors. Although “satisfaction with democracy” is still a contested concept (Anderson, 2002; Canache et al., 2001; Linde and Ekman, 2003), it has been generally thought of as referring to citizens’ evaluations of the way democracy works in practice (e.g. Aarts and Thomassen, 2008; Anderson and Guillory, 1997; Blais et al., 2015; Weitz-Shapiro, 2007). As such, it concerns the middle level of Norris’ classification between the diffuse and specific levels of political support (Ferland, 2015; Norris, 1999) and is correlated with, but distinct both conceptually and empirically from, approval of the principles of democracy (Bratton and Mattes, 2001; Clarke et al., 1993; Klingemann, 1999).

The literature shows that the level of satisfaction with democracy is a function of citizens’ evaluations of government output and opportunities for input into the democratic process. Citizens base their evaluations of government output on the delivery of economic and political goods. With regard to economic goods, the economic performance of a country and public perception of personal and national economic conditions are the main factors (Anderson and Guillory, 1997; Clarke et al., 1993; Criado and Herreros, 2007; Finkel et al., 1989; Hofferbert and Klingemann, 1999; Lühiste, 2014).

Several aspects of political goods affect democratic satisfaction. Studies on new democracies show that perceived increases in freedom (Hofferbert and Klingemann, 1999; Mishler and Rose, 2001a) and government responsiveness (Evans and Whitefield, 1995) contribute to higher levels of regime support. Examining the effects of political institutions, scholars argue that citizens are more satisfied in systems allowing more proportional representation, partly because this enhances the perceived fairness of electoral procedures (Birch, 2008; Blais and Loewen, 2007). Studies also highlight the importance of impartial institutions. For example, Dahlberg and Holmberg (2014) and Linde (2012) demonstrate that the quality of government institutions, including fair treatment by public authorities, strongly affects public support for regime performance. Similarly, other studies suggest that citizens’ perceptions of corrupt activity by politicians or government officials are negatively correlated to satisfaction with democracy and related indicators, including institutional trust, subjective well-being, and legitimacy of the regime. This contention is supported by both cross-country analyses (Anderson and Tverdova, 2003; Canache and Allison, 2005; Mishler and Rose, 2001b; Seligson, 2002, 2006; Tavits, 2008) and evidence from single countries (Linde and Erlingsson, 2013; Morris and Klesner, 2010). The theoretical explanation is that corruption undermines the democratic principles of impartiality and accountability; seeing public authorities influenced by special interests creates popular discontent and alienates citizens from the system (Booth and Seligson, 2009; Rothstein, 2009; Tavits, 2008).

With regard to citizens’ opportunities for political inputs, the higher levels of satisfaction among electoral winners have been studied extensively (Anderson and Guillory, 1997; Blais and Gélineau, 2007; Blais et al., 2015; Clarke et al., 1993; Curini et al., 2011; Delgado, 2016; Singh, 2014; Singh et al., 2012). The winner/loser gap is partly due to the pleasure of being on the winning side (Anderson et al., 2005; Sing et al., 2012) and can also be explained by citizens’ policy-oriented considerations (Brunell and Buchler, 2009; Kim, 2009): citizens express more satisfaction when their policy preferences are represented (Ezrow and Xezonakis, 2011; Stecker and Tausendpfund, 2016). However, winning/losing elections can mean different things in different systems. In “consensual” systems, which involve more power sharing and provide minorities with more opportunities to be represented in decision-making, the differences in system support between winners and losers are lower than in “majoritarian” systems (Anderson and Guillory, 1997; Bernauer and Vatter, 2012).

The impacts of clientelism

Clientelistic linkage modes involve the provision of targeted goods and services by politicians to individuals or small groups of voters in exchange for votes. Under programmatic accountability mechanisms, voters’ demands are fulfilled through broad policy packages, while voters in a clientelistic relationship expect to obtain party-sponsored benefits directly tailored to specific narrow constituencies. Clientelism is generally seen as having pernicious effects on the quality of democracy (Keefer, 2007).

We argue that in countries where parties rely more heavily on clientelism, citizens show generally lower levels of satisfaction with democracy. There are several reasons for this expectation. First, clientelism leads to a narrower and less efficient distribution of economic goods, which has been considered as an important determinant of performance evaluation. The negative consequences of clientelism on governance outcomes include larger public deficits and public sector inefficiencies (Hicken, 2011). Parties/politicians divert public resources to invest in patronage networks and monitoring mechanisms, which may lead to less public goods provision (Hicken and Simmons, 2008; Keefer, 2007).

More importantly, clientelism negatively affects democratic satisfaction by reducing citizens’ opportunities for political inputs and the provision of political goods. As Stokes et al. (2013) point out, clientelism weakens the public’s interest in elections as instruments of political inputs. When votes are secured through targeted benefits, they do not convey much information about voters’ policy preferences or incumbents’ performance. The electoral mechanism of political representation, which is helpful in boosting satisfaction, is thus impaired.

Moreover, clientelism undermines the principles of democratic equality. The delivery of goods is contingent on citizens’ votes for the winning party in a clientelistic relationship. Only loyalists of clientelistic parties and/or swing voters are likely to receive benefits, and supporters of other parties are significantly disadvantaged. Citizens’ equal access to political process and the fair application of laws and policies are weakened. Studies on the impacts of institutional arrangements and corruption all agree that procedural fairness and equality substantially boost satisfaction. Literature in other areas also suggests that the perceived fairness of decision-making procedures is crucial for legitimacy beliefs in legal, political, and managerial settings (see MacCoun, 2005; Tyler, 2006, for reviews). Esaiasson et al. (2012: 790) point out that citizens value procedural fairness because “it is a moral right fulfilled, it assures them that they are respected by the decision-making authority, it indicates that the decision-making authority is trustworthy and it enables them to reduce uncertainty about the fairness of substantial outcomes.”

