Party system institutionalization,partisan affect,and satisfaction with democracy

Abstract

Citizens’ attitudes about the political parties in their countries have been linked to their overall satisfaction with their democracy, with those feeling great love (hate) for parties feeling more (less) satisfied with the democracy. Such strong positive and negative emotions require time and clear targets to form. This study demonstrates that the influence of interparty affect is greater where the party system has institutionalized. Where the public can be familiar with the parties, their positions, and their relative status in the party system, citizens’ attitudes toward the democracy are more informed by their feelings about the parties in the system. This article draws on 20 years of surveys from the Comparative Study of Electoral Systems. It demonstrates that the positive effect of positive party affect is greater in institutionalized party systems, while the negative effect of negative party affect is not. This article thus contributes to political science’s understanding of affective polarization, the benefits of party system institutionalization, and popular democratic commitment.

Keywords

Public opinion democracy partisanship affective polarization satisfaction with democracy

Introduction

Around the world, voters tend to be fond of the parties they support in elections; they like them emotionally as well as ideologically. In many countries, this in-group affection is coupled with negative sentiments towards other political parties (Boxell et al., 2020; Iyengar and Westwood 2015). Researchers have expressed concern that out-party animosity is increasing, leading to stark differences between in-party and out-party sentiment for some voters (Wagner 2020; Westwood et al., 2018). While loving one of the players can facilitate love for the game, these negative partisans are less prone to democratic satisfaction (Ridge 2020). Thus, strong and bourgeoning emotions can present both a boon and a challenge to democratic stability.

This project introduces a moderating element into the relationship between partisan affect and democratic attitudes: party system institutionalization. Party system institutionalization is the development of consistent parties, positions, and interactions in a country (Mainwaring et al. 2018; Chiaramonte and Emanuele 2017; Piñeiro Rodríguez and Rosenblatt 2020). Voters must know the parties in order to love them or to hate them. It takes time and stability for party brands to form and disseminate. If the parties frequently change, citizens may not form strong attachments or antipathies, and they may not connect those emotions as readily to the system as a whole. An institutionalized party system facilitates citizens’ development of party affect and clarifies the connection the parties in the system have to each other and to the regime as a whole. In institutionalized systems, citizens can form these attachments, and these attachments can drive their attitudes towards their democracy. Thus, the relationship between party affect and democratic institutional satisfaction concentrates in more institutionalized systems.

This article evaluates that proposition using the Integrated Module Dataset of the Comparative Study of Electoral Systems (CSES), a series of post-election surveys conducted over 20 years. It finds that the positive effect of positive party affect grows with party system institutionalization. The effect of negative party affect though is not conditioned by system institutionalization. The story of institutionalization and satisfaction with democracy is more a love story than a hate story. The heightened influence of affect with institutionalization is greatest among those responding emotionally to the largest parties in their system. The article concludes by considering the implications of this research for studies of affective polarization and for democratic stability.

Love, hate, and democracy

Voters have shown themselves ready to love and to hate the political parties in their democracies. Positive party affect (PPA) is an expression of citizens’ positive sentiments for their political parties; this is a party they like. Negative party affect (NPA) is the opposite. It is when citizens hate a political party in their democracy.¹ Partisan affect is related to but distinct from partisanship, which is party membership or self-identification with a party (Russo and Mayer 2021). It is the like-dislike – or rather the love-hate – dimension.

Emotional investment in parties is more feasible when parties have stable identities (Thames and Robbins 2007). This partisan affect, positive and negative, has been linked to myriad political behaviors. For instance, positive connections to parties influence electoral turnout, vote choice, volunteering with a party, and signing petitions (Caruana et al., 2015; Medeiros and Noël 2014; Vlachová 2001; McGregor et al. 2015). Negative connections with a party have been effect party membership, voter turnout and vote choice, joining a protest, and signing petitions (Caruana et al., 2015; Mayer 2017; Medeiros and Noël 2014; Vlachová 2001; Çakır 2021).

Researchers have been concerned about the consequences of these positive and negative emotions. Attitudes towards parties can be marshalled against the members of the liked and disliked parties. For instance, people favor members of their own party for jobs and scholarships, are more likely to show generosity to in-partisans, and prefer to spend time with co-partisans, while being less willing to have out-partisans marry into their families (Sheffer 2020; Iyengar and Westwood 2015; Iyengar et al. 2012; Sheffer 2020; Knudsen 2021).² In terms of political consequences, strong negative sentiment towards political parties has been proposed as a threat to democratic commitment.³ Ridge (2020) identifies a negative relationship between NPA and citizens’ satisfaction with democracy.

Conversely, PPA may support democracy. Russo and Mayer (2021) argue that individuals who identify with a party – positive partisans – develop positive affect (PPA) towards it. It is part of the party institutionalization process (Bolleyer and Ruth 2018). Individuals who like the parties in their democracy transfer that positive sentiment to the system itself. Converse (1969) argues that citizens who are attached to parties have cognitively bought into their democratic system and will work to keep their party structures and the electoral systems in which they operate in place. Thus, it is reasonable that, in line with the positive partisans, citizens holding PPA will also feel more satisfied with their democracies.

For citizens to develop party affect and for their party affect to influence their democratic attitudes, citizens need to be able to recognize parties and the parties need to stick around long enough to be loved or hated. This is where party system institutionalization comes into the story: “An institutionalized party system is one in which a stable set of parties interact regularly in stable ways” (Mainwaring 2016, 692). Actors can develop their political expectations and behaviors within that party framework. These parties are able to include new demands and societal interests (adapting) without obliterating the extant party structure (stability) (Piñeiro Rodríguez and Rosenblatt 2020). Party system institutionalization includes regularized party competition of organized parties with “stable roots in society” (Kuenzi and Lambright 2001, 441).⁴ These processes take time (Mainwaring et al. 2018).

