Partisan Polarization and the Effect of Congressional Performance Evaluations on Party Brands and American Elections

Abstract

Scholars of political parties and elections around the globe have devoted extensive study toward understanding how party polarization affects the criteria voters use to determine their party preferences in elections. In American politics, previous empirical research suggests that congressional job performance ratings do not affect public attitudes toward parties and therefore have no significant ramifications for the electoral fortunes of parties. Building on the existing literature from American and comparative traditions, I demonstrate that congressional performance evaluations can be important for American parties and elections, but this depends on the extent of party polarization in Congress. First, when congressional parties are more distinct, congressional performance evaluations have a greater effect on the relative favorability of the majority party brand: when citizens like how Congress is performing under the leadership of the majority party, this boosts the favorability of that party overall. Second, because partisan polarization increases the effect of congressional performance evaluations on party brand favorability, partisan polarization also increases the role played by congressional performance evaluations in a variety of elections outside of Congress. These results inform our understanding of public opinion, parties, and elections, both in the U.S. and from a comparative perspective.

Keywords

party brands polarization political parties public opinion voter psychology U.S. Congress

In recent years, scholars of political parties and elections around the globe have devoted extensive study toward understanding how party polarization affects the criteria voters use to determine their party preferences (e.g., Clark and Leiter 2014; Dalton 2008; J. Green 2007; Pardos-Prado 2012). These ongoing debates in the comparativist literature raise an interesting question in the American setting. Although survey research consistently finds that performance evaluations of a president play a significant role in the public’s overall attitudes toward his party (e.g., Jacobson 2012), it has not found this to be the case for performance evaluations of Congress. Seminal scholarship in American politics suggests that citizens’ evaluations of the performance of the U.S. Congress do not affect their attitudes toward the parties (e.g., Stokes and Miller 1962). However, these studies date from a period of historically low polarization of congressional parties. One possibility, suggested in part by the comparativist literature (e.g., Clark and Leiter 2014; Pardos-Prado 2012), is that in times of greater party polarization in Congress—such as the current era of American politics—congressional performance may play an enhanced role in shaping citizens’ feelings toward the parties and their preferences in elections. If so, this would not only upend conventional scholarly¹ wisdom in American politics, but it would also inform an important debate among comparative scholars of parties and elections.

This study examines whether variation in the level of party polarization in Congress—and thus the clarity of congressional politics—makes any difference in how Americans evaluate U.S. political parties. First, I argue that the greater the level of partisan polarization in Congress, the more likely citizens’ evaluations of congressional performance will shape their attitudes toward the majority party brand specifically. This is because more distinct parties in Congress make it easier for Americans to associate congressional performance with a single party—the majority party—as opposed to a muddled collectivity. If citizens like how Congress is performing under the leadership of the majority party, this boosts the overall brand favorability of that party.

Second, as a consequence of the first argument, I argue that greater partisan polarization in Congress also increases the effect that congressional performance evaluations have in a variety of elections—even those outside of Congress. The literature shows that a party’s overall brand affects the electoral fortunes of all candidates running under that party label. Thus, any factor that harms or benefits a party’s brand has the potential to affect all partisan elections. If greater partisan polarization in Congress increases the effect of congressional performance on party brand favorability (see my first argument), then polarization should also increase the effect of congressional performance on a variety of elections up and down the ballot, including presidential elections and state legislative elections.

Perhaps because of the longstanding conventional wisdom first articulated by Stokes and Miller, these arguments have yet to be systematically tested in the context of American politics. This study attempts to do fill this gap. I begin by reviewing the literature on the electoral importance of party brand favorability and its connection, or lack thereof, to congressional performance evaluations. Next, I present the theoretical basis for my argument regarding polarization and identify specific hypotheses to be tested. The analysis then proceeds in two parts. First, I analyze how variation in the actual level of partisan polarization in Congress conditions the effect of congressional performance evaluations on favorability toward each party. Second, I analyze how partisan polarization in Congress conditions the effect of congressional performance evaluations on presidential elections and state legislative elections. Finally, I conclude with a discussion of how the results inform our understanding of public opinion, parties, and elections, both in the U.S. and from a comparative perspective.

Literature

Various literatures exist on particular aspects of citizens’ attitudes toward parties. Some works focus on party identification (e.g., Campbell et al. 1960), others on the specific content of party images that citizens hold in their heads (e.g., Brewer 2009), and others on the relative clarity of party ideological stands (e.g., Grynaviski 2010). This study focuses on the overall favorability—the degree of positive or negative affect—that citizens feel toward a party at a given point in time (e.g., Jacobson 2012). This same basic concept is also sometimes referred to as a party evaluation (D. Green, Palmquist, and Schickler 2002) or party brand favorability (Cox and McCubbins 2005).

How favorably citizens feel toward a party has important consequences in elections. In particular, political psychologists have found that the feelings a voter has toward a political party brand transfer to any political candidate running under that party label, even in the presence of contradictory candidate-specific information (Rahn 1993). The reason may be that a party’s brand label is an “information shortcut” voters use when assessing candidates (Popkin 1991). Consistent with this research, the public’s attitudes toward each party’s brand have been found to have a significant effect—independent from party identification—in American elections of varying types and levels, including congressional elections (Pope and Woon 2006), presidential elections (Brewer 2009; Matthews and Prothro 1966; Pope and Woon 2006; Trilling 1976), and state legislative elections (Butler and Powell 2014).

Given the substantial evidence that citizens’ favorability toward each party’s brand affects elections, there is considerable interest in identifying what information Americans use to form and update these attitudes. Of particular relevance to this study is the literature suggesting that party favorability varies over time in relation to the perceived performance of its members in government (e.g., Brewer 2009; Pope and Woon 2009; Popkin 1991).