Clientelistic exchange undermines the notion that citizens are entitled to equal legal treatment. Witnessing that only followers of clientelistic parties have preferential access to social welfare or public sector employment, and only firms close to those parties get government contracts and enjoy favorable regulatory treatment, citizens experience violations of the principle of impartiality. When clientelism is widespread, concerns of democratic equality and fairness are expected to move individuals to express disapproval of the performance of democracy.

Even voters who benefit from clientelism may be less content compared to citizens in a generally less clientelistic party system. Their policy preferences or opinions about incumbents’ performance can be largely ignored, because their support is bought with targeted benefits. Additionally, a clientelistic party’s decision on whether, how, and whom to distribute goods may be opaque or erratic from election to election. Voters’ uncertainty about distributive outcomes is not reduced. Research on the effects of procedural fairness has also shown that unjust procedures engender negative emotions even among individuals who benefit from them (Krehbiel and Cropanzano, 2000; Weiss et al., 1999).

This negative relationship between parties’ reliance on clientelism and democratic satisfaction has yet to be tested systematically. This article aims to test the following hypothesis:

Hypothesis 1: In societies where parties rely more heavily on clientelistic strategies, citizens on average show lower levels of satisfaction with democracy.

We further argue that the general negative associations between the prevalence of clientelism and system satisfaction are not uniform within a country. More specifically, within countries, supporters of parties that do not engage in clientelism are more dissatisfied. In contrast, supporters of clientelistic parties, who are likely to be beneficiaries of targeted goods distribution, may view clientelism as an efficient way to hold their representatives accountable (Lindberg, 2010; Lust-Okar, 2009). In addition to democratic principles, citizens’ perceptions of government output are also closely related to system support. Therefore, citizens with an allegiance to clientelistic parties are likely to be more satisfied with democracy than fellow citizens who support parties that distribute fewer targeted goods.

Similar theoretical expectations about the trade-off between norms and material benefits have been established in studies on corruption. For example, perceived political corruption has less negative influence on presidential approval when economic conditions are favorable (Zechmeister and Zizumbo-Colunga, 2013). In the absence of fair procedures, economic performance matters more for regime evaluation (Magalhães, 2016). Specifically, research finds that citizens are likely to overlook the bad reputation of corrupt politicians and still support them if they deliver goods to their constituencies (Anduiza et al., 2013; Manzetti and Wilson, 2007; Winters and Weitz-Shapiro, 2013). Manzetti and Wilson (2007) explain that corrupt governments may survive if they are able to buy off voters through clientelistic networks. That is, material benefits can be viewed as “compensation” for the weakened democratic principles.

This article aims to examine more explicitly the trade-offs between norms of democratic equality and targeted benefits. It is expected that citizens are more tolerant toward clientelism if they support those clientelistic parties and thus are likely to be beneficiaries of patronage. Compared to citizens supporting clientelistic parties, those attached to non-clientelistic parties may particularly feel alienated from this biased system because they do not get favors as compensation for weakened democratic norms. A hypothesis is proposed:

Hypothesis 2: Within countries, supporters of parties that make substantial clientelistic efforts show higher levels of democratic satisfaction than other citizens.

Conditions of within-country differences

We also hypothesize that within a country, the gap in regime evaluations between supporters of clientelistic and non-clientelistic parties expands or contracts, depending on parties’ ability to really deliver clientelistic benefits. Following Hypothesis 2, the argument that supporters of clientelistic parties are more content compared to other citizens is based on the assumption that these supporters are beneficiaries of targeted goods distribution. Therefore, it is expected that the supporters are particularly tolerant when they can obtain targeted benefits more easily.

The literature has pointed out parties’ resources and support networks that make the delivery of clientelistic goods possible. As Szwarcberg (2014) suggests, capacity of politicians to provide targeted benefits is determined by the combination of their access to particularistic goods and ability to distribute these goods to voters. Access to resources depends on politicians’ party affiliation. Incumbents with control over state resources and administrative branches can use this to focus on private transfers rather than provide programmatic policies (Calvo and Murillo, 2013; Magaloni et al., 2012). Larger parties are more likely to have organizational and material means to develop clientelistic relationships. Therefore, the within-country gap in democratic satisfaction between supporters of clientelistic and non-clientelistic parties is likely to depend on parties’ majority and minority status: supporters of winning clientelistic parties tend to evaluate the performance of democracy more positively because these parties are better able to effectively distribute committed benefits.

Parties’ ability to target and transfer particularistic goods also depends on their networks of local intermediaries (Koter, 2013; Szwarcberg, 2014). Extensive local organizations facilitate clientelistic relationships and enhance delivery of goods. Parties that have local leaders as their intermediates are able to access a broader base of voters. Kitschelt and Kselman (2011) show that parties’ informal organization, non-partisan notables (such as neighborhood, religious, or community leaders) who are delegated the task of maintaining contact with voters on parties’ behalf, can assist clientelistic exchange. These non-partisan local leaders have preestablished relationship to the constituencies, which can provide parties with individual-level information about whom to target. Additionally, local networks can be utilized to transmit benefits and possibly monitor opportunistic voter behavior. Overall, parties’ informal local intermediaries enable more effective practice of clientelistic exchange.