Globally, party system institutionalization is the “exception rather than the norm” (Mainwaring 2016, 693). It has been studied in Africa (Kuenzi and Lambright 2001), Asia (Hicken and Kuhonta 2015), Europe (Chiaramonte and Emanuele 2017, 2019), Latin America (Bolleyer and Ruth 2018; Piñeiro Rodríguez and Rosenblatt 2020; Mainwaring and Scully 1995), post-Communist countries (Bielasiak 2002), and cross regionally (Thames and Robbins 2007; Mainwaring and Torcal 2006). Rates of institutionalization vary across and within regions. In some places, voters are becoming more party-embedded, including reducing ticket-splitting and evincing strong emotions about the parties (Abramowitz and Webster 2016). In other states, party membership rates are declining (Van Biezen, Mair, and Poguntke 2012; Scarrow and Gezgor 2010). In those contexts, researchers fear de-institutionalization, with voters being less loyal and parties becoming less durable (Chiaramonte and Emanuele 2017, 2019; Piñeiro Rodríguez and Rosenblatt 2020). Chiaramonte and Emanuele (2019, 1) postulate that scholars have so “taken for granted for a long time” the “stability and predictability in the patterns of inter-party competition” in Western Europe that they have failed to recognize the variations in that interaction.

Nascent parties and party systems are fragile and may dissolve or evolve, and it takes time for citizens’ “habituation” or “socialization” with that structure to occur (Converse 1969, 140; see also Dalton and Weldon 2007). That said, institutionalization is not merely a function of age. Systems that are old but in which parties never change positions or incorporate new dimensions are “ossified” (Piñeiro Rodríguez and Rosenblatt 2020, 253). Evolving systems can reflect fluctuations in public will, as-yet-unknown party brands, diluted party brands,⁵ or changing electoral rules that require parties to break down or merge (Bielasiak 2002; Mainwaring 2016). Institutionalized party systems allow parties to age, growing roots in the political landscape, without stagnating.

Party system institutionalization has been linked to more democratic outcomes. For instance, it increases the accountability, effectiveness, and legitimacy of governments (Mainwaring and Torcal 2006; Thames and Robbins 2007; Piñeiro Rodríguez and Rosenblatt 2020). Citizens are better able to punish bad politicians when they know who they are and how to vote against them. Party system institutionalization makes the system more intelligible to the electorate by encouraging “strong party brands” with clear programmatic differences (Mainwaring 2016, 699). Institutionalization is positively correlated with consolidated democracy (Thames and Robbins 2007; Mainwaring and Scully 1995; Kuenzi and Lambright 2001). Durable and responsive party systems are seen as necessary for democratic institutionalization (Mainwaring et al. 2018; Piñeiro Rodríguez and Rosenblatt 2020; Converse 1969). Conversely, low levels of institutionalization mean systems are more susceptible to capture by charismatic and anti-party politicians, who may break the rules or undermine democracy (Mainwaring and Torcal 2006; Mainwaring and Scully 1995). These findings would be consistent with a positive direct effect of institutionalization on democratic satisfaction, since they tell the story of a functioning democracy. Research has already shown that satisfaction with democracy is lower among citizens who cannot tell the parties in their system apart (Ridge 2021b).

This study demonstrates the role of institutionalization in the above story of partisan affect and democratic satisfaction. Where party systems are not institutionalized, parties may be weak or personalistic (Mainwaring and Scully 1995). If the parties are in flux, continually forming or shifting, then individuals would be less able to form strong emotions about them – or at least emotions that relate to a truth about the parties. They would as well have less reason to form opinions about the system as a whole based on their attitudes towards the present parties. Afterall, those parties seem to be short-lived. The learning process about parties is faster in the older democracies, probably due to their relatively greater party stability (Dalton and Weldon 2007). In more institutionalized party systems, respondents will have a clearer sense of who the players are. Citizens will have observed how the parties interact with each other and the democratic institutions, and the parties will have histories that are hard to escape. These histories could even follow them through re-branding attempts. In older party systems, then, voters are more likely to have formed emotional attachments to some parties and potential emotional opposition to other parties. The sense that these parties have been able to participate and draw supporters over time as well as an awareness of the role they play in the state – whether or not they are ever in power – would then more reasonably transfer into the citizens’ attitudes towards their democracy. Afterall, their democracy as it is functioning is now linked to those parties, their participation, and their participants. In the institutionalized system, in which partisan affect is easier to develop, and in which loved and hated parties are repeat players, the effect of affect would then be heightened.

The combined benefit of party system age and stabilization in terms of forming stronger party attitudes and the reasonableness of transferring those attitudes to the system as a whole increasing with time leads to the central prediction of this article. Namely, the influence of partisan affect on citizens’ satisfaction with democracy will become stronger as party systems congeal (H1).⁶ Broken down by positive and negative partisan affect, this would look like a stronger positive effect of PPA with institutionalization (H1a) and a more negative effect of NPA with institutionalization (H1b).

An additional factor can also be considered – party magnitude. The relationship Ridge (2020) identifies between NPA and satisfaction with democracy was heavily determined by the political magnitude of the party involved. Hating the largest parties in the system is significantly detrimental to system satisfaction, but hating a minor party is not significantly related to satisfaction with the democracy. This nuance has not yet been introduced into the examination of positive party affect. It is included in this study.