To date, empirical research in support of the idea that performance in government affects party brands has focused mainly on the presidency. For instance, Jacobson (2012) shows that public ratings of President Clinton’s job performance in office affected the Democratic Party’s favorability during his administration, President George W. Bush’s performance ratings harmed the Republican Party’s favorability during his term, and President Obama’s performance ratings have affected the Democratic Party’s favorability while he has been in office. Similarly, Pope and Woon (2006) find effects of presidential popularity of Reagan, Clinton, and both Bushes on their respective parties’ brand reputations.

In theory, the public’s favorability toward a party could also be influenced by its perceived performance in Congress. But empirical research has been relatively more pessimistic about the possible role of congressional performance evaluations in shaping attitudes toward parties. In particular, work written in the 1960s and 1970s is fairly unanimous that Americans do not think about Congress in partisan terms (e.g., Fenno 1978; Mayhew 1974; Stokes and Miller 1962; but see Jones 2010a).

Yet the parties in Congress are often much more polarized by party than they were when this earlier research was conducted (McCarty, Poole, and Rosenthal 2006). There are several reasons to suspect that higher levels of congressional polarization may alter the way voters view Congress and its relevance to parties. First, from a theoretical perspective, the earlier literature explicitly speculates that a lack of party discipline in Congress may be responsible for Americans’ inability to conceptualize Congress in partisan terms. For example, the seminal American Political Science Association (1950, 17) report notes that a partisan basis for accountability is hindered whenever “both parties are . . . confused by internal divisions.”. Stokes and Miller (1962, 545) attribute their findings to the fact that “the legislative parties speak not as two voices but as a cacophony of blocs and individuals.” If so, clearer party distinctions could change the situation.

Second, more recent studies of congressional behavior find that in an era of higher party polarization, members of Congress act as if congressional actions affect party reputations. Sinclair (2006, 265) reports that members of Congress “have now come to expect their party leaders [in Congress] . . . to protect and enhance the party’s image.” Lee (2009, 132) argues that in the Senate, members act “with an eye to how floor votes and debates will affect their party’s overall image.” Scholars such as Cox and McCubbins (2005) and Grynaviski (2010) note the agenda-setting powers that majority party members have delegated to party leaders and argue that the explanation for this arrangement lies with member concern over how favorably the public views the party brand. Rohde (2013, 857) argues that since the government shutdown battles of 1996, “protecting the party brand has been one of the main concerns of party leaders.” However, none of these works demonstrate empirically that congressional performance ratings actually affect a party’s brand favorability, or that this relationship varies with polarization in Congress.²

Third, existing empirical studies of elite polarization suggest this phenomenon can have important effects on public opinion regarding parties and on partisan preferences in elections. Outside the United States, recent cross-national research finds evidence that greater party polarization can increase the electoral importance of citizens’ evaluations of party competence (Clark and Leiter 2014; Pardos-Prado 2012). In particular, Clark and Leiter (2014, 188) conclude that in more highly polarized contexts, parties are held more accountable for “good governance.” In the American context, Grynaviski (2010), Hetherington (2001), and Pope and Woon (2009) all find that greater partisan polarization in Congress has increased both the extent to which citizens perceive differences between the parties and the number of likes and dislikes they offer about the parties. None of these studies look at how congressional polarization may have increased the effects of public evaluations of congressional performance on party favorability.

Jones (2010b) explicitly examines the interaction between congressional polarization and congressional performance ratings and their effects in U.S. House elections. Among other results, this study finds that in the period 1954–2006, greater levels of polarization in Congress increased the positive effect of congressional performance ratings on the vote share received by majority party candidates. In a general sense, this finding is consistent with a model in which congressional performance ratings affect party brand favorability. Alternatively, however, it could be that citizens are merely voting retrospectively within the domain of Congress, and not drawing larger conclusions about the parties generally. As such, the findings cannot be taken as definitive evidence of a link between congressional performance ratings and overall party favorability.

In sum, the literature offers evidence consistent with the idea that greater partisan polarization in Congress affects whether and how citizens use congressional performance evaluations to help form attitudes toward the parties, thereby influencing elections. However, the literature has yet to formulate explicit hypotheses or conduct direct empirical tests of this idea.

Hypotheses

Partisan polarization in Congress has multiple components. Some studies have focused on an underlying ideological component, others on the nonideological “teamsmanship” that accounts for almost half of all observed polarization (Lee 2009). This study’s argument does not necessitate distinguishing between these components. What is important here is the visible manifestation of partisan polarization that is salient to the voting public: parties voting more differently from one another.³

This study makes two related arguments about the conditioning effects of congressional partisan polarization. My first argument is that when partisan polarization in Congress is low, more actions of Congress have support from members from both parties. In turn, the salient actions of Congress that citizens perceive in such a scenario are more likely to be characterized by support from members of both parties. In this case, congressional actions provide less of a clear idea about which party is responsible for legislative outcomes. As a result, citizens will find it more difficult to use the performance of Congress as a basis for updating their evaluations of the parties, and both parties will be only weakly associated with perceived congressional performance. (Citizens will still hold party evaluations, but these will mostly consist of impressions based on factors outside of Congress.)

In contrast, when the level of partisan polarization in Congress is higher, the public face of Congress changes. With greater partisan polarization, members of the majority party vote more as a unified team, opposed by the minority party. The more this pattern occurs, the more likely citizens are to perceive that the parties are distinct from each other, and that it is primarily the majority party that is responsible for the work product of Congress. As a result, with a relatively high level of polarization, a citizen who negatively evaluates the job performance of Congress is more likely to see this as a failing of the majority party and a citizen who positively evaluates Congress is more likely to give credit to the majority party.⁴

Research demonstrates that citizens update their overall attitude toward a political party based on salient behavior by public officials from that party (Brewer 2009; Pope and Woon 2009; Popkin 1991). As greater partisan polarization in Congress makes people increasingly aware of congressional politics and the central role played by the majority party (Hetherington 2001), it should also increase the extent to which citizens use congressional job performance to update their overall impressions of the majority party specifically.⁵ Thus, my first hypothesis is as follows:

Hypothesis 1a (H1a): The greater the level of partisan polarization in Congress, the more citizens’ ratings of congressional job performance will affect their favorability toward the majority’s party brand.