Based on these arguments, parties with more resources or well-established networks of local intermediaries can distribute benefits more effectively and thus make their supporters particularly satisfied, compared to supporters of other parties. Two hypotheses are proposed:

Hypothesis 3.1: The within-country differences in democratic satisfaction between voters supporting clientelistic parties and those attached to less clientelistic parties increase with the clientelistic parties’ size.

Hypothesis 3.2: The within-country differences in democratic satisfaction between voters supporting clientelistic parties and those attached to less clientelistic parties increase as the clientelistic parties have extensive networks of local intermediaries.

In addition to parties’ ability to deliver clientelistic benefits, it is expected that the within-country differences in satisfaction are also conditional on a country’s quality of democracy. The extent to which particularistic benefits contribute to citizens’ evaluations of government outputs depends on the relative importance of those benefits. Scholars have argued that in well-established democracies, politicians are more likely to build a credible reputation for delivering programmatic policies, and targeted exchange is thus less attractive (Keefer, 2007). Moreover, in not-fully consolidated democracies, contestation and participation are more restricted (Hafner-Burton et al., 2013; Lindberg, 2009; Schedler, 2002). With lower quality of representation and fewer opportunities for political input, material goods delivered in a clientelistic relationship may become a more important component of citizens’ evaluations of government performance. Therefore, it is hypothesized that clientelistic efforts tend to generate satisfaction among beneficiaries in countries with lower quality of democracy.

Hypothesis 3.3: The within-country differences in democratic satisfaction between voters supporting clientelistic parties and those attached to less clientelistic parties decrease in well-established democracies.

Data and measures

Dependent variable

The main dependent variable is citizens’ satisfaction with democracy. This variable captures the levels of support for how the democratic system works in practice (Canache et al., 2001; Linde and Ekman, 2003) and reflects citizens’ assessments of regime output and opportunities for inputs in the democratic process, rather than their attitudes toward democracy as an ideal regime type. The variable is measured at the individual level using data from public opinion surveys, including Afrobarometer of 2008–2012 (Round 4 and 5, 12 countries), Asianbarometer of 2010 (Wave 3, 8 countries), the European Social Survey (ESS) of 2010 (Round 5, 29 countries), and the Latin American Public Opinion Project (LAPOP) of 2010 (21 countries). In these surveys, respondents were asked “On the whole, how satisfied are you with the way democracy works in [country]?” The wording of the question was consistent in all surveys. This indicator has been used in many studies on regime support (e.g. Aarts and Thomassen, 2008; Anderson and Tverdova, 2003; Anderson and Guillory, 1997; Ezrow and Xezonakis, 2011). In the analysis, responses are coded as a binary variable, for which 0 indicates “not very satisfied or not at all satisfied,” and 1 means “satisfied” or “very satisfied.”¹

Independent and contingent variables

The measures of parties’ linkage strategies and organizational features are from the DALP data set (Kitschelt, 2013). The data include all countries of at least two million inhabitants and with minimally competitive democratic elections. To meet the latter criteria, a country needs to have scored an average of at least 4.0 (partly free) on the Freedom House’s political rights index for 2007 and 2008. For each country, the project contains expert judgments on electoral practices of all politically relevant parties as of the last elections prior to 2009, when the expert surveys were implemented. The project covers 506 parties in 88 countries. For nearly all countries, the data on the independent variables were collected before the dependent variable of citizens’ democratic satisfaction was measured.

The main independent variable, the extent to which parties rely on clientelism, is based on experts’ assessment of the effort parties make in offering five types of benefits to target individuals or small groups of citizens in exchange for votes. The five types of benefits include consumer goods, preferential access to social policy, employment in the public sector, government contracts, and influence over the application of government regulations.² They are the most common types of clientelistic exchanges and constitute a direct measure of clientelism. The expert judgments on the five items are coded on a 1–4 scale, in which “1” refers to negligible or no effort at all, while “4” means a major effort to provide the benefit. The five components are highly correlated with each other (r = 0.75–0.9). Experts’ ratings of the five components for each party are aggregated to generate the party-level uncentered indicator of clientelism.

Recall that Hypothesis 1 suggests a negative relationship between the general prevalence of clientelism in a society and citizens’ average level of satisfaction; while following Hypothesis 2, within countries, supporters of parties that make substantial clientelistic efforts are expected to be more satisfied than other citizens. To test these two hypotheses, country-level average clientelism and parties’ deviation from their country mean are distinguished. For the country-level general clientelistic tendency of the entire party system, the averages of parties’ scores on clientelism in each country, weighted by parties’ vote shares in the most recent two elections before 2009, are computed. In addition, a country-mean centered variable of clientelism, that is, the differences between a party’s score and the weighted national average, is generated to measure the extent to which a party is “more clientelistic” than other parties in the same country. This party-level variable is dubbed exceptional clientelism. Positive (negative) values on exceptional clientelism refer to more (less) clientelistic efforts relative to the national average.

In the public opinion surveys, respondents were asked which party (or presidential candidate) they voted for in the last national election. The individual-level data from public opinion surveys (the dependent variable) and the party-level independent variables are matched with the help of this survey question.³

As specified in Hypothesis 3, the within-country differences in satisfaction between supporters and non-supporters of clientelistic parties are expected to be conditional on parties’ ability to deliver goods and a country’s democratic quality. The indicator of party seat shares in the legislature (when DALP survey was implemented) is utilized to measure parties’ access to resources. It is assumed that larger parties have preferential access to resources essential for clientelistic practices. The other indicator of party’s ability to distribute targeted goods is its local intermediaries. It measures the extent to which parties have local intermediaries such as neighborhood leaders, local notables, or religious leaders who operate in local constituencies on parties’ behalf. The variable is also from DALP, coded on a three-point scale from 0 (no reliance on local intermediaries) to 2 (extensive reliance). The quality of democracy in a country is based on the Polity IV scores when DALP was conducted.