The largest parties in a system have the greatest potential for durability; they would be the fixtures of an institutionalized party system. They are also objects of great love and support, as well as disaffection (Appendix 1). As Ridge (2020) argues, affect towards parties that represent a larger share of the population has the greatest direct effect on citizens’ satisfaction with their democracies; “if parties with “a ‘majority bent’ are rejected, this can cause polarization” and make alternations in government “impossible” (Rose and Mishler 1989, 226). Strong sentiments towards smaller or niche parties are less likely to make citizens question changeovers in power, the nature of the electorate, or their institutional structure more broadly. Sentiments towards the major parties would be harder to shake, particularly as they cement, and the major parties would be harder to rebrand. Voters may also have lesser expectations that minor parties will play a sizeable political role (Ridge 2020). Furthermore, these small parties, by virtue of their smallness, may not be as long-lived, even in a system that experiences greater overall stability. Thus, it is conceivable that the amplifying effect of institutionalization on affect is linked to the major parties. That is to say, a positive effect of PPA towards a major party (H2a) and a negative effect of NPA towards a major party (H2b) becomes more potent with institutionalization. This argument does not preclude a relationship between minor party affect and institutionalization; however, it does not make a specific prediction about them.

Materials and methods

To examine these relationships, this study relies on the Integrated Module Dataset (IMD) from the Comparative Study of Electoral Systems (CSES). The CSES is a repeat series of post-election surveys conducted by countries around the world that has been fielded since 1996. The IMD includes questions that have been asked in multiple survey modules between 1996 and 2016. The dataset offers, through the number of countries included and the length of time, a sufficient variation in party system institutionalization for this study. Affective polarization is highly salient in the time close to an election (Hernández, Anduiza, and Rico 2021; Sheffer 2020). Thus, the CSES IMD is well-suited to analyze the relationship between interparty affect and democratic satisfaction.

For these tests, the operationalization of positive and negative affect is necessary. The CSES asks respondents to rate the political parties in their country from zero (dislike) to ten (like). Inquiries were made about up to nine parties per country, depending on the country. Priority was given to the largest parties in the system. Like/dislike rating scales are common metrics for measuring party affect (Druckman and Levendusky 2019; Boxell et al., 2020; Reiljan 2020; Wagner 2020). Respondents are deemed to have negative affect towards a party if they rate one of the parties four or lower. Four was chosen is for consistency with the previous literature (Mayer 2017; Ridge 2020). Reciprocally, they are deemed to have a positive affect to a party if they rate one of the parties six or higher. Five, the midpoint on the scale, is treated as neutral. These are binary indicators.

Affect towards a major or minor party is determined by the same question. Respondents who have positive (negative) affect toward one of the two largest parties in their system, based on the parties’ vote share in that election, are deemed to have positive (negative) affect toward a major party. Having positive (negative) affect toward any of the smaller parties is coded as positive (negative) affect towards a minor party. These variables are included as a factor variable with a reference category of no positive (negative) affect. The frequency of each classification is varied across the country samples (Appendix 1).

The outcome of interest in this study is the respondent’s satisfaction with the functioning of the democracy in his country. Survey takers were asked, “On the whole, are you very satisfied, fairly satisfied, not very satisfied, or not at all satisfied with the way democracy works in [this country]” (CSES IMD). Responses are coded from one to four: not at all satisfied, not very satisfied, fairly satisfied, and very satisfied. This question targets how democracy works in practice in that country – specific support – rather than support for democracy in general – diffuse support (Linde and Ekman 2003; Anderson 2002). Democratic satisfaction is “located between diffuse notions of support for democratic principles and specific attitudes toward political actors” (Blais et al. 2017, 86). Specific support, however, can be linked to diffuse support (Linde and Peters 2020; Ridge, 2021a).

Some scholars have expressed concern about the measure’s correlation with other indicators, like regime support, in countries that are just becoming democracies (Canache et al. 2001). How can citizens meaningfully evaluate the functioning of a democracy in a regime that is not a democracy? Furthermore, party system institutionalization does not mean the same thing in democracies as in autocracies, where regularization is a function of the state control, rather than public will (Mainwaring 2016); Mainwaring and Scully (1995) go so far as to require democratic elections in their system of identifying institutionalized party systems. To address these concerns, the study follows Thames and Robbins (2007) in only including countries rated six or higher on the Polity scale in the sample.⁷

Party system institutionalization must also be measured. There is not a single metric for the formalization of a country’s party system. The measurements target stability – often associated with the number of years the parties exist. Kuenzi and Lambright (2001) note that age indicates the parties have put down roots in the system; they weight it by electoral volatility and the quality of the country’s democracy for their study. Others weight age by barriers to party formation and voter turnout (Piñeiro Rodríguez and Rosenblatt 2020). Electoral volatility and cumulative volatility have also been used (Bielasiak 2002; Mainwaring et al. 2018; Thames and Robbins 2007); these studies look at “the net change from one election to the next for each party that participated in both elections and dividing this by the sum of the total vote for each party in both elections” (Thames and Robbins 2007, 5). Bielasiak (2002) also considers the effective number of political parties. This method is limited because more parties does not mean the system is necessarily more or less solidified. Ideological inconsistency can proxy for uninstitutionalization (Mainwaring et al. 2018). Fundamentally, there are “two dimensions of party system institutionalization” empirically, the “stability of interparty competition and the depth of party roots (or anchoring) in society” (Mainwaring and Torcal 2006, 204).