Implicit in this argument is a key assumption regarding causal direction. To facilitate testing this assumption, I state it explicitly here as a subhypothesis of the first argument:

Hypothesis 1b (H1b): Ratings of congressional job performance affect subsequent favorability toward the majority’s party brand.

As for favorability toward the minority party, I do not expect partisan polarization in Congress to have much of an interactive effect. When polarization is low, congressional performance evaluations will have only a small effect on minority party brand because citizens are less likely to be thinking about Congress in terms of parties. When polarization is high, congressional performance evaluations will have only a small effect on the minority party brand because citizens identify mainly the majority party, not the minority party, as responsible for congressional activity.

My second main argument concerns electoral implications. A party’s brand image affects the willingness of citizens to vote for candidates from that party (Brewer 2009; Butler and Powell 2014; Matthews and Prothro 1966; Pope and Woon 2006; Trilling 1976). When the Democratic brand is particularly poor, Democratic candidates will find themselves at an electoral disadvantage up and down the ballot. When the Republican brand is particularly poor, Republican candidates will be similarly disadvantaged. Given that party brands affect elections, any factor that has a heightened impact on party brands should also have a heightened impact in elections. Specifically, if greater partisan polarization in Congress causes an increased effect of congressional performance evaluations on a party’s brand, it could also lead to an increased effect of congressional performance evaluations in elections.

Just as poor presidential performance ratings often harm the electoral prospects for all candidates from the president’s party—including both federal races (e.g., Tufte 1975) and local races (e.g., Simon, Ostrom, and Marra 1991)—I argue that poor congressional performance ratings can negatively affect all candidates from the majority party (not just those running for Congress).⁶ In both cases, one logical reason is party brands. A candidate running under a given party label is at a greater disadvantage whenever the performance of that party in office leads voters to view that party label in a negative light. Although this is not a new idea in reference to presidential performance evaluations, it remains a largely unexplored idea in regard to congressional performance evaluations.

As party brands have been shown to be relevant in both high-salience national races, such as presidential elections (Brewer 2009; Matthews and Prothro 1966; Pope and Woon 2006; Trilling 1976), and in low-salience local races, such as state legislative elections (Butler and Powell 2014), I hypothesize that congressional performance evaluations will be relevant in such races, conditional on congressional polarization. Specifically, my second set of hypotheses is as follows:

Hypothesis 2a (H2a): The greater the level of partisan polarization in Congress, the more citizens’ ratings of congressional job performance will affect the majority party’s electoral prospects in presidential elections.

Hypothesis 2b (H2b): The greater the level of partisan polarization in Congress, the more citizens’ ratings of congressional job performance will affect the majority party’s prospects in state legislative elections.

It is important to note that the hypotheses outlined above apply only when there is a single majority in Congress. During Congresses featuring split-party control of the House and Senate, the theory predicts that because each party is the majority in one chamber and the minority in the other, the overall effect of congressional performance ratings should always be about the same on both parties’ brand favorability, regardless of the level of polarization. As a result, there is no theoretical reason to expect polarization to increase one particular party’s relative electoral culpability for perceived congressional performance. The appendix presents evidence consistent with these expectations. However, the main analysis excludes split-party-control Congresses for important reasons. First, the goal of the study is to illuminate “normal” politics in the United States, and split-control Congresses are relatively infrequent events (only five cases in the eight decades from 1933 to 2013). Second, the infrequency of these split-control Congresses poses an empirical problem: it is almost impossible to accurately estimate the conditioning effects of polarization across split-control Congresses with only four (due to other data restrictions) values of polarization available for analysis. Of particular concern is that any results are likely to be heavily influenced by single Congresses that may or may not be typical of split control. Interested readers can find additional information in the appendix.

In sum, this project’s argument consists of two sets of hypotheses. First, partisan polarization in Congress increases the degree to which congressional performance ratings affect favorability toward the majority party’s brand. Second, partisan polarization in Congress increases the degree to which congressional performance ratings affect partisan voting across a variety of elections. Each set of hypotheses will be analyzed in turn.

Analyzing Effects on Party Favorability

To analyze how individuals’ evaluations of congressional performance affect their favorability toward each party, it is necessary to identify surveys that ask about both of these concepts in the same questionnaire. The American National Election Studies (ANES) is the only regular academic survey to do so. Beginning in 1980, ANES has asked the standard congressional performance question, “Do you approve or disapprove of the way the U.S. Congress has been handling its job?”⁷ Consistent with other studies (e.g., D. Green, Palmquist, and Schickler 2002; Grynaviski 2010; Jacobson 2012), I measure party favorability using questions that ask respondents to rate “the Democratic Party” and “the Republican Party” on 0- to 100-point favorability scales.

The central question is how variation in the level of partisan polarization in Congress changes the effect of congressional performance evaluations on party brand favorability. To measure congressional partisan polarization, I use the Rice index of party difference. Specifically, I calculate the absolute difference between the percentage of Democrats voting yes and the percentage of Republicans voting yes on each roll call vote, averaged across all votes in each chamber in the two most recent Congresses.⁸ Although I believe this to be the most straightforward measure of my conceptual definition of partisan polarization, results are substantively similar if other common measures of polarization are used, such as the absolute difference in party means measured with Poole and Rosenthal’s DW-NOMINATE scores (results of this alternative analysis are provided in the Online Supplementary Appendix at http://prq.sagepub.com/supplemental/).

Figure 1 illustrates the range of polarization in the data used in the various analyses in the paper. For the present analysis of effects on party brands, this encompasses the period 1980–2008 (for a later aggregate analysis of elections, this encompasses the period 1958–2010). One potential methodological concern is that the polarization data display a bit of a time trend, with higher values tending to occur later in the series. To ensure that any effects of the hypothesized interaction between congressional performance evaluations and congressional partisan polarization are valid, and not attributable to other possible interactions that are not directly related to my hypothesis, I tried adding several alternative interactions to the model as empirical controls. Specifically, I tested interactions of congressional performance ratings with temporal variables such as a simple time trend, polarization in presidential ideology, polarization of the parties overall (not specifically Congress), aggregate campaign spending, and southern realignment. When added to the model presented here, none of these alternative interactions was statistically significant and none outperformed this study’s hypothesized interaction, which generally retained its statistical significance and substantive effect across the alternative models. Because these alternative interactions were not significant (and not theorized to have any specific effect), they are not included in the final model.

Figure 1.

Partisan polarization in Congress, 1958–2010.

Although all of the figures in the paper use the original polarization scale, in the analysis I center my polarization variable such that its mean equals 0 before interacting it with the variable for congressional performance. This formulation is useful because it means that in the tables, the baseline effect of the congressional performance evaluation variable reflects its impact when polarization is at an average level.

The model also includes key control variables. Respondents’ evaluations of presidential performance⁹ are interacted with a variable representing whether or not the president is from the same party as the majority in Congress, thereby creating two control variables: one capturing the effect of presidential performance ratings during years when the president is from the same party as the congressional majority (zero otherwise) and one capturing the effect during years when the president is from the opposing party (zero otherwise). I also control for respondents’ party identification¹⁰ and include dummy variables capturing fixed effects for each year (not shown).¹¹ Controls for majority party size and its interaction with congressional performance were found to be insignificant, and are thus not included in the analysis.

To test the first two hypotheses, I regress majority party brand favorability and minority party brand favorability on the independent variables. Here, “majority party” is defined as the party that holds the most seats in both chambers of Congress and “minority party” is defined as the party that holds fewer seats in both chambers of Congress. Years with split-party control of the House and Senate are not included in the present analysis but are discussed separately in the appendix, for reasons explained therein.

The results in Table 1 provide strong support for this study’s arguments regarding party brands. First, the findings support this study’s underlying premise that when citizens update their impressions of each party, they not only take into account their longstanding party identification and their evaluation of recent presidential performance, but they also factor in their perceptions of recent congressional performance. The congressional performance rating variable is statistically significant in both models, but on average its effect is nearly three times as large on the majority brand as on the minority party brand. This differential is consistent with the argument that voters perceive the majority party as primarily responsible for what goes on in Congress.

Table 1.

Effect of Congressional Performance Evaluations on Party Brand Favorability, 1980–2008.

	Majority party favorability		Minority party favorability
	Coefficient	(SE)	Coefficient	(SE)
Party identification	14.410*	(0.204)	−12.991*	(0.206)
Presidential performance evaluation (president same as majority)	7.065*	(0.306)	−3.251*	(0.309)
Presidential performance evaluation (president same as minority)	−3.158*	(0.193)	8.628*	(0.195)
President from the same party as majority in Congress	−1.746*	(1.789)	7.410*	(1.801)
Congressional polarization	0.101	(0.278)	0.051	(0.028)
Congressional performance evaluation	3.635*	(0.165)	1.334*	(0.166)
Congressional performance evaluation × Congressional polarization	0.062*	(0.023)	−0.003	(0.023)
Constant	55.010*	(2.757)	48.479*	(2.775)
Adjusted R²	.39		.40
No. of cases	16,613		16,644

Entries are regression coefficients. Year fixed effects are included in the model but not shown here. Years with split-party control of Congress are omitted.

p < .05 (one-tailed tests).

Second, the results show that partisan polarization in Congress conditions the effect of congressional performance evaluations, particularly for the majority party brand. As expected by H1a, the interaction between polarization and congressional approval is positive and significant in the majority party model, indicating that greater polarization increases the effect that congressional performance ratings have on evaluations of the majority party. In contrast, the interaction term in the minority party model is neither substantive nor statistically significant, indicating that, as expected, greater polarization does not increase the effect that congressional performance ratings have on evaluations of the minority party.

The substantive meaning of these findings is graphically presented in Figure 2, where the vertical scale represents estimated effect of congressional performance evaluations (approving vs. disapproving) on party favorability, and the horizontal scale represents the actual range of congressional partisan polarization from the data used in Table 1. Solid lines plot the conditional effect for the majority party (black) and for the minority party (gray). Dashed lines represent confidence intervals. The figure shows that at the lowest level of polarization in the data, approving of congressional performance (as opposed to disapproving) increases a person’s thermometer rating of the majority party by 5.6 points. But at the highest levels of polarization in the data, the effect of congressional performance evaluations increases to 8.7 points. In contrast, the effect of congressional performance evaluations on ratings of the minority party hovers at around 2.6 points, without any meaningful change as congressional partisan polarization increases. Thus, with greater polarization, citizen disapproval of Congress is disproportionately problematic for the majority party’s brand.

Figure 2.

Polarization in Congress and the effect of congressional performance evaluations on party brand favorability.

To put in perspective the size of the effect of congressional performance on the majority party brand, one can compare it with the effect of presidential performance. First, when the president is from the same party as the congressional majority, approving of the president (as opposed to disapproving) increases a citizen’s thermometer rating of the majority party by 14.1 points. In this case, the maximum effect of congressional performance evaluations (8.7 points) is almost two-thirds as large as the effect of presidential performance evaluations. Second, when the president is not from the same party as the congressional majority, his performance ratings have (understandably) less influence on that party’s brand. Specifically, approving of a president from the opposite party decreases a citizen’s ratings of the majority party by 6.3 points. In this case, the maximum estimated effect of congressional performance evaluations is 38 percent larger than the (absolute) effect of presidential performance evaluations (8.7 vs. 6.3).

Overall, these data are consistent with the argument that greater clarity of party differences in Congress leads citizens to increasingly use perceived congressional performance as a factor when forming their attitudes toward the majority party. Skeptics, however, might wonder about an alternative interpretation of the results in which citizens are perhaps using their favorability toward a party to form their evaluation of congressional performance. Although this alternative causal interpretation is more difficult to defend from a theoretical perspective,¹² it cannot be ruled out empirically with the cross-sectional survey data used in Table 1.

To demonstrate empirically that a causal arrow (at least in the sense of time order) runs from congressional performance ratings to party favorability, I conduct a test of H1b using data from an ANES panel survey of respondents interviewed in the fall of 1993 and again in the fall of 1994. Because time order is important for demonstrating causality, I test whether a citizen’s evaluation of congressional performance in the fall of 1993 affects his or her favorability toward the congressional majority party in 1994. To ensure that any effects of the predictors are on future attitudes (not simply continuations of cross-sectional effects from 1993), the model follows the standard practice of controlling for a citizen’s initial favorability toward the majority party. This specification provides the most stringent possible test of temporal causality. Also, as in Table 1, the model controls for party identification and presidential performance ratings.

When this panel model is estimated in Table 2, the coefficient for the congressional performance evaluation variable is still positive and significant. This result confirms that present evaluations of Congress do affect future favorability toward the majority party.¹³ Furthermore, an additional test (not shown here) demonstrates that the reverse is not true: present favorability toward the majority party has no significant effect on future evaluations of congressional performance. In other words, one’s evaluations of congressional performance influence one’s feeling toward the parties, not vice versa.

Table 2.

Testing for Causal Direction with Panel Data.

	Majority party favorability in 1994
	Coefficient	(SE)
Majority party favorability₁₉₉₃	0.45*	(0.04)
Party identification compatibility₁₉₉₃	6.52*	(0.82)
Presidential performance evaluation₁₉₉₃	3.97*	(0.74)
Congressional performance evaluation ₁₉₉₃	1.76*	(0.66)
Constant	28.00*	(2.07)
Adjusted R²	.58
No. of cases	614

Entries are regression coefficients.

p < .05.

A more recent ANES panel study, the 2008–2009 panel, does not use the same thermometer ratings of parties used in Tables 1 and 2, but does ask respondents similar questions regarding the degree to which they “like” or “dislike” each party on a seven-point scale. Reestimating the model in Table 2 with these alternative data from the 2008–2009 panel study produces the same basic finding: present evaluations of Congress significantly affect future favorability toward the majority party.¹⁴ Overall, the various panel data suggest that the results in Table 1 are not an artifact of reverse causality and are therefore properly interpreted as providing strong support for this study’s argument regarding party brands.

Analyzing Implications for Elections

The second set of hypotheses posits that greater party polarization in Congress also increases the effect of congressional performance evaluations on the majority party’s electoral fortunes across a variety of contests. In this section, I conduct two analyses looking for evidence of such effects in, respectively, presidential elections and state legislative elections.

Presidential Elections

For the analysis of effects of congressional performance evaluations on presidential elections, I operationalize the dependent variable as a vote for or against the presidential candidate from the party that has a majority of seats in the Congress. Congressional performance ratings and partisan polarization in Congress are measured as in the previous section. The model controls for other common influences on presidential voting, including presidential performance ratings, the economy, and party identification. Each of these variables is coded such that higher values are expected to lead to a positive response in the dependent variable.¹⁵ As this recoding is based on whether the party of Congress and the president match, the model necessarily includes a control for this factor as well. Year fixed effects are also included in the model, but not shown in the results.

Table 3 presents the logistic regression results. Consistent with H2a, model 1 shows that the interaction between congressional performance evaluations and congressional polarization is positive and significant. As partisan polarization in Congress increases, congressional performance evaluations have an increasingly larger effect on whom Americans vote for in presidential elections.

Table 3.

Effect of Congressional Performance Evaluations on Voting for President, 1980–2008.

	Model 1		Model 2
	Coefficient	(SE)	Coefficient	(SE)
Party identification	1.909*	(0.055)	0.838*	(0.067)
Presidential performance evaluation (president same as majority)	1.478*	(0.090)	1.143*	(0.098)
Presidential performance evaluation (president same as minority)	−1.364*	(0.048)	−0.923*	(0.054)
Economy (president same as majority)	0.526*	(0.140)	0.404*	(0.154)
Economy (president same as minority)	−0.353*	(0.071)	−0.213*	(0.078)
Congress and president of same party	−0.117	(0.150)	−0.036	(0.167)
Congressional polarization	1.362	(1.107)	3.861*	(1.213)
Congressional performance evaluation	0.272*	(0.043)	0.173*	(0.048)
Congressional performance evaluation × Congressional polarization	1.505*	(0.561)	1.308*	(0.608)
Majority party brand favorability			0.059*	(0.003)
Minority party brand favorability			−0.055*	(0.003)
Constant	0.111	(0.104)	−0.175	(0.218)
Nagelkerke R²	.71		.78
No. of cases	7,594		7,594

Entries are logistic regression coefficients. Year fixed effects are included in the model but not shown here. Years with split-party control of Congress are omitted.

p < .05.

My argument is that congressional performance evaluations affect elections by altering attitudes toward the majority party. If so, then when those attitudes are included as a predictor in the model, the effect of majority party favorability should be positive and significant, just as the literature shows, and the effect of congressional performance should diminish. Model 2 shows that this is the case. Although congressional performance evaluations do display some direct effects above those transmitted through brand attitudes,¹⁶ both the baseline and interactive coefficients for congressional performance evaluations are indeed smaller than in model 1. To verify that a significant portion of the effects of congressional evaluations shown in model 1 are transmitted through majority party favorability, I use the Preacher, Rucker, and Hayes (2007) bootstrapping method of moderated mediation analysis. Although the nature of the analysis prevents statements about the relative size of direct versus indirect effects (Preacher, Rucker, and Hayes 2007), the tests confirm that the mediated effects are indeed statistically significant.

Figure 3 graphically depicts the results of the moderated mediation analysis. The vertical axis represents the increase in the odds of voting for a presidential candidate from the same party as the majority in Congress if an individual approves of Congress as opposed to being neutral. The horizontal axis displays the actual range of congressional partisan polarization in the data. The solid, dark line plots the effect of congressional approval, mediated through majority party favorability, at each level of polarization. Dashed lines are bias-corrected bootstrap confidence intervals. With low partisan polarization in Congress, approving of Congress increases the odds of supporting the presidential candidate from the majority party by only 18 percent. With high partisan polarization in Congress, approving of Congress increases those odds by 28 percent.

Figure 3.

Effect of congressional performance evaluations, mediated by majority party favorability, on voting for president.

State Legislative Elections

In this section, I test whether partisan polarization in Congress increases the effect of congressional performance evaluations on the majority party’s prospects in state legislative elections. Neither the ANES nor any other survey organization has systematically collected survey data on voting in state legislative elections over an extended time period. Therefore, this analysis follows the lead of other studies in this area and focuses of partisan seat change (e.g., Rogers 2012; Simon, Ostrom, and Marra 1991). Specifically, I measure the percentage seat gain in each state’s lower chamber by the party that holds the majority in Congress.¹⁷

I measure the public’s evaluation of congressional performance using Stimson’s (1999) algorithm to combine the results from different survey questions on this topic into a single, quarterly measure that is comparable over time. To ensure that congressional performance ratings are driving voting outcomes and not the other way around, I use the rating from the third quarter prior to each election. This variable is interacted with the same congressional partisan polarization variable used throughout this study. For the mediation analysis, the Stimson algorithm is also used to create a single, yearly measure of majority party favorability and minority party favorability. Overall, omitting years of split-party control of Congress, there are a total of nineteen years between 1958 and 2010 for which suitable ratings of Congress and party favorability could be found.¹⁸

To guard against the possibility that any observed effects of congressional evaluations are spurious, the model controls for other factors that have been found to affect state legislative outcomes (e.g., Chubb 1988; Rogers 2012; Simon, Ostrom, and Marra 1991). Presidential evaluations are measured using the average Gallup approval rating of the president in the third quarter of the election year. The state of the national economy is measured as the percentage change in real per capita disposable income over the year prior to the third quarter of the election year.¹⁹ Just as with the analysis of presidential elections, both variables are recoded so that higher values are expected to increase the dependent variable. Because this recoding is based on whether the party of Congress and the president match, the model necessarily includes a control for this factor as well. As a control for seat exposure, I include a variable measuring the previous seat change in the legislature. I also control for the tendency of the president’s party to lose legislative seats in midterm elections.²⁰ As the data include repeated measures of the same units over time, I use hierarchical linear modeling (also called multilevel modeling) with a first-order autoregressive covariance structure (Peugh and Enders 2005). Data are weighted to account for the fact that there are more seats in some states than in others.

Table 4 presents the results. As expected in H2b, model 1 shows that the interaction between congressional performance evaluations and congressional polarization is positive and statistically significant. When there is higher partisan polarization in Congress, Congress’s performance ratings have a greater positive effect on the congressional majority party’s fortunes in state legislative elections across the country. In fact, the average effect of congressional performance evaluations appears larger than that of presidential performance evaluations when it comes to state legislative races. This finding is consistent with research showing that attitudes toward state legislatures are more closely associated with attitudes toward the national legislature than toward the president (Cook 1979).

Table 4.

Effect of Congressional Performance Evaluations on Partisan Seat Change in State Legislatures, 1958–2010.

	Model 1		Model 2
	Coefficient	(SE)	Coefficient	(SE)
Presidential performance evaluation	0.139*	(0.026)	0.112*	(0.030)
Economy	0.828*	(0.209)	0.661*	(0.253)
Congress and president of same party	−2.218*	(0.638)	−1.773*	(0.666)
Seat exposure	−0.014	(0.011)	0.008	(0.013)
Midterm penalty	−2.485*	(0.536)	−1.967*	(0.552)
Congressional polarization	0.000	(0.064)	0.095	(0.068)
Congressional performance evaluation	0.231*	(0.041)	0.157*	(0.049)
Congressional performance evaluation × Congressional polarization	0.017*	(0.007)	0.008	(0.008)
Majority party brand favorability			0.557*	(0.169)
Minority party brand favorability			0.097	(0.102)
Constant	2.125*	(0.445)	−30.192*	(8.101)
Akaike’s information criterion	−1,865.569		1,858.286
No. of cases	804		804

Entries are linear multilevel modeling coefficients. Years with split-party control of Congress are omitted.

p < .05 (one-tailed tests).