One potential limitation of measures from the DALP is that experts may have less information about certain parties. For robustness checks, we rerun the analysis excluding parties that few experts provided judgments and those that experts had diversified ratings (see Online Appendix 3). Online Appendix 3 also includes more detailed discussion of the survey instrument and analysis utilizing country-level clientelism measures from an alternative source.

Controls

Several controls complement the analyses of the relationship between clientelism and regime evaluations. At the individual level, survey respondents’ demographic variables, including gender, age, and education, are controlled. Consistent with the existing research on democratic satisfaction, individuals’ electoral majority/minority status is included. This variable is based on whether the party they voted for had a cabinet minister (in parliamentary systems) or was the president’s party (in presidential systems) when the survey was conducted. Because the literature shows that citizens’ perception of economic performance affects their levels of satisfaction, respondents’ assessments of their country’s general economic situation are included. Similarly, respondents’ evaluations of household economic situation and their general interests in public affairs are also controlled.⁴

In addition to citizens’ perceptions of economic conditions, at country level, two objective measures of economic performance, the level and growth rate of per capita GDP, are included. The literature suggests that political goods also affect citizens’ regime evaluation (Hofferbert and Klingemann, 1999; Mishler and Rose, 2001a). Countries’ quality of democracy (Polity IV scores) and experience with democracy are therefore controlled. For the latter, the democracy stock measure developed by Gerring et al. (2005) is utilized. Two institutional controls that account for differences between consensual and majoritarian systems are included. The electoral formula variable is coded as majoritarian (coding of 1), mixed (2), or proportional (3) (Reynolds et al., 2005). The indicator of whether a country is a presidential (coded 1) or parliamentary system is also incorporated. These indicators have been suggested to be related to both clientelism (Ames, 1999; Linz and Valenzuela, 1994; Ordeshook, 1995) and regime support.

At party level, in addition to parties’ clientelistic efforts, the extent to which parties rely on programmatic strategies is included. The literature suggests that the quality of programmatic representation is related to regime evaluations (Ezrow and Xezonakis, 2011). It is necessary to control for the possibility that the effects of clientelism are due to the fact that these parties also make substantial programmatic efforts, and it is the reliance on both strategies that especially leads to citizens’ satisfaction. The measure of parties’ programmatic efforts is from Kitschelt and Freeze (2010) based on the DALP data to capture parties’ issue cohesion, salience, and distinctiveness.

Analysis

Figure 1 displays the relationship between clientelism and the level of citizens’ satisfaction with democracy. These two variables are country-level averages. This figure suggests that on average, citizens living in countries where clientelism prevails tend to show a lower level of regime satisfaction. It provides some evidence supporting the correlations proposed in Hypothesis 1. To examine whether the correlation results from other spurious variables or the contrasts between established and new democracies, regression analysis is conducted.

Figure 1.

Clientelism and satisfaction with democracy at country level.

The hypotheses are tested using mixed-effects logistic regression models. The dichotomous dependent variable takes on values 1 (very satisfied or satisfied with the way democracy works) or 0 (not very satisfied or not at all satisfied) and is at the individual level. The country-level average clientelism and party-level (country-mean centered) exceptional clientelism are distinguished. Moreover, several individual-level determinants of citizens’ regime support are controlled. The hierarchical models not only account for potential fixed effects across countries but also allow some coefficients to vary from group to group, though fixed within them. Each model is run with standard errors clustered at the country level to address the issue of heteroskedasticity.⁵

Table 1 presents the results of models estimating the effects of country-level average and party-level exceptional clientelism on citizens’ satisfaction with democracy. Model 1 includes only the country-average measure. The results are consistent with Hypothesis 1: respondents in countries where parties in general more heavily utilize clientelism are less satisfied. Model 2 includes both the country mean and party-level exceptional clientelism measures. The results provide unambiguous evidence in support of Hypotheses 1 and 2. The coefficients on these two variables are in opposite directions: while individuals in countries with generally high levels of clientelism tend to evaluate the performance of the political system negatively, supporters of parties making above (country) average efforts to provide targeted goods show a higher level of satisfaction than other citizens. The negative relationship between country-level clientelism and satisfaction is still robust when the within-country variation of parties’ clientelistic efforts is included.⁶ Model 2 suggests that a 1 SD increase in the country-level clientelism is associated with around 70% lower odds of being satisfied with democracy. A 1 SD increase in the party-level exceptional clientelism corresponds to 8% positive difference in the estimated odds of being satisfied.

Table 1.

Logistic regression analysis of satisfaction with democracy on clientelism.