Three measurements for party system institutionalization tapping these dimensions are brought to bear here.⁸ Table 1 shows their distribution.⁹ The first measurement reports the age of the party system. As noted, the party system age indicates the “parties’ rootedness in society” (Chiaramonte and Emanuele 2017, 377) and that it has “captured the long-term loyalties of some social groups” (Mainwaring and Scully 1995, 13). This variable shows the length of time during which citizens were able to form their affective response and to live politically with it. To address the elements of stability and volatility, it incorporates the turnover in major parties in the system. The variable measures “the average of the ages of the first government party (1GOVAGE), second government party (2GOVAGE), and first opposition party (1OPPAGE), or the subset of these for which age of party is known” (Cruz et al. 2021). It thus shows how long the major players have been in the game. It ranges from 4.33 (South Korea 2000) to 161 (USA, 2012). The measurement is included in centuries for model convergence and comparability to the other metrics (0.043–1.61). While this variable does not account for change up in all the parties in the system, it accounts for the stability among the most prominent government and opposition parties –institutionalization among the major players.

Table 1.

Party system institutionalization summary statistics.

Measurement	Mean	Median	Standard deviation
Party system age	0.559	0.513	0.369
Party institutionalization index	0.887	0.922	0.145
Party system institutionalization	0.93	0.94	0.031

The second metric is Chiaramonte and Emanuele’s (2019) party system institutionalization score (PSI). Chiaramonte and Emanuele target the routinization and stabilization of the interaction between parties in a system. PSI combines electoral volatility – the shifting of votes among parties – and party system regeneration – the replacement of old parties by new parties – to measure system institutionalization. It shows the continuity of the players and their relative status. Chiaramonte and Emanuele only assign ratings to 19 countries in Western Europe, only some of which participate in the CSES. The models using the PSI variable thus only include a subset of election-years from the CSES IMD. The score ranges from 0 to 100 in theory, but in practice the scores are more constrained because Western Europe is home to many long-standing democracies. For the elections included here, it ranges from 83.95 (Greece, 2012) to 96.04 (Germany, 2002). For the analyses, the scores are scaled.

The final metric is the Varieties of Democracy (V-Dem) Project’s Party Institutionalization Index (v2xps_party) (PII). As noted above, where parties are more institutionalized, the system becomes more institutionalized (Bolleyer and Ruth 2018). Mainwaring and Scully (1995) posit that parties must be socially embedded and formally organized internally for a system to be institutionalized. PII specifically taps into the political organization and embeddedness of the party: “Party institutionalization refers to various attributes of the political parties in a country, e.g. level and depth of organization, links to civil society, cadres of party activists, party supporters within the electorate, coherence of party platforms and ideologies, party-line voting among representatives within the legislature” (Coppedege et al. 2020). These are means for the public to become aware of the players. Like the others, party size is a factor here. The PII includes “the attributes of all parties with an emphasis on larger parties, i.e. those that may be said to dominate and define the party system.” The variable ranges from 0 to 1, with higher scores indicating greater institutionalization. For the election years included here, it ranges from 0.184 (Philippines 2016) to 1 (Sweden, 2002/2006).

The partisan affect, party system age, and satisfaction scores in a few example countries offers some context. These are selected for geographic diversity as well as variation in these key variables. In the United States, party affect is common: 86% report PPA and 69% report NPA. The US (2012) party age is 161 years, and the mean satisfaction with democracy is 2.94 out of 4. In the Netherlands (2010), PPA and NPA are more common (96% and 93% respectively). The party age is 36, and the mean satisfaction is 2.83. In Peru (2016), mean satisfaction is far lower (1.88). The party age is 5.5, and 87% report PPA while 92% report NPA. Lastly, Korea (2008) has a party age of 11.6. The mean satisfaction is 2.35. 85% report PPA and 84% report NPA. Thus, the dataset includes variations on these key features.

A few demographic covariates are included as well in keeping with prior literature on satisfaction with democracy and on partisan affect. A scaled variable for respondent age, measured in years, is included. Age has been associated with both partisan affect (Rose and Mishler 1998) and with satisfaction with democracy (Anderson and Guillory 1997; Luhiste 2014). Gender is included as a binary indicator for male. Gender has been linked to partisan affect (Rose and Mishler 1998) and to satisfaction with democracy (Curini et al., 2012; Dahlberg and Linde, 2016). Education is included as a binary indicator for college education. Education has been linked to partisan attitudes as a proxy for socioeconomic status (Medeiros and Noël 2014; Rose and Mishler 1998) and for political sophistication (Mayne and Hakhverdian 2017). Socioeconomic status is also included as a binary indicator for being in the fourth or fifth income quintile. These traits have been linked to satisfaction with democracy (Anderson and Guillory 1997; Dahlberg and Linde 2016; Aldrich et al., 2020).

A variable for respondent ideology is included. It is a zero to ten self-report rating from “left” to “right” (CSES IMD). The ideology rating is scaled for this analysis. Individual’s ideological position predicts their sentiments towards parties in their system (Medeiros and Noël 2014). Attitudes can reflect both ideological proximity and the electoral competition a party represents to the citizen’s preferred party. In forming their evaluations of the political parties and their democracy, respondents are “sensitive to the realities of the ideological menu before them” (Mayne and Hakhverdian 2017, 842). Conservatives are also more likely to report satisfaction with their democracy (Luhiste 2014; Ridge 2021b).