A final question is whether these effects of congressional evaluations are mediated through attitudes toward the majority party brand. For this to be true, a necessary condition is that majority party favorability must have a significantly positive effect on the same dependent variable. Model 2 demonstrates that this is the case. Using the Preacher, Rucker, and Hayes (2007) bootstrapping method of moderated mediation analysis, I am able to confirm that a significant portion of the effects of congressional evaluations shown in model 1 are transmitted to the dependent variable through majority party favorability.²¹

Figure 4 provides a graphical representation of the results of the moderated mediation analysis. The vertical axis represents the effect of a one-point increase in approval of Congress on the percentage of state legislative seats gained in a state by the party in control of Congress. The horizontal axis represents the actual range of congressional partisan polarization from the data used in Table 4. The solid, dark line plots the effect of congressional evaluations, mediated through majority party favorability, at each level of polarization. Dashed lines are bias-corrected bootstrap confidence intervals. The left side of the figure shows that with low levels of congressional partisan polarization, the mediated effect of congressional evaluations is negligible. In contrast, as expected, with greater partisan polarization in Congress, the mediated effect of congressional evaluations on state legislative elections grows strongly positive. In addition to illustrating this expected conditional relationship, the figure also helps show the substantive impact. At polarization’s peak, a ten-point increase in Congress’s job performance rating (about one standard deviation) produces an estimated 1.8 percentage-point increase in the share of a state’s lower chamber seats held by the party in control of Congress.

Figure 4.

Effect of congressional performance evaluations, mediated by majority party favorability, on partisan seat change in state legislatures.

Discussion

As parties in Congress grow more distinct from each other, evaluations of congressional performance have a greater effect on the brand favorability of the majority party. In turn, partisan polarization in Congress increases the role played by congressional evaluations in American elections both up and down the ballot. The consistency of the pattern of election results is impressive given the substantial differences in the types of races analyzed (local and national; executive and legislative; high salience and low salience) and in the types of data used in the analyses (votes and seats; survey data and official results).

These results expand our understanding of American politics in several ways. First, they upend conventional scholarly wisdom that citizens’ evaluations of Congress do not affect their attitudes toward and support for the parties overall. Second, for both presidential and state elections, the results introduce a new explanatory variable that, in certain cases, is as powerful as or more powerful than presidential approval. Third, the results help explain why strategic actors in the United States sometimes act counter to the interests of their own electoral base. Consider the example of Democratic leaders allowing passage of wiretapping legislation in 2007, and Republican leaders vowing not to shut down the government in 2015—stands that were relatively controversial among rank-and-file congressional party members but more popular in the general public. According to this study, broadly popular actions that boost perceptions of Congress can help the overall party brand, thereby boosting the electoral fortunes of all who run under the party label. Fourth, if the actions of legislators can affect the popularity of a party’s brand, as shown here, this suggests that these brands might be affected by other political actors as well. Future research might investigate, for example, whether perceptions of a party’s salient candidates—even those not yet in office—also carry important electoral implications for the entire party. Fifth, the results contribute to the literature on how citizens think about parties. A dominant view is that Americans hold stable psychological attachments to a particular political party that largely preordain vote choice (e.g., D. Green, Palmquist, and Schickler 2002). The results of this study help show that citizens sometimes also think about parties in more cognitively demanding ways, such as changeable brand images that are affected by government performance ratings.

The findings also inform debates by theorists of democracy about the conditions under which democratic governments are more likely to be responsive to citizenry. In particular, there is concern in the literature about the effect of partisan polarization (e.g., Ladewig 2014). My findings suggest both positive and negative impacts for democracy. On one hand, the results suggest that with greater polarization, a system like the United States may come a bit closer to the responsible party government model—one in which parties are held accountable for the acts of their agents in both branches of the national government, not just in the White House. On the other hand, recognizing that in a polarized environment perceptions of a strong congressional performance can benefit candidates from the majority party in various contests, members of the minority party would appear to have heightened incentives to obstruct and denigrate the work of Congress.

The results also speak to an important debate in the broader comparativist literature on political parties and elections. This field contains conflicting findings about whether greater party polarization increases the electoral role of concerns such as evaluations of party competence and party performance (e.g., Clark and Leiter 2014; Pardos-Prado 2012), or decreases it (e.g., J. Green 2007). To date, however, these studies have focused exclusively on European systems. The present study is the first to explore this general question from an American perspective. In doing so, it not only provides additional evidence pointing toward a positive rather than a negative relationship, but it also provides a potential framework to extend this research to other full-presidential systems.

Finally, one issue left unresolved by this study is the question of which particular aspects of congressional performance evaluations make the most impact on party favorability and elections. For example, the literature analyzing the sources of Americans’ evaluations of Congress distinguishes between the factors of congressional competence, congressional processes, congressional policy orientation, and national conditions (e.g., Jones 2014). As each of these factors has been shown to significantly influence Americans’ evaluations of Congress (Jones and McDermott 2009), future research should attempt to tease out which component of congressional performance evaluations transfers more readily to evaluations of the majority party overall, and thereby to partisan electoral preferences.

Footnotes

Appendix

Author’s Note

Data used in this article are available at .

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

Support for this project was provided by a PSC-CUNY Award, jointly funded by The Professional Staff Congress and The City University of New York (award #67612-00 45) and by funding from the Weissman School of Arts and Sciences at Baruch College, CUNY.

Notes

References

American Political Science Association, Committee on Political Parties. 1950. “Toward a More Responsible Two-Party System”. American Political Science Association.

Brewer

Mark D.

2009. Party Images in the American Electorate. New York: Routledge.

Butler

Daniel M.

Powell

Eleanor N.

2014. “Understanding the Party Brand”. Journal of Politics 76 (2): 492–505.

Campbell

Angus

Converse

Philip E.

Miller

Warren E.

Stokes

Donald E.

1960. The American Voter. Ann Arbor: University of Michigan Press.

Chubb

John E.

1988. “Institutions, the Economy, and the Dynamics of State Elections”. American Political Science Review 82 (1): 133–54.

Clark

Michael

Leiter

Debra

. 2014. “Does the Ideological Dispersion of Parties Mediate the Electoral Impact of Valence? A Cross-National Study of Party Support in Nine Western European Democracies”. Comparative Political Studies 47 (2): 171–202.

Cook

Timothy E.

1979. “Legislature vs. Legislator: A Note on the Paradox of Congressional Support”. Legislative Studies Quarterly 4 (1): 43–52.

Cox

Gary

McCubbins

Mathew

. 2005. Setting the Agenda: Responsible Party Government in the U.S. House of Representatives. New York: Cambridge University Press.

Dalton

Russell J.

2008. “The Quantity and the Quality of Party Systems: Party System Polarization, Its Measurement, and Its Consequences”. Comparative Political Studies 41 (7): 899–920.

10.

Fenno

Richard F.

Jr . 1978. Home Style: House Members in their Districts. Boston: Little, Brown.

11.

Glueck

Katie

. 2014. “Republican Governors Are Already Running against the Republican Congress”. Politico, November 11, http://www.politico.com/story/2014/11/gop-congress-2016-governors-112796.html.

12.

Green

Donald P.

Palmquist

Bradley

Schickler

Eric

. 2002. Partisan Hearts and Minds: Political Parties and the Social Identities of Voters. New Haven: Yale University Press.

13.

Green

Jane

. 2007. “When Voters and Parties Agree: Valence Issues and Party Competition”. Political Studies 55 (3): 629–55.

14.

Grynaviski

Jeffrey D.

2010. Partisan Bonds: Political Reputations and Legislative Accountability. Cambridge: Cambridge University Press.

15.

Hayes

Andrew F.

2013. Introduction to Mediation, Moderation, and Conditional Process Analysis: Methodology in the Social Sciences. New York: Guilford Press.

16.

Hetherington

Marc J.

2001. “Resurgent Mass Partisanship: The Role of Elite Polarization”. American Political Science Review 95 (3): 619–32.

17.

Jacobson

Gary C.

2012. “The President’s Effect on Partisan Attitudes”. Presidential Studies Quarterly 42 (4): 683–718.

18.

Jones

David R.

2010a. “Evaluations of Congress and Voting in House Elections: Revisiting the Historical Record”. Public Opinion Quarterly 74 (4): 696–710.

19.

Jones

David R.

2010b. “Partisan Polarization and Congressional Accountability in House Elections”. American Journal of Political Science 54 (2): 323–37.

20.

Jones

David R.

2014. “A More Responsible Two-Party System? Accountability for Majority and Minority Party Performance in a Polarized Congress”. Polity 46 (3): 470–92.

21.

Jones

David R.

McDermott

Monika L.

2009. Americans, Congress, and Democratic Responsiveness. Ann Arbor: University of Michigan Press.

22.

Ladewig

Jeffrey W.

2014. “Polity Symposium: Partisan Polarization and American Democracy”. Polity 46 (3): 407–10.

23.

Lee

Frances

. 2009. Beyond Ideology. Chicago: University of Chicago Press.

24.

Matthews

Donald R.

Prothro

James W.

1966. Negroes and the New Southern Politics. New York: Harcourt, Brace & World.

25.

Mayhew

David R.

1974. Congress: The Electoral Connection. New Haven: Yale University Press.

26.

McCarty

Nolan

Poole

Keith

Rosenthal

Howard

. 2006. Polarized America: The Dance of Ideology and Unequal Riches. Cambridge: MIT Press.

27.

Pardos-Prado

Sergi

. 2012. “Valence beyond Consensus: Party Competence and Policy Dispersion from a Comparative Perspective”. Electoral Studies 31 (2): 342–52.

28.

Peugh

James L.

Enders

Craig K.

2005. “Using the SPSS Mixed Procedure to Fit Cross-Sectional and Longitudinal Multilevel Models”. Educational and Psychological Measurement 65 (5): 717–41.

29.

Pope

Jeremy C.

Woon

Jonathan

. 2006. “The Dynamics of Party Reputations: Sources and Consequences”. Paper delivered at the APSA annual meeting, Philadelphia.

30.

Pope

Jeremy C.

Woon

Jonathan

. 2009. “Measuring Changes in American Party Reputations, 1939-2004”. Political Research Quarterly 62 (4): 653–61.

31.

Popkin

Samuel L.

1991. The Reasoning Voter. Chicago: University of Chicago Press.

32.

Preacher

Kristopher J.

Rucker

Derek D.

Hayes

Andrew F.

2007. “Addressing Moderated Mediation Hypotheses: Theory, Methods, and Prescriptions”. Multivariate Behavioral Research 42 (1): 185–227.

33.

Rahn

Wendy M.

1993. “The Role of Partisan Stereotypes in Information Processing about Political Candidates”. American Journal of Political Science 37:472–96.

34.

Ramirez

Mark D.

2009. “The Dynamics of Partisan Conflict on Congressional Approval”. American Journal of Political Science 53 (3): 681–94.

35.

Rogers

Steve.

2012. “Collective Accountability in State Legislative Elections”. Typescript, Princeton University, Princeton.

36.

Rohde

David W.

2013. “Reflections on the Practice of Theorizing: Conditional Party Government in the Twenty-First Century”. Journal of Politics 75 (4): 849–64.

37.

Simon

Dennis M.

Ostrom

Charles W.

Jr. Marra

Robin F.

1991. “The President, Referendum Voting, and Subnational Elections in the United States”. American Political Science Review 85 (4): 1177–192.

38.

Sinclair

Barbara

. 2006. Party Wars. Norman: University of Oklahoma Press.

39.

Stimson

James A.

1999. Public Opinion in America. 2nd ed. Boulder: Westview Press.

40.

Stokes

Donald E.

Miller

Warren E.

1962. “Party Government and the Saliency of Congress”. Public Opinion Quarterly 26 (4): 531–46.

41.

Trilling

Richard J.

1976. Party Image and Electoral Behavior. New York: John Wiley & Sons.

42.

Tufte

Edward R.

1975. “Determinants of the Outcomes of Midterm Congressional Elections”. American Political Science Review 69 (3): 812–26.

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

0.19 MB