	(1)	(2)	(3)	(4)	(5)
Dependent Variable (DV)	Satisfaction with democracy
Exceptional clientelism (country-mean centered)		0.0463*** [0.0162]	0.0506*** [0.0183]	0.0934*** [0.0187]	0.0958*** [0.0189]
Governmental party	0.468*** [0.0619]	0.391*** [0.0645]	0.394*** [0.0648]	0.244** [0.0963]	0.226** [0.0977]
Programmaticism			−0.203 [0.372]
Programmaticism × Exceptional Clientelism			−0.0269 [0.0495]
Country-level clientelism (country mean)	−0.0987*** [0.0292]	−0.0975*** [0.0291]	−0.0968*** [0.0284]	−0.0872** [0.0416]	−0.116** [0.0468]
(ln) GDP pc Purchasing Power Parity (PPP)	−0.162 [0.138]	−0.163 [0.139]	−0.124 [0.153]	−0.105 [0.169]	−0.0567 [0.161]
GDP pc growth	−0.00876 [0.0338]	−0.00712 [0.0335]	−0.0105 [0.0330]	0.0952* [0.0519]	0.0810* [0.0485]
Polity	0.00823 [0.0357]	0.00602 [0.0369]	−0.00584 [0.0369]	0.0203 [0.0359]	0.0226 [0.0377]
Democracy stock	0.000677** [0.000318]	0.000685** [0.000311]	0.000842*** [0.000299]	0.0000754 [0.000651]	0.000341 [0.000659]
Electoral formula proportionality				−0.0985 [0.111]	−0.181 [0.140]
Presidentialism				−0.141 [0.250]	0.0447 [0.332]
Education	−0.00273 [0.0149]	−0.00203 [0.0148]	−0.00328 [0.0151]	−0.0610*** [0.0174]	−0.0637*** [0.0175]
Age	0.00253*** [0.000849]	0.00251*** [0.000847]	0.00239*** [0.000878]	0.00578*** [0.00100]	0.00617*** [0.00105]
Female	−0.0383* [0.0218]	−0.0385* [0.0218]	−0.0512** [0.0217]	0.00729 [0.0295]	−0.0108 [0.0290]
Country’s financial condition	0.599*** [0.0328]	0.599*** [0.0328]	0.593*** [0.0335]	0.382*** [0.0302]	0.363*** [0.0276]
Household financial condition				0.174*** [0.0275]	0.169*** [0.0292]
Interests in politics				0.124*** [0.0219]	0.130*** [0.0225]
Constant	1.269 [1.392]	1.335 [1.399]	1.131 [1.444]	0.366 [1.769]	0.526 [1.798]
Country-level variance	0.213	0.215	0.190	0.261	0.262
Party-level variance	0.143	0.133	0.132	0.111	0.103
N (individual/party/country)	62,975/333/69	62,975/333/69	59,826/320/66	30,197/137/39	27,821/130/37
Akaike information criterion	70,071	70,062	66,598	35,822	33,119
Log likelihood	−35022.5	−35017.2	−33283.2	−17892.9	−16541.6

Note: Clustered standard errors in brackets. The individual-level controls of household financial condition and interests in politics are not available for countries covered by the ESS. In models 4 and 5, the magnitude of the coefficient on exceptional clientelism measure increases. This partly reflects the pattern that the within-country positive association is stronger in lower quality democracies. ESS: the European Social Survey; pc: per capita.

*p < 0.1; **p < 0.05; ***p < 0.01.

Figure 2 shows the predicted probabilities of democratic satisfaction against the party-level exceptional clientelism for different levels of country-average clientelism based on Model 2. The solid, dotted, and dashed lines represent the relationships between parties’ reliance on clientelism and their supporters’ regime evaluation when the country averages of clientelistic effort are fixed at the first quartile, median, and third quartile, respectively. The points where they cross the line x = 0 suggest that in countries where parties on average more heavily utilize clientelism, citizens are less satisfied. Compared to the national average, if a party makes less clientelistic effort, their supporters tend to evaluate the regime more negatively than those who support clientelistic parties. In general, the between-country effects are larger than the within-country differences.

Figure 2.

Country-average clientelism, exceptional clientelism, and democratic satisfaction.

Model 3 includes a measure of parties’ reliance on programmatic strategies and the interaction with parties’ clientelistic strategies. The results show that the within-country positive effects of clientelism are not contingent on whether parties also make programmatic promises. Supporters of parties that combine both strategies do not show higher levels of satisfaction. Model 4 includes additional controls for countries’ institutional features and individual attitudes. Model 5 excludes advanced democracies in Western Europe and North America to examine whether the negative effects of country-level clientelism merely result from the contrast between them and developing countries. The results still prevail in these models.⁷

To test Hypothesis 3 that within-country differences in democratic satisfaction between supporters of clientelistic and non-clientelistic parties are contingent on parties’ organizational features and countries’ quality of democracy, we include several interaction terms in the analysis. The results are presented in Table 2. Model 1 includes parties’ seat shares as the measure of parties’ access to resources and capacity to distribute goods. The coefficient on the interaction is positive and significantly different from zero. Consistent with Hypothesis 3.1, the results suggest that compared to other citizens supporters of clientelistic parties are particularly satisfied if these parties are large. The within-country differences are attenuated when parties are smaller. The marginal effect of supporting an “exceptionally” clientelistic party is displayed in the upper panel of Figure 3. The solid curve represents the change in the predicted probability of the dependent variable when the exceptional clientelism indicator increases from 0 to 1 SD higher, across different party sizes. When a party’s seat share is above 20%, the marginal effect of its relatively clientelistic effort becomes significantly greater than zero.

Table 2.

Logistic regression analysis of the contingent effect of clientelism on satisfaction with democracy.