Some system-level variables were also included. A binary indicator identifies presidential as opposed to parliamentary and mixed political systems; system structure informs satisfaction with democracy and the form in which the party system congeals (Anderson and Guillory 1997; Bielasiak 2002; Mainwaring and Scully 1995). GDP per capita is also included. This accounts for the positive influence of state deliverables on satisfaction with democracy and development that likely happened in states over 20 years (Luhiste 2014). Party system institutionalization has been positively correlated with country wealth (Mainwaring and Torcal 2006).¹⁰

As satisfaction with democracy is a four-point ordered variable, ordered logistic regression models are employed. Hierarchical models, grouping responses by election, are appropriate to account for the fact that surveys were taken for particular local elections. Failure to do so can artificially increase the measured significance of national-level variables (Wells and Krieckhaus 2006). In this case, key independent variables – the system institutionalization scores – are the same for all respondents in each election survey. The IMD sample includes a sufficient number of election cases that multilevel models are useable (Bryan and Jenkins 2016). Hierarchical modeling thus reduces the likelihood of overstating the results. Multilevel ordered models have been used in previous satisfaction with democracy studies (Curini et al. 2012; Ridge 2020, 2021b). As in prior studies, mixed effects models with election random effects and fixed effects for the other variables are used.¹¹

Results and discussion

In general, PPA is positively associated with satisfaction with democracy, while NPA is negatively associated with democratic satisfaction (Appendix 2). The positive effect of PPA on satisfaction with democracy is of greater magnitude than the negative effect of NPA on satisfaction. The effects are stronger for the affect towards larger parties than towards the smaller parties, both for positive and negative party affect. Citizens in more institutionalized party systems are also generally more satisfied (Appendix 2). This direct effect is consistent with the party system institutionalization literature’s statements about improved democracy through institutionalization.

To assess the impact of party system institutionalization, conditional effects are examined. Table 2 presents the interactions between the party system institutionalization variables and the respondents’ affect binaries. There is a positive interaction between PPA and system institutionalization. The results are significant for the Party System Age (Model 1), PSI (Model 3), and Party Institutionalization (Model 5) metrics. This is consistent with Hypothesis 1a. The NPA interactions are not significant. Hypothesis 1b is then not confirmed. Thus, it seems that citizens are more likely to be satisfied with their democracy when they have a party participating that they like and more so when the parties are institutionalized – when that favored party will be sticking around and they can predict how it will interact with the other parties. The stabilization of the party system though is not making the negative effect of party animosity stronger.

Table 2.

Effect of institutionalization and partisan affect on satisfaction with democracy.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Positive affect	0.53^***	—	0.76^***	—	0.21^*	—
Positive affect	(0.03)	—	(0.03)	—	(0.08)	—
Negative affect	−0.29^***	—	−0.29^***	—	−0.30^***	—
Negative affect	(0.03)	—	(0.03)	—	(0.09)	—
Positive affective (major party)	—	0.58^***	—	0.82^***	—	0.18^*
Positive affective (major party)	—	(0.03)	—	(0.03)	—	(0.09)
Positive affective (minor party)	—	0.14^***	—	0.28^***	—	0.30^**
Positive affective (minor party)	—	(0.04)	—	(0.03)	—	(0.11)
Negative affective (major party)	—	−0.36^***	—	−0.36^***	—	−0.33^***
Negative affective (major party)	—	(0.03)	—	(0.03)	—	(0.09)
Negative affective (minor party)	—	−0.07	—	−0.06	—	−0.20
Negative affective (minor party)	—	(0.04)	—	(0.04)	—	(0.11)
Party system age	1.25^***	1.25^***	—
Party system age	(0.21)	(0.21)	—
PSI	—		0.27^*	0.27^*	—
PSI	—		(0.12)	(0.12)	—
PII	—				0.55	0.52
PII	—				(0.52)	(0.55)
Positive Affect:Party system age	0.20^***	—
Positive Affect:Party system age	(0.04)	—
Party system Age:Negative affect	−0.02	—
Party system Age:Negative affect	(0.04)	—
Positive affective (major Party):Party system age	—	0.20^***	—
Positive affective (major Party):Party system age	—	(0.04)	—
Positive affective (minor Party):Party system age	—	0.11	—
Positive affective (minor Party):Party system age	—	(0.06)	—
Party system Age:Negative affective (major party)	—	0.01	—
Party system Age:Negative affective (major party)	—	(0.04)	—
Party system Age:Negative affective (minor party)	—	0.02	—
Party system Age:Negative affective (minor party)	—	(0.05)	—
Positive Affect:PSI	—		0.05^*	—
Positive Affect:PSI	—		(0.03)	—
PSI:Negative affect	—		−0.05	—
PSI:Negative affect	—		(0.03)	—
Positive affective (major Party):PSI	—			0.05	—
Positive affective (major Party):PSI	—			(0.03)	—
Positive affective (minor Party):PSI	—			−0.02	—
Positive affective (minor Party):PSI	—			(0.03)	—
PSI:Negative affective (major party)	—			−0.05	—
PSI:Negative affective (major party)	—			(0.04)	—
PSI:Negative affective (minor party)	—			0.01	—
PSI:Negative affective (minor party)	—			(0.04)	—
Positive Affect:PII	—				0.49^***	—
Positive Affect:PII	—				(0.09)	—
PII:Negative affect	—				0.00	—
PII:Negative affect	—				(0.10)	—
Positive affective (major Party):PII	—					0.58^***
Positive affective (major Party):PII	—					(0.10)
Positive affective (minor Party):PII	—					−0.11
Positive affective (minor Party):PII	—					(0.12)
PII:Negative affective (major party)	—					−0.03
PII:Negative affective (major party)	—					(0.11)
PII:Negative affective (minor party)	—					0.15
PII:Negative affective (minor party)	—					(0.12)
Age	0.00	0.00	0.00	0.00	0.00	0.00
Age	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)
Male	0.07^***	0.07^***	0.12^***	0.11^***	0.07^***	0.07^***
Male	(0.01)	(0.01)	(0.02)	(0.02)	(0.01)	(0.01)
Higher education	0.17^***	0.18^***	0.21^***	0.23^***	0.17^***	0.18^***
Higher education	(0.01)	(0.01)	(0.02)	(0.02)	(0.01)	(0.01)
Ideology	0.16^***	0.14^***	0.10^***	0.05^***	0.16^***	0.14^***
Ideology	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)
High income	0.22^***	0.22^***	0.30^***	0.29^***	0.22^***	0.21^***
High income	(0.01)	(0.01)	(0.02)	(0.02)	(0.01)	(0.01)
GDP per capita	0.21^**	0.20^*	0.43^**	0.41^**	0.51^***	0.50^***
GDP per capita	(0.08)	(0.08)	(0.14)	(0.14)	(0.08)	(0.08)
Presidential system	−0.14	−0.15	—		0.31	0.30
Presidential system	(0.16)	(0.16)	—		(0.18)	(0.18)
1\|2	−1.43^***	−1.46^***	−2.25^***	−2.28^***	−1.50^**	−1.55^**
1\|2	(0.12)	(0.12)	(0.14)	(0.14)	(0.47)	(0.50)
2\|3	0.62^***	0.60^***	−0.17	−0.18	0.54	0.50
2\|3	(0.12)	(0.12)	(0.14)	(0.14)	(0.47)	(0.50)
3\|4	3.58^***	3.60^***	3.18^***	3.20^***	3.50^***	3.50^***
3\|4	(0.12)	(0.12)	(0.14)	(0.14)	(0.48)	(0.50)
AIC	303939.88	302146.39	123260.72	122372.48	306919.77	305055.64
N	143398	143398	63722	63722	144571	144571
Groups (election)	135	135	52	52	137	137
Variance: Election: (Intercept)	0.45	0.45	0.68	0.68	0.59	0.59