	(1)	(2)	(3)
DV	Satisfaction with democracy
Exceptional clientelism (country-mean centered)	−0.0250 [0.0256]	−0.0882** [0.0366]	0.139*** [0.0310]
Governmental party	0.348*** [0.0622]	0.362*** [0.0658]	0.392*** [0.0622]
Seat share	0.00495* [0.00273]
Reliance on local intermediaries		0.166* [0.0867]
Exceptional Clientelism × Seat share	0.00384*** [0.00111]
Exceptional Clientelism × Local Intermediaries		0.105*** [0.0263]
Exceptional Clientelism × Polity			−0.0122*** [0.00380]
Country-level clientelism (country mean)	−0.108*** [0.0300]	−0.105*** [0.0301]	−0.0962*** [0.0292]
(ln) GDP pc PPP	−0.166 [0.147]	−0.176 [0.146]	−0.167 [0.140]
GDP pc growth	−0.00832 [0.0367]	−0.00879 [0.0362]	−0.00727 [0.0358]
Polity	0.0105 [0.0379]	0.00905 [0.0383]	−0.0000311 [0.0404]
Democracy stock	0.000584* [0.000330]	0.000660** [0.000311]	0.000693** [0.000312]
Electoral formula proportionality	−0.00895 [0.0818]	−0.0138 [0.0782]	−0.0209 [0.0781]
Presidentialism	−0.0526 [0.144]	−0.0460 [0.135]	−0.0350 [0.137]
Education	−0.00210 [0.0149]	−0.00191 [0.0148]	−0.00220 [0.0149]
Age	0.00249*** [0.000846]	0.00250*** [0.000845]	0.00252*** [0.000847]
Female	−0.0387* [0.0218]	−0.0386* [0.0218]	−0.0386* [0.0218]
Country’s financial condition	0.599*** [0.0328]	0.599*** [0.0328]	0.599*** [0.0328]
Constant	1.371 [1.487]	1.346 [1.497]	1.459 [1.439]
Country-level variance	0.225	0.224	0.225
Party-level variance	0.119	0.123	0.126
N (individual/party/country)	62,975/333/69	62,975/333/69	62,975/333/69
Akaike information criterion	70,053	70,058	70,061
Log likelihood	−35008.5	−35011.1	−35013.7

Note: Clustered standard errors in brackets. pc: per capita.

*p < 0.10; **p < 0.05; ***p < 0.01.

Figure 3.

Marginal effect of exceptional clientelism.

Model 2 in Table 2 includes the interaction variable for parties’ local networks and exceptional clientelism. The middle panel of Figure 3 shows the estimated marginal effect. Consistent with Hypothesis 3.2, the positive association between supporting exceptionally clientelistic parties and democratic satisfaction is especially pronounced when these parties can rely on local notables as their brokers. Supporters of clientelistic parties with extensive linkage associations are more likely to evaluate the political system positively.

Model 3 includes the cross-level interaction between countries’ quality of democratic governance and parties’ clientelistic efforts relative to their national average. The estimated marginal effect of party-level exceptional clientelism across different Polity IV scores is displayed in the lower panel of Figure 3. The results are consistent with Hypothesis 3.3: the differences in satisfaction between supporters of clientelistic parties and other citizens are larger in lower quality democracies. Targeted benefits are less likely to have effects on regime evaluation in fully established democracies.⁸

Conclusion

Building on studies that document the negative effects of clientelism on democratic principles, this article demonstrates a systematic analysis of the relationship between parties’ reliance on clientelistic strategies and citizens’ evaluations of regime performance. The results suggest that citizens on average show a lower level of satisfaction with democracy in countries where parties generally rely more heavily on clientelism to attract votes. However, the association between clientelism and satisfaction is not uniform within a country. Supporters of parties that substantially engage in clientelism are more satisfied than fellow citizens who support non-clientelistic parties. The within-country differences are strengthened when the parties are large, have extensive external linkages, or in countries with lower democratic quality.

The results have several implications. First, consistent with normative democratic concerns, citizens tend to evaluate a political system more negatively if clientelism is widespread. Our results further suggest that it is supporters of non-clientelistic parties who are particularly dissatisfied. Second, the results that within-country differences in satisfaction are strengthened in less established democracies are consonant with previous literature, suggesting that clientelistic policies are especially attractive in young democracies. Finally, negative evaluations of regime performance do not directly lead to actions that challenge democracy. Further studies are required to identify whether democracies where clientelism prevails tend to be more vulnerable.

Supplemental material

Supplemental Material, sj-docx-1-ppq-10.1177_1354068818784965 - Clientelistic parties and satisfaction with democracy

Supplemental Material, sj-docx-1-ppq-10.1177_1354068818784965 for Clientelistic parties and satisfaction with democracy by Yi-Ting Wang in Party Politics

Footnotes

Acknowledgements

The author wishes to thank Herbert Kitschelt and the participants of the 2011 Workshop on Democratic Accountability Strategies at Duke University and 2013 Quality of Government Conference at the University of Gothenburg for their comments on earlier versions of this manuscript.

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 received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Yi-Ting Wang

Supplemental material

Supplemental material for this article is available online.

Notes

References

Aarts

Thomassen

(2008) Satisfaction with democracy: do institutions matter? Electoral Studies 27(1): 5–18.

Ames

(1999) Electoral rules, constituency pressures, and pork barrel: bases of voting in the Brazilian congress. Journal of Politics 57(2): 324–343.

Anderson

(2002) Good questions, dubious inferences, and bad solutions: some further thoughts on satisfaction with democracy. Binghamton University, Center on Democratic Performance Working Paper.

Anderson

Guillory

(1997) Political institutions and satisfaction with democracy. American Political Science Review 91(1): 66–81.

Anderson

Tverdova

(2003) Corruption, political allegiances, and attitudes toward government in contemporary democracies. American Journal of Political Science 47(1): 91–109.

Anderson

Blais

Bowler

, et al. (2005). Losers’ Consent: Elections and Democratic Legitimacy. Oxford: Oxford University Press.

Anduiza

Gallego

Muñoz

(2013) Turning a blind eye: experimental evidence of partisan bias in attitudes toward corruption. Comparative Political Studies 46(12): 1664–1692.