***p < 0.001; **p < 0.01; *p < 0.05.

Now consider the distinction between major and minor parties. With respect to Party System Age, with the influence of institutionalization PPA’s influence is heightened. Institutionalization significantly increases the positive effect of PPA for major and minor parties (p=0.7) (Model 2). For PII, the interaction is only significant for major party PPA (Model 6). In the PSI scale, the interaction for the major party approaches significance (p=0.09) (Model 4). In all three cases, the effect for major parties is significantly stronger than the effect for minor parties. The NPA effects are not significantly dependent on institutionalization in these models.

This result provides support for Hypothesis H2a. Party system institutionalization makes having a preferred party even more impactful. However, when the party that one likes is one of the smaller parties, the party is seemingly getting locked into a subordinated position. In a parliamentary system, the party can still play a role and may even be a kingmaking coalition member, thereby achieving influence. However, institutionalization does not magnify the effect of that affect for these partisans the same way institutionalization seems to for PPA holders of the larger parties. Hypothesis H2b does not receive support.

To put these results in context, consider a hypothetical individual. She is of average age and ideology. She is not in the top-two income quartiles or college educated; the country is a parliamentary system with average GDP per capita. In a system where the party age is 15 years, in her probability of being very satisfied with her democracy when she has PPA is 2.3 percentage points higher than when she does not, and her probability of being not very satisfied is 7.5 points lower. When party age is 50 years, her probability of being very satisfied is 4.0 points higher, and her probability of being not very satisfied is 10.3 points lower. When party age is 85 years, her probability of being very satisfied is 6.5 points higher, and her probability of being not very satisfied is 11.2 points lower.

Similarly, if the Party Institutionalization Index score is 0.2, in her probability of being very satisfied with her democracy when she has PPA is 1.1 percentage points higher than when she does not, and her probability of being not very satisfied is 3.7 points lower. When the score is 0.5, her probability of being very satisfied is 2.1 points higher, and her probability of being not very satisfied is 6.5 points lower. When the score is 0.8, her probability of being very satisfied is 3.4 points higher, and her probability of being not very satisfied is 9.5 points lower.

Conclusion

Researchers have focused on citizens’ satisfaction with their democracies because they tie reservoirs of democratic goodwill to democratic stability (Aldrich et al., 2020; Anderson and Guillory 1997; Curini et al. 2012). This specific support for a democracy, over time, can promote or erode diffuse support for democracy itself (Easton 1975; Linde and Peters 2020). This led to affirmation of Converse’s (1969) argument on the role of party commitments and democratic commitment by way of promoting and undermining satisfaction with democracy (Aldrich et al., 2020; Ridge 2020). This study extends the examination of party affect and satisfaction with democracy.

In particular, it focuses on the role of party system institutionalization in transmuting PPA and NPA to democratic satisfaction. Broadly, the literature has shown, PPA increases citizens’ propensity to be satisfied with their democracies, while NPA decreases that propensity. Institutionalized party systems facilitate partisan attachment and connect that attachment to the political system overall. As party systems congeal and become more predictable, citizens better understand the parties in their democracy, the roles they are likely to play in future governments, and the policies that they would promote. With this insight, they are better able to form emotional responses to these parties. The responses are heightened by the citizens’ awareness of the parties circumstances and potential for power, especially major parties. Thus, party system institutionalization can magnify the effects of partisan affect.

The effect of PPA on satisfaction is significantly stronger in more institutionalized systems. This relationship seems to be driven by the amplified effect of PPA for the largest parties in a system. In systems in flux, attachments are less likely or weaker, and they may not seem as system relevant. In stabilized systems, the parties are knowable and they are embedded in the political environment. Citizens who like them know that they will likely continue to be part of that democracy and that their current democracy – including its rules for party formation and power distribution – creates space for that party. The study does not find, however, consistent evidence that the negative effect of NPA changes with the level of system institutionalization. The effect of institutionalization seems then to be a love story. It is magnifying the positive effect of affection without necessarily magnifying the harms of hate.

This study, as all survey studies, must be presented with some caveats. Countries select into the CSES panel. As such, the surveys feature a non-random and incomplete sample of world democracies. The surveys also are taken once an election has already occurred, so the studies cannot account for ex ante attitudes. It also means that timing of the studies is not the same across all countries, leading some countries to appear more often than others. Elections are not always exogenous to country circumstances. Though it is not suspected that the elections are timed based on partisan affect or satisfaction with democracy, they are not random with respect to country circumstances. Panel surveys would be useful for future research aiming to extend political scientists’ understanding of partisanship in satisfaction with democracy.

Nonetheless, it is not to conclude that heightening levels of interparty affect are not a risk to democracy. The negative effect of NPA is not necessarily increasing with system institutionalization, although this study does not rule that out. That finding does not mean the negative effect does not exist. Strong negative sentiment towards parties in the democracy still undermines democratic satisfaction. Ameliorating negative interparty affect or attenuating its relationship to satisfaction with democracy would be beneficial in addressing the negative impact of affective partisanship on democracy. The positive effect of PPA and party system institutionalization both speak to potential avenues for supporting democratic systems.

Supplemental Material

Supplemental Material - Party system institutionalization, partisan affect, and satisfaction with democracy

Supplemental Material for Party system institutionalization, partisan affect, and satisfaction with democracy by Hannah Ridge in Party Politics

Footnotes

Acknowledgements

I would like to thank Markus Wagner and the participants of the Affective Polarization Around the World: Measurement, Causes, and Consequences workshop for helpful feedback on an earlier version of this article. This article also benefited from insightful comments by the anonymous reviewers.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Hannah M. Ridge

Supplemental Material

Supplemental material for this article is available online.

Notes

Author biography

Hannah M. Ridge (she/her) is a postdoctoral scholar in the University of Chicago Center for Human Rights. Her research focuses on public opinion on democracy, religion and secularism politics, and the Middle East. Her research has been published in Political Behavior, Party Politics, and Democratization. She earned a PhD in Political Science at Duke University.

References

Abramowitz

Webster

(2018) Negative partisanship: Why Americans dislike parties but behave like rabid partisans. Political Psychology 39: 119–135.

Abramowitz

Webster

(2016) The rise of negative partisanship and the nationalization of US elections in the 21st century. Electoral Studies 41: 12–22.

Aldrich

Austin

Krishnamurthy

Madan

Renberg

Ridge

(2020) Does a partisan public increase democratic stability? Research Handbook on Political Partisanship. Edward Elgar Publishing.

Anderson

(2002) Good Questions, Dubious Inferences, and Bad Solutions: Some Further Thoughts on Satisfaction with Democracy. Binghamton, NY: Binghamton University.

Anderson

Guillory

(1997) Political institutions and satisfaction with democracy: A cross-national analysis of consensus and majoritarian systems: 66–81.

Bielasiak

(2002) The institutionalization of electoral and party systems in postcommunist states. Comparative Politics 189–210.

Blais

Morin-Chassé

Singh

(2017) Election outcomes, legislative representation, and satisfaction with democracy. Party Politics 23(no. 2): 85–95.

Bolleyer

Ruth

(2018) Elite investments in party institutionalization in new democracies: a two-dimensional approach. The Journal of Politics 80(1): 288–302.

Boxell

Gentzkow

Shapiro

(2020) Cross-country Trends in Affective Polarization. National Bureau of Economic Research.

10.

Bryan

Jenkins

(2016) Multilevel modelling of country effects: A cautionary tale. European Sociological Review 32(1): 3–22.

11.

Çakır

(2021) Does Party Ambivalence Decrease Voter Turnout? A Global Analysis. Party Politics.

12.

Canache

Mondak

Seligson

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

13.

Caruana

Michael McGregor

Stephenson

(2015) The power of the dark side: Negative partisanship and political behaviour in Canada. Canadian Journal of Political Science/Revue canadienne de science politique 771–789.

14.

Chiaramonte

Emanuele

(2017) Party system volatility, regeneration and de-institutionalization in Western Europe (1945–2015). Party Politics 23(4): 376–388.

15.

Chiaramonte

Emanuele

(2019) Towards turbulent times: measuring and explaining party system (de-) institutionalization in Western Europe (1945–2015). Italian Political Science Review 49(1): 1–23.

16.

Converse

(1969) Of time and partisan stability. Comparative Political Studies 2(2): 139–171.

17.

Coppedge

Gerring

, et al. (2020) V-dem Codebook V10" Varieties of Democracy. V-Dem) Project.

18.

Curini

Jou

Memoli

(2012) Satisfaction with democracy and the winner/loser debate: The role of policy preferences and past experience. British Journal of Political Science 241–261.

19.

Cruz

Keefer

Scartascini

(2021) The Database of Political Institutions 2020 (DPI2020). DOI: 10.18235/0003049

20.

Dahlberg

Linde

Jonas

Holmberg

(2015) Democratic discontent in old and new democracies: Assessing the importance of democratic input and governmental output. Political Studies 63: 18–37.

21.

Dahlberg

Linde

Jonas

(2016) Losing happily? The mitigating effect of democracy and quality of government on the winner–loser gap in political support. International Journal of Public Administration 39(9): 652–664.

22.

Dalton

Weldon

(2007) Partisanship and party system institutionalization. Party Politics 13(2): 179–196.

23.

Druckman

Levendusky

(2019) What Do We Measure When We Measure Affective Polarization? Public Opinion Quarterly 83(1): 114–122.

24.

Easton

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

25.

Hernández

Anduiza

Guillem

(2021) Affective polarization and the salience of elections. Electoral Studies 69: 102203.

26.

Hicken

Martinez Kuhonta

(eds), (2015) Party system institutionalization in Asia: democracies, autocracies, and the shadows of the past. Cambridge University Press.

27.

Iyengar

Sood

Lelkes

(2012) Affect, not ideology: a social identity perspective on polarization. Public Opinion Quarterly 76(3): 405–431.

28.

Iyengar

Westwood

(2015) Fear and loathing across party lines: New evidence on group polarization. American Journal of Political Science 59(3): 690–707.

29.

Kuenzi

Lambright

(2001) Party system institutionalization in 30 African countries. Party Politics 7(4): 437–468.

30.

Knudsen

(2021) Affective Polarization in Multiparty Systems? Comparing Affective Polarization Towards Voters and Parties in Norway and the United States. Scandinavian Political Studies 44(1): 34–44.

31.

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.

32.

Linde

Peters

(2020) Responsiveness, support, and responsibility: How democratic responsiveness facilitates responsible government. Party Politics 26(3): 291–304.

33.

Luhiste

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

34.

Mainwaring

(2016) Party system institutionalization, party collapse and party building. Government and Opposition 51(4): 691–716.

35.

Mainwaring

Bizzarro

Petrova

(2018) Party System Institutionalization, Decay, and Collapse” in. Cambridge: Cambridge University Press.

36.

Mainwaring

Scully

(1995) Introduction: Party Systems in Latin America” in. Scott Mainwaring and Timothy Scully. 1-34. Stanford: Stanford University Press.

37.

Mainwaring

Torcal

(2006) Party system institutionalization and party system theory after the third wave of democratization. Handbook of Party Politics 11(6): 204–227.

38.

Mayer

(2017) How negative partisanship affects voting behavior in Europe: Evidence from an analysis of 17 European multi-party systems with proportional voting. Research & Politics 4(no. 1): 2053168016686636.

39.

Mayne

Hakhverdian

(2017) Ideological congruence and citizen satisfaction: Evidence from 25 advanced democracies. Comparative Political Studies 50(6): 822–849.

40.

McGregor

Stephenson

(2015) Negative partisanship in a multi-party system: the case of Canada. Journal of Elections, Public Opinion and Parties 25(3): 300–316.

41.

Medeiros

Noël

(2014) The forgotten side of partisanship: Negative party identification in four Anglo-American democracies. Comparative Political Studies 47(7): 1022–1046.

42.

Piñeiro Rodríguez

Rosenblatt

(2020) Stability and incorporation: Toward a new concept of party system institutionalization. Party Politics 26(2): 249–260.

43.

Reiljan

(2020) Fear and Loathing across Party Lines” (also) in Europe: Affective Polarisation in European Party Systems. European Journal of Political Research 59(2): 376–396.

44.

Ridge

(2021) Electoral outcomes and support for Westminster democracy. Journal of Elections, Public Opinion and Parties 1–20.

45.

Ridge

(2020) Enemy Mine: Negative Partisanship and Satisfaction with Democracy. Political Behavior 1–25.

46.

Ridge

(2021) Just like the others: Party differences, perception, and satisfaction with democracy. Party Politics 1354068820985193.

47.

Rose

Mishler

(1998) Negative and positive party identification in post-communist countries. Electoral Studies 17(2): 217–234.

48.

Russo

Mayer

(2021) Divide and Conquer: Disentangling Negative Partisanship and Affective Polarization.

49.

Scarrow

Gezgor

(2010) Declining memberships, changing members? European political party members in a new era. Party Politics 16(6): 823–843.

50.

Sheffer

Lior.

(2020) Partisan in-group bias before and after elections. Electoral Studies 67: 102191.

51.

Thames

Robbins

(2007) Party System Institutionalization and the Level of Democracy Annual Meeting of the American Political Science Association.

52.

Van

Ingrid

Poguntke

(2012) Going, going,… gone? The decline of party membership in contemporary Europe. European Journal of Political Research 51(1): 24–56.

53.

Vlachová

(2001) Party identification in the Czech Republic: inter-party hostility and party preference. Communist and Post-Communist Studies 34(4): 479–499.

54.

Wagner

(2020) Affective Polarization in Multiparty Systems. Electoral Studies.

55.

Wells

Krieckhaus

(2006) Does national context influence democratic satisfaction? A multi-level analysis. Political Research Quarterly 59(4): 569–578.

56.

Westwood

Iyengar

Walgrave

, et al. (2018) The tie that divides: Cross‐national evidence of the primacy of partyism. European Journal of Political Research 57(2): 333–354.

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

0.61 MB