Bernauer

Vatter

(2012) Can’t get no satisfaction with the Westminster model? European Journal of Political Research 51(4): 435–468.

Birch

(2008) Electoral institutions and popular confidence in electoral processes. Electoral Studies 27(2): 305–320.

10.

Blais

Gélineau

(2007) Winning, losing and satisfaction with democracy. Political Studies 55(2): 425–441.

11.

Blais

Loewen

(2007) Electoral systems and evaluations of democracy. In: Cross

(ed) Democratic Reform in New Brunswick. Toronto: Canadian Scholars Press, pp. 39–57.

12.

Blais

Morin-Chasse

Singh

(2015) Election outcomes, legislative representation, and satisfaction with democracy. Party Politics 22(1): 1–15. DOI: 10.1177/1354068815583200

13.

Booth

Seligson

(2009) The Legitimacy Puzzle in Latin America. Cambridge: Cambridge University Press.

14.

Bratton

Mattes

(2001) Support for democracy in Africa: intrinsic or instrumental? British Journal of Political Science 31(03): 447–474.

15.

Brunell

Buchler

(2009) Ideological representation and competitive congressional elections. Electoral Studies 28(3): 448–457.

16.

Calvo

Murillo

(2013) When parties meet voters: assessing political linkages through partisan networks and distributive expectations in Argentina and Chile. Comparative Political Studies 46(7): 851–882.

17.

Cameron

Miller

(2015) A practitioner’s guide to cluster-robust inference. Journal of Human Resources 50(2): 317–372.

18.

Canache

Allison

(2005) Perceptions of political corruption in Latin American democracies. Latin American Politics and Society 47(3): 91–111.

19.

Canache

Mondak

Seligson

(2001) Meaning and measurement in cross-national research on satisfaction with democracy. Public Opinion Quarterly 65(4): 506–528.

20.

Clarke

Dutt

Kornberg

(1993) The political economy of attitudes toward polity and society in Western European democracies. The Journal of Politics 55(04): 998–1021.

21.

Criado

Herreros

(2007) Political support: taking into account the institutional context. Comparative Political Studies 40(12): 1511–1532.

22.

Curini

Jou

Memoli

(2011) Satisfaction with democracy and the winner/loser debate. British Journal of Political Science 42(02): 241–261.

23.

Dahlberg

Holmberg

(2014) Democracy and bureaucracy: how their quality matters for popular satisfaction. West European Politics 37(3): 515–537.

24.

Delgado

(2016) How governing experience conditions winner-loser effects. Electoral Studies 44: 76–84.

25.

Easton

(1975) A re-assessment of the concept of political support. British Journal of Political Science 5(4): 435–457.

26.

Esaiasson

Gilljam

Persson

(2012) Which decision-making arrangements generate the strongest legitimacy beliefs? European Journal of Political Research 51(6): 785–808.

27.

Evans

Whitefield

(1995) The politics and economics of democratic commitment: support for democracy in transition societies. British Journal of Political Science 25(04): 485–514.

28.

Ezrow

Xezonakis

(2011) Citizen satisfaction with democracy and parties’ policy offerings. Comparative Political Studies 44(9): 1152–1178.

29.

Ferland

(2015) A rational or a virtuous citizenry?–The asymmetric impact of biases in votes-seats translation on citizens’ satisfaction with democracy. Electoral Studies 40: 394–408.

30.

Finkel

Muller

Seligson

(1989) Economic crisis, incumbent performance and regime support. British Journal of Political Science 19(03): 329–351.

31.

Fox

(1994) The difficult transition from clientelism to citizenship. World Politics 46(2): 151–184.

32.

Kitschelt

Freeze

(2010) Programmatic party system structuration. Presented at the Annual Meeting of the American Political Science Association, Washington, DC, 2–5 September 2010.

33.

Gelman

(2007) Struggles with survey weighting and regression modeling. Statistical Science 22(2): 153–164.

34.

Gerring

Bond

Barndt

, et al. (2005) Democracy and economic growth. World Politics 57(3): 323–364.

35.

Hafner-Burton

Hyde

Jablonski

(2013) When do governments resort to election violence? British Journal of Political Science 44(1): 149–179.

36.

Hicken

(2011) Clientelism. Annual Review of Political Science 14(1): 289–310.

37.

Hicken

Simmons

(2008) The personal vote and the efficacy of education spending. American Journal of Political Science 52(1): 109–124.

38.

Hofferbert

Klingemann

(1999) Remembering the bad old days: human rights, economic conditions, and democratic performance in transitional regimes. European Journal of Political Research 36(2): 155–174.

39.

Howell

Justwan

(2013) Nail-biters and no-contests: the effect of electoral margins on satisfaction with democracy in winners and losers. Electoral Studies 32(2): 334–343.

40.

Keefer

(2007) Clientelism, credibility, and the policy choices of young democracies. American Journal of Political Science 51(4): 804–821.

41.

Kim

(2009) Cross-national analyses of satisfaction with democracy and ideological congruence. Journal of Elections, Public Opinion and Parties 19(1): 49–72.

42.

Kitschelt

(2013) Democratic Accountability and Linkages Project. Durham: Duke University. Available at: http://sites.duke.edu/democracylinkage/ (accessed 25 October 2013).

43.

Kitschelt

Kselman

(2011) Organizational extensiveness and political clientelism. Presented at the Annual Meeting of the American Political Science Association, Seattle, WA, 1–4 September 2011.

44.

Kitschelt

Wilkinson

(2007) Patrons, Clients, and Policies. Cambridge: Cambridge University Press.

45.

Klingemann

(1999) Mapping political support in the 1990s. In: Norris

(ed) Critical Citizens: Global Support for Democratic Governance. Oxford: Oxford University Press.

46.

Koter

(2013) King makers: local leaders and ethnic politics in Africa. World Politics 65(02): 187–232.

47.

Krehbiel

Cropanzano

(2000) Procedural justice, outcome favorability and emotion. Social Justice Research 13(4): 339–360.

48.

Lindberg

(2009) Democratization by elections: a mixed record. Journal of Democracy 20(3): 86–92.

49.

Lindberg

(2010) What accountability pressures do MPs in Africa face and how do they respond? The Journal of Modern African Studies 48(01): 117–142.

50.

Linde

(2012) Why feed the hand that bites you? Perceptions of procedural fairness and system support in post-communist democracies. European Journal of Political Research 51(3): 410–434.

51.

Linde

Ekman

(2003) Satisfaction with democracy: a note on a frequently used indicator in comparative politics. European Journal of Political Research 42(3): 391–408.

52.

Linde

Erlingsson

GÓ

(2013) The eroding effect of corruption on system support in Sweden. Governance 26(4): 585–603.

53.

Linz

Valenzuela

(eds) (1994) The Failure of Presidential Democracy. Baltimore: Johns Hopkins University Press.

54.

Lühiste

(2014) Social protection and satisfaction with democracy: a multi-level analysis. Political Studies 62(4): 784–803.

55.

Lust-Okar

(2009) Competitive clientelism in the Middle East. Journal of Democracy 20(3): 122–135.

56.

MacCoun

(2005) Voice, control, and belonging: the double-edged sword of procedural fairness. Annual Review of Law and Social Science 1(1): 171–201.

57.

Magalhães

(2016) Economic evaluations, procedural fairness, and satisfaction with democracy. Political Research Quarterly 69(3): 522–534.

58.

Magaloni

Diaz-Cayeros

Estevez

(2012) Strategies for Vote Buying: Democracy, Clientelism, and Poverty Relief in Mexico. Cambridge: Cambridge University Press.

59.

Manzetti

Wilson

(2007) Why do corrupt governments maintain public support? Comparative Political Studies 40(8): 949–970.

60.

Mishler

Rose

(2001a) Political support for incomplete democracies: realist vs. idealist theories and measures. International Political Science Review 22(4): 303–320.

61.

Mishler

Rose

(2001b) What are the origins of political trust? Testing institutional and cultural theories in post-communist societies. Comparative Political Studies 34(1): 30–62.

62.

Morris

Klesner

(2010) Corruption and trust: theoretical considerations and evidence from Mexico. Comparative Political Studies 43(10): 1258–1285.

63.

Norris

(ed) (1999) Critical Citizens: Global Support for Democratic Governance. Oxford: Oxford University Press.

64.

Ordeshook

(1995) Institutions and incentives. Journal of Democracy 6(2): 46–60.

65.

Reynolds

Reilly

Ellis

(eds) (2005) Electoral System Design: The New International IDEA Handbook. Stockholm: International IDEA.

66.

Rothstein

(2009) Creating political legitimacy: electoral democracy versus quality of government. American Behavioral Scientist 53(3): 311–330.

67.

Schedler

(2002) The menu of manipulation. Journal of Democracy 13(2): 36–50.

68.

Seligson

(2002) The impact of corruption on regime legitimacy. Journal of Politics 64(2): 408–433.

69.

Seligson

(2006) The measurement and impact of corruption victimization: survey evidence from Latin America. World Development 34(2): 381–404.

70.

Singh

(2014) Not all election winners are equal: satisfaction with democracy and the nature of the vote. European Journal of Political Research 53(2): 308–327.

71.

Singh

Karakoç

Blais

(2012) Differentiating winners: how elections affect satisfaction with democracy. Electoral Studies 31(1): 201–211.

72.

Stecker

Tausendpfund

(2016) Multidimensional government-citizen congruence and satisfaction with democracy. European Journal of Political Research 55(3): 492–511.

73.

Stokes

(2005) Perverse accountability: a formal model of machine politics with evidence from Argentina. American Political Science Review 99(3): 315–325.

74.

Stokes

(2007) Political clientelism. In: Boix

Stokes

(eds) Oxford Handbook of Comparative Politics. Oxford: Oxford University Press, pp. 604–627.

75.

Stokes

Dunning

Nazareno

, et al. (2013). Brokers, Voters, and Clientelism. London: Cambridge University Press.

76.

Szwarcberg

(2014) Political parties and rallies in Latin America. Party Politics 20(3): 456–466.

77.

Tavits

(2008) Representation, corruption, and subjective well-being. Comparative Political Studies 41(12): 1607–1630.

78.

Tyler

(2006) Psychological perspectives on legitimacy and legitimation. Annual Review of Psychology 57(1): 375–400.

79.

Weiss

Suckow

Cropanzano

(1999) Effects of justice conditions on discrete emotions. Journal of Applied Psychology 84(5): 786–794.

80.

Weitz-Shapiro

(2007) The local connection: local government performance and satisfaction with democracy in Argentina. Comparative Political Studies 41(3): 285–308.

81.

Winters

Weitz-Shapiro

(2013) Lacking information or condoning corruption: when do voters support corrupt politicians? Comparative Politics 45(4): 418–436.

82.

Zechmeister

Zizumbo-Colunga

(2013) The varying political toll of concerns about corruption in good versus bad economic times. Comparative Political Studies 46(10): 1190–1218.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB