The Member-Level Determinants and Consequences of Party Legislative Obstruction in the U.S. Senate

Abstract

Are Senators cross-pressured by party influence and constituent demands on procedural votes? We present a theory positing that Senators are indeed cross-pressured between these two demands and that electorally vulnerable members use the increased salience of procedural votes to differentiate themselves from their party. In an analysis of Senate procedural votes since the 92nd Congress, we develop a new spatial measure of obstruction preference and demonstrate our finding in three ways. First, obstruction preferences have become more polarized as the Senate experiences more intense legislative obstruction. Second, electorally vulnerable members are most likely to break from their party on procedural votes. Third, we find that voters electorally reward majority party Senators who break from their party and behave like minority party obstructionists. Our findings suggest that, unlike the House, Senate procedural votes are special opportunities for lawmakers to separate themselves from their party.

Keywords

legislative gridlock congressional polarization minority party obstruction Senate elections collective accountability

In their study of the House, Jessee and Theriault (2014) find that “not all roll-call votes are the same.” Some votes are anchored in the ideological divide between conservative and liberal policy positions while others are rooted in legislative procedure. The latter have become a larger focus of recent studies on Congress. Lee (2009) finds that many of today’s political battles, particularly in the U.S. Senate, are fought on partisan rather than ideological terms with respect to policy items. As a consequence, procedural votes are more consequential and salient to the majority party’s ability to control the legislative agenda and overcome partisan gridlock (Den Hartog & Monroe, 2011; Lee, 2009; Smith, Ostrander, & Pope, 2013). In this article, we ask if the ramifications from partisan pressures of procedural votes are similar to the repercussions from partisan pressures on final passage votes. To do this, we turn to the contemporary Senate, where legislators garner far more procedural power than their counterparts in the House and where procedural votes are more salient to the legislative process.

We build on recent studies of the Senate by focusing on the individual-level dynamics faced by Senators on procedural votes and the potential electoral incentives they face in this realm of congressional voting. Using comprehensive roll call data on Senator procedural votes and a unique data set on Senate election outcomes, we assess the following two questions critical to understanding the determinants and consequences of legislative obstruction in the contemporary Senate: (a) Do Senate procedural votes provide Senators an opportunity to differentiate themselves from their party? and (b) Do Senators face punishment from voters for joining their party on procedural votes? In this analysis, we explore these questions using new measures of Senator preference for legislative obstruction. We theorize that vulnerable Senators view procedural votes as an opportunity to separate themselves from their party and reduce the electoral costs of partisan loyalty (Carson, Koger, Lebo, & Young, 2010). For minority party Senators, procedural votes provide an opportunity to break with their party and the negative associations with legislative obstruction and gridlock (Hibbing & Theiss-Morse, 1995). By contrast, procedural votes provide majority party Senators the opportunity to differentiate their electoral brand from their national partisan brand. Our analysis provides two principal pieces of evidence to support our theoretical framework. First, we find that minority party Senators break with their party on procedural votes in the Congress preceding their reelection bid by sporting more of an anti-obstruction legislative preference than their minority party colleagues not facing a reelection bid. This differentiation is more salient for minority party Senators representing electorally marginal states and states with a partisan preference for the majority party. Second, majority party Senators who fail to break with their party on procedural votes face electoral consequences for their partisan loyalty on legislative procedure in their subsequent reelection bid. This electoral implication is especially pronounced by majority party Senators representing states with a clear partisan preference for the minority party.

Our theoretical framework of the electoral dynamics of legislative obstruction in the Senate builds on recent work positing that procedural battles are salient events that draw the attention of voters. Indeed, the 60-vote threshold in the Senate to overcome the legislative filibuster is often cited as the strongest veto point in the legislative process, with most legislation requiring 60 votes de facto (Fong & Krehbiel, 2018). Scholars note that the growth of legislative gridlock and partisan conflict, particularly in the Senate (i.e., Binder, 2011; Binder, Lawrence, & Smith, 2002; Koger, 2010), not only draws the attention of the mass public, but also informs their political attitudes. To that point, the greater salience of obstruction in the contemporary Senate coincides with lower congressional job approval assessments (Flynn & Harbridge, 2015; Ramirez, 2009), lower trust in the national government (Hibbing & Theiss-Morse, 2001), and lower public confidence that Congress can fulfill its policymaking constitutional prerogative (Jones, 2015; Keele, 2007). Moreover, recent work finds that voters, particularly those who are politically sophisticated and strong partisans, base their views of legislative procedure on their preference for the underlying policy subject to procedural maneuvering (Doherty, 2015; Park & Smith, 2016; Smith & Park, 2013). Taken together, recent work suggests that voters are not only aware of growing obstruction, but are using these events to inform political attitudes. As a consequence, Senate procedural votes act as the ideal opportunity for both vulnerable members to make a statement about their loyalties or for voters to determine whether the lawmaker is beholden to her constituency or her party. Given that overcoming procedural obstruction requires a super majority, this opportunity is evident for electorally vulnerable members (e.g., a majority Senate Democrat in a Republican-leaning state) who often end up as the key votes in Senate procedural votes that determine the fate of legislation.

We move forward by introducing our theoretical framework and assessing the relatively high cost of overcoming legislative obstruction in the U.S. Senate compared with the U.S. House. Similar to previous scholars (e.g., Den Hartog & Monroe, 2011; Smith et al., 2013), we argue that legislative organization in the Senate is predicated on almost complete unity in legislative preference with operationally no room for individual majority party defection given the greater challenge by the Senate majority party to control the legislative agenda (Campbell, Cox, & McCubbins, 2002). We then present our theory arguing that this greater emphasis on unified legislative preference provides Senators with a salient opportunity to differentiate themselves from their party, with the hopes of reaping electoral benefits from a more distinct personal brand. Next, we introduce our measure of Senator preference for legislative procedure. Then, we use procedural voting preferences in two steps. First as a dependent variable, and second as an independent variable in a model of reelection vote-shares. The first model shows that Senators up for reelection become less polarized in their procedure preferences. The second set of models show that Senators who associate with their party on procedural votes face negative electoral consequences. Finally, we consider our findings in the context of models of legislative organization in the House.

Public Opinion and Senate Procedure

A majority of the work on American party organization has occurred in the House. The prevailing argument is that members of a party are willing to hand over power to party leaders to obtain group victories and thus improve the party brand (Aldrich & Rohde, 2000; Cox & McCubbins, 2005, 2007). However, individual lawmakers face a backlash from voters for choosing the party over constituents (Carson et al., 2010; Lebo, McGlynn, & Koger, 2007; Patty, 2008). This central tension, a collective action problem, sets up the basis of party control in the House. Individual members are willing to give up the least amount of power possible to secure group victories. Do these findings carry to the Senate?

Party organization in the Senate is not as well known as it is in the House. There are fundamental differences in the rules that make organization more difficult. In the Senate, there is not a powerful Rules Committee that sets the parameters of debate on legislation like there is in the House. Instead, there is a 60-vote threshold to end debate. This makes procedure in the Senate less predictable and party cohesion more important to accomplish any large legislative achievement. As we argue later, the less organized nature of the Senate makes procedural votes more salient than procedural votes in the House. We know that parties in the Senate still desire to achieve cohesion to create group benefits. Senators, like their counterparts in the House, are cross-pressured by their party and constituents. Recent studies, such as Doherty (2015), show that voters prefer that Senators place the needs of their state over federal issue domains. Smith and Park (2013) find that Americans place more importance on their own preferences than they do for ideas of minority rights in the Senate. This is to say that Americans understand that the Senate is a nonmajoritarian institution, but still prefer that their representatives break with the party when the party’s desires split with the state’s.

Our contribution to this work is twofold. First, we ask whether Senators are aware that their procedural votes affect their standing with voters. We find evidence that vulnerable Senators do indeed break with their party more on procedural votes in the term preceding an election. Second, we ask whether Senators who do vote the partisan line on procedural votes are punished by constituents in a general election. We find evidence that indeed vulnerable members are punished for voting the party line on procedural votes. Our findings show that party organization in the Senate is made more difficult by voters. The next section highlights why Senate procedural votes are more salient to voters.

Salience of Procedural Votes

Recent scholarship into the determinants of congressional roll call voting uncovers differences in how members vote on procedural votes and substantive policy votes (Ansolabehere, Snyder, & Stewart, 2001; Roberts & Smith, 2003; Rohde, 1991; Theriault, 2008). The intuition behind the observed difference is that procedural votes are critical votes for congressional parties but are not salient to the public. To advance their legislative agenda, the majority party cannot afford significant defections from rank-and-file members on procedural votes, given that final passage votes are conditional on successful passage of procedural motions (Roberts & Smith, 2003). However, these findings focus on the House. The imperative for unity on procedural votes is particularly crucial in the contemporary Senate, as most important legislation needs to overcome the 60-vote threshold of the filibuster and other forms of procedural obstruction to consider legislation on the floor (McCarty, Poole, & Rosenthal, 2006, Chapter 6). For the minority party, it is also important to hold the line on procedural votes. If successful in defeating the majority party on procedural votes, the minority party may extract policy concessions for cooperation on the next round of procedural votes, moving legislation closer to their collective preferences (Green, 2015).

We argue that the lessons learned from the House are that procedural votes are of little or no electoral consequence to the political fortunes of Senators and that they are less salient to voters relative to final passage and substantive policy votes. Moreover, we argue that procedural votes present a unique opportunity for Senators looking to differentiate themselves from their party that final passage votes do not. Procedural votes in the Senate, unlike those in the House, garner large amounts of media attention (Binder, 1997; Matthews, 1959). Indeed, Rohde (1991) articulates this institutional difference in procedural obstruction by arguing that “the potential for extended debates (commonly referred to as filibusters) has no parallel in the House, and it vests considerable power in minorities” (p. 178). In many cases, majority party leadership is forced to invoke cloture in an attempt to quell minority-led opposition to bringing legislation to the floor. Individual Senators can use this procedural opportunity to signal opposition to majority-backed legislation that could be politically unpopular back at home. Scholars note that the procedural cartel theory, with respect to gaining sufficient loyalty on procedural motions, is weaker when applied to the Senate (Campbell et al., 2002). This is due to the fact that Senators defect on procedural votes with greater frequency than their House colleagues and the Senate majority party must generally make policy accommodations to the minority to advance their agenda (Rohde, 1991; Smith et al., 2013).

Moreover, procedural votes, particularly in the form of cloture votes, are becoming a larger part of the Senate legislative calendar. Figure 1 shows the total proportion of Senate roll call votes from 1971 to 2016 that are classified as procedural votes in the form of tabling and cloture motions. Following the lead of Den Hartog and Monroe (2011), we consider amendment votes as substantive votes on policy, similar to final passage votes, and consider tabling and cloture motions as procedural obstruction measures.

Figure 1 shows that the nature of obstruction in the Senate has undergone a significant change since the 1970s. On average, procedural votes in the form of cloture and tabling motions represent a greater proportion of votes than in the past, a finding corroborated by similar analysis of Senate roll call voting (Smith et al., 2013). In contemporary Congresses, the tabling motions have given way to cloture votes, a higher profile and more costly form of obstruction (Binder et al., 2002; Grimmer, 2010). While tabling motions may be considered minor threats to the ability of the majority to set the legislative agenda, filibusters are certainly high-profile events. Confronted with a direct threat to agenda-setting power given the presence of the filibuster, majority party leaders are forced to either concede defeat or invoke cloture to break the filibuster (Lee, 2009). In many cases, majority party leaders force votes on cloture to “effectively communicate what a congressional majority party stands for, even when they achieve nothing in policy terms . . . and dramatize being blocked by the opposition” (Lee, 2016, p. 66). The implications of the changing nature of obstruction, from tabling motions to cloture votes, increase the salience of obstruction in the contemporary Senate. We contend that votes seeking to curtail legislative obstruction, much like votes on substantive policy, are salient votes that attract significant attention and have bearing on a Senator’s electoral fortunes.

Figure 2 articulates this point by showing the quarterly trend of U.S. Senate filibuster news story mentions in the New York Times from 1971 to 2016. Given the variation over time in news data quality across various news sources, we follow the lead of other scholars (e.g., Althaus, Edy, & Phalen, 2001; Boydstun, 2013; Woolley, 2000) and focus on the New York Times data archive provided by the LexisNexis database. We focus on coverage of Senate filibusters, a costly form of obstruction by the minority party and one that requires direct action by the Senate Majority Leader, in the form of invoking a cloture motion, to overcome.¹ As Figure 2 shows, the growth in media coverage of Senate filibusters follows the same positive trend as the growth in Senate cloture votes and the proportion of Senate votes that are procedurally articulated in Figure 1. While salient to a certain degree from the 1970s to the 1990s, mentions of the Senate filibuster grow dramatically during the 2000s, particularly after the election of President Barack Obama and a filibuster-proof Democratic Senate Majority in 2008.² Moreover, this growing salience in media coverage about the Senate filibuster coincides with a dramatic increase in the proportion of the Senate legislative calendar dedicated to overcoming minority-led Senate filibusters. Taken together, it is clear that to a certain degree, Senate obstruction has always been salient and this salience is increasing as the Senate is confronted with growing levels of obstruction.

Expectations of Senate Member and Voter Behavior on Procedural Votes

Our theory considers procedural voting in the Senate in two different domains. The theory first addresses how Senators act while casting procedural votes, while also addressing how voters will react to procedural votes at the voting booth. Our expectations for the first stage relate to both the salience of Senate procedural votes and the need for Senators to differentiate themselves from their party. We view procedural votes as the ideal means for Senators to demonstrate to their constituents that they are not controlled by their party.

We expect that Senators will distance themselves from their party in the term preceding their reelection campaign. We use prior studies of party unity on final votes to inform our expectations for procedural votes in the Senate. In the House, several theories of party organization state that parties are only as cohesive as needed due to the potential harm caused to members forced to vote against constituent demands (Aldrich & Rohde, 2000; Carson et al., 2010; Lebo et al., 2007; Patty, 2008; Volden & Bergman, 2006). Several of these theories (Aldrich & Rohde, 2000; Patty, 2008; Rohde, 1991) show that members will limit the amount of power they give to party leadership to minimize the repercussions from following party orders. It is this direct tension between the benefits of party unity (winning close votes) and the costs (voter punishment for conformity) that are the underlying assumptions that drive our theory.

By casting votes in the 2 years prior to a campaign, a Senator can distance herself from her party on the most recent issues. Therefore, in the two terms not leading up to reelection, we expect a Senator to move in the opposite direction from her partisan median when looking at procedural preferences.

Senators, like their counterparts in the House, are cross-pressured by the needs of their party and the needs of their constituents. Senators would like to benefit from the collective benefits derived from party victories but would still like to appear as serving their constituents over their party. These two interests should collide when a Senator’s constituents prefer a different outcome than the median member of a Senator’s party. These cross-pressures are increased during procedural votes. In the Senate, a defection is much costlier to majority party control during procedural votes than final passage votes due to the 60-vote threshold. However, due to the salience of Senate procedural votes, these votes provide Senators with an ideal situation deviate from their party.

Which members will then choose constituents over party? We expect Senators facing difficult reelection to show the strongest procedural defection in the reelection term. There is considerable variation in electoral vulnerability across Senators. Senators from states in which their party enjoys a stark partisan advantage, such as Oklahoma’s Republican Senators, have little fear of an opposite party challenger. By contrast, Senators representing states in which their party faces a partisan disadvantage, such as Sen. Joe Manchin (D-WV), very much need to distance themselves from their party in fear of losing the upcoming election. Therefore, we expect Senators facing similar electoral dynamics as Senator Manchin to be more likely to break with their party on procedural votes than Oklahoma’s Republican Senators.

Our theory then considers the electoral ramifications of the procedural voting records held by Senators. We believe that voters will punish Senators who are too partisan and do not deviate from their party’s procedural preference. Since Senators are under high scrutiny from their constituents during high salience procedural votes, we expect Senators who vote in line with their party on procedural votes to be punished by their constituents. Similar to our theory on Senator defection on procedural votes, we expect the punishment to be more severe for Senators who sit in vulnerable seats.

Data and Analysis

Measuring Senator Obstruction

To evaluate the empirical effect of obstruction on electoral support, we specify a Senator-level measure of obstruction preference. We agree with Cox and McCubbins (2005) that majority party leadership places greater weight on procedural votes and that they demand greater partisan loyalty on these votes compared with votes on final passage. However, we challenge the notion that procedural votes are less salient than final passage votes in the context of the U.S. Senate. The following section explains the theoretical motivation and empirical estimation of our main independent variable, a measure of Senator preference for obstruction.

Analysis of congressional roll call voting assumes a spatial model. Legislators have an ideal point in which they want policy to move. This ideal point is located in a latent ideological space containing all legislator ideal points, with this space usually defined in terms of liberal or conservative preferences (Carroll, Lewis, Lo, Poole, & Rosenthal, 2013; Poole, 1985). The model dictates how legislators will vote on policy proposals, with the goal of moving policy as close to their ideal point as possible, thus maximizing their utility. Legislators evaluate policy on an ideological scale, relative to the status quo, and will vote for policies that move the status quo closer to their ideal point or vote against policies that move the status quo further away from their ideal point (Carroll et al., 2013).

While application of the spatial model is straightforward with respect to policies, application of the model is less straightforward with respect to procedural votes. Legislators may treat procedural votes as de-facto policy votes, voting in favor of procedural votes that advance their preferred policy to the floor (Groseclose & Milyo, 2010; McKelvey & Niemi, 1978). In this specification, the same ideological ideal point that dictates legislator preference on substantive policy votes will also dictate legislator preference on procedural votes. On the contrary, Senators may exhibit diverging ideologies when voting on policy and procedural votes. For example, Senators Mark Pryor (D-AR) and Mary Landrieu (D-LA) voted for cloture on the Supreme Court nomination of Samuel Alito, but ultimately against confirmation (Seo & Theriault, 2012). Moreover, both Landrieu and Pryor represented states that went decisively for President George W. Bush, the president who nominated Alito to the Supreme Court, indicating perhaps a potential cost to voting against the salient cloture vote. Since Alito’s confirmation was a foregone conclusion once cloture was invoked, the vote for final passage was less salient, leaving both Senators unconstrained on the confirmation vote (Seo & Theriault, 2012).

To assess whether Senators really do have obstruction preferences that are different than preferences on policy, we follow the same methodological approach that Jessee and Theriault (2014) take in the context of the U.S. House. First, we code all procedural votes from the 92nd to the 114th Congresses.³ Following the lead of other work measuring obstruction in the Senate (e.g., Wawro & Shickler, 2004), we follow the coding scheme for identifying procedural votes as defined by Binder and Smith (1997). We consider tabling motions, cloture votes (i.e., motions to end debate), and motions to recess as procedural votes designed to put an end to disruption and obstruction of the majority party’s legislative agenda. The overwhelming number of these votes are tabling motions and cloture votes.

Once we code all procedural votes, we specify a procedural vote matrix and scale legislator preferences on obstruction using Poole and Rosenthal’s (1997) W-NOMINATE binary choice model for each Congress.⁴ We specify the model so that positive scores indicate a greater preference for obstruction and delaying the legislative process, and negative scores indicate less of a preference for obstruction. Given the W-NOMINATE model, these scores are bounded in a range from −1 (anti-obstruction) to 1 (pro-obstruction). To orient the scale, we expect the Senate majority whip to have a high negative score (indicating a strong anti-obstruction preference) and the Senate minority whip to have a high positive score (indicating a strong pro-obstruction preference) given that it is the formal role of these party leaders to toe the party line and convince their caucus to do the same (Cox & McCubbins, 2007).⁵ We also repeat the scaling procedure for the nonprocedural votes such as votes on amendments, nominations, and final passage. Indeed, these roll call votes are purged of procedural votes and allow for the manifestation of ideological cleavages that define the modern Congress (Jessee & Theriault, 2014). The resulting ideal point derived from this estimation of the nonprocedural matrix can be interpreted as liberal-conservative ideological preferences on substantive policy votes.⁶ In the case of amendments, this represents an opportunity to vote and potentially influence the ideological direction of a legislative proposal under consideration by the floor (Smith et al., 2013). We again follow the insights provided by Jessee and Theriault’s (2014) scaling of House votes by using the W-NOMINATE model to derive legislator ideal points on nonprocedural policy items.

Figure 3 presents scatter plots of our estimated obstruction and ideological ideal points by Congress, with the gray dots indicating Democratic Senators and black dots indicating Republican Senators. This presents the first scaling of procedural votes in the contemporary Senate and the results of the scaling procedure yield insight into the nature of roll call voting by Senators. We find that the two spatial measures move across time in similar ways, with Senators polarizing both on procedural and explicitly ideological grounds.⁷ In the 92nd Congress, members from both parties overlapped on both the ideological and obstruction dimensions. Following well-known trends in polarization, the 114th Congress shows no overlap in either the procedural or ideological dimension (Gailmard & Jenkins, 2008). Also, the extremity of polarization over time appears to be even on both dimensions. This would mean that party effects on member behavior are not stronger in one dimension than they are in the other.⁸

Another key insight of the obstruction measure is that it is predicated on which party is in the majority. Indeed, Republicans and Democrats exhibit a spatial preference for obstruction when they are in the minority, as reflected in the positive values on the procedural dimension. Recall that we also expect the party whips to have extreme ideal points on procedural votes, with a negative score indicating an anti-obstruction preference and a positive score indicating a pro-obstruction preference. The whips also have scores in the expected direction during this Congress. Senate Majority Whip John Cornyn (R-TX) features an obstruction ideal point of −0.68 while Senate Minority Whip Dick Durbin (D-IL) holds a score of 0.75. Moreover, as procedural votes become more salient during the contemporary Congresses, there is less heterogeneity in procedural preferences of the two parties (particularly the majority), underscoring the solidification of procedural preferences within both parties (Lee, 2009; Smith et al., 2013).⁹

Anticipating an Electoral Cost of Senator Obstruction

Our theory posits that Senators are well aware that being out of step with their constituents carries potential electoral costs. Given the high-profile nature of Senate elections, Senators who are out of step on salient procedural votes may draw higher quality and more spirited challengers. Previous studies of congressional gridlock note that citizens place a normative value on representatives who seek common ground on procedural matters rather than those who engage in partisan-motivated obstruction (Flynn & Harbridge, 2015; Harbridge & Malhotra, 2011). As a result, we expect that minority Senators anticipate a cost to obstruction, and thus change their obstruction preference closer to the majority party if they are up for reelection. Conversely, majority Senators may be anticipating a gain and thus change their obstruction closer to the minority party line. To state this anticipatory hypothesis more formally,

Hypothesis 1 (H1): We expect Senators entering their reelection cycle to move their obstruction preferences closer to the opposite party’s preferences compared with Senators who are not up for reelection. Majority Senators become more pro-obstructionist if they are up for reelection and minority Senators become more anti-obstructionist if they are up for reelection.

We expect the anticipation hypothesis articulated in H1 to be conditioned by the type of state a given Senator represents. Senators representing states that lean toward the opposing party have a strong incentive to change their obstruction in the face of a reelection bid compared with Senators who represent marginal or safe states. Senators from safer states face differing incentives to change their obstruction compared with Senators from out-partisan states and marginal states. These Senators can toe the party line with relative impunity while their colleagues from more electorally volatile states cannot, especially given the high-profile nature of procedural votes. This line of reasoning leads to the following hypothesis:

Hypothesis 2 (H2): Senators up for reelection and from out-partisan and marginal states should change their obstruction preferences to move away from their own party’s procedural preferences while Senators from safer states should not. Majority Senators from marginal states or states that lean toward the minority should be more anti-obstructionist than their colleagues in safer states. Minority Senators from marginal states or states that lean toward the majority should be more pro-obstructionist than their colleagues in safer states.

To test the H1, we specify a regression model evaluating whether Senators change their obstruction in the 2 years preceding an election in anticipation of an electoral sanction for their legislative procedural behavior. The dependent variable is the Senator’s obstruction preference score for a given Congress. The ordinary least squares (OLS) model is specified below and the unit of analysis is a Senator serving in a given Congress.¹⁰

\begin{matrix} {\hat{Obstruct}}_{t} = α + β_{1 t} \times {Election}_{t} + β_{2 t} \times {MinoritySen}_{t - 1} + \\ β_{3 t} \times ({Election}_{t} \times {MinoritySen}_{t - 1}) + β_{i t} \times C_{t - 1} + ε_{i t} . \end{matrix}

(1)

We include a binary variable, ${Election}_{t}$ , indicating whether the Senate seat is up for election at the conclusion of the Congress. For example, all Senators in Class III Senate seats are coded as 1 for the 114th Congress (2015-2017) since this seat class was up for election in the 2016 election cycle. We capture Senator majority status using the ${MinoritySen}_{t - 1}$ variable, coded 1 if the Senator is in the minority for the given Congress and coded 0 if the Senator is in the majority. To test whether Senators moderate their obstruction preference if they are up for reelection, with the direction of moderation conditioned by majority status (H1), we include the interaction term ${Election}_{t} \times {MinoritySen}_{t - 1}$ into the model. Given our argument that minority Senators up for reelection move their obstruction preferences in the direction of the majority, this interaction should be significant and negative given that lower scores of obstruction indicate a more anti-obstruction preference. Since we also posit that majority party Senators become more pro-obstructionist if they are up for reelection, we expect the constituent term of ${Election}_{t}$ to be positive and significant, since higher levels of obstruction indicate a more pro-obstruction preference.¹¹ We evaluate the model using data from the 92nd to the 114th Congress, the Congresses preceding the 1972 and 2016 elections, respectively.

We turn to Figure 4, which articulates the interactive effect of majority status and election year on a Senator’s obstruction preference. Figure 4 provides partial evidence for H1.¹² There is a significant difference between the obstruction preferences of minority Senators who are up for election and those who are not. Minority Senators with seats up for election change their obstruction preference in the majority’s direction, toward the anti-obstruction position.¹³ Indeed, the predicted obstruction preference for minority party Senators not up for reelection is approximately 0.57 while, by contrast, the predicted obstruction preference for minority party Senators up for reelection is approximately 0.50. Given that scaling of the obstruction preference is bounded from −1 to 1, this seven-unit difference presents a modest 4% difference in obstruction preferences between minority party Senators who are up for reelection and those who are not.¹⁴ However, there is no significant difference in obstruction preferences between majority Senators who are up for election and those who are not.¹⁵ This observation, that majority party Senators do not adjust their obstruction preference when they are up for election while minority party Senators do, is evidence that it is far more critical for the majority party to secure support on procedural votes than it is for the minority party (Cox & McCubbins, 2005).

After all, it is the majority party’s agenda at stake in these procedural votes, with passage of the agenda providing the record of accomplishments by which all party members seek reelection. This can raise the stakes for majority party leaders to secure the votes to move the agenda through the legislative process. Minority Senators may fear the electoral consequence of being branded as an obstructionist at home, thus feeling compelled to toe more of an anti-obstruction preference in the Congress preceding their reelection bid (Flynn & Harbridge, 2015; Hibbing & Theiss-Morse, 1995).

Although we find partial evidence that Senators adjust their obstruction preference in the wake of a potential campaign, it is unclear which type of Senators are more susceptible to defecting from their party on obstruction preference. H2 posits that the Senators we should observe defecting from their procedural preferences toward to opposite party in the wake of a potential campaign are those who represent marginal seats and seats that lean toward the opposing party. To evaluate this hypothesis, we define these three distinct electoral categories of seats a given Senator may represent by recoding state partisanship into two dummies, one specifying out-partisan seats and the other specifying safe seats (marginal seats are specified as the omitted category).¹⁶ Following previous studies of Senate elections (e.g., Highton, 2000), our measure of state partisanship is the mean of the incumbent party’s two-party presidential vote-share in the preceding three presidential elections, thus controlling for short-term partisan swings.¹⁷ Senators representing states that lean 5% or more toward the opposing party are coded as holding out-partisan seats. Senators representing states that lean 5% or more toward their own party are coded as holding politically safe seats. Finally, Senators representing states in which no party has at least a 5% advantage are coded as holding marginal seats.¹⁸ To avoid a triple interaction and for ease of interpretation, we subset the data to include only seats up for election, holding election year constant.¹⁹ We interact these dummy variables with majority status to evaluate the hypothesis that the extent by which majority and minority Senators change their procedural preference is conditioned by their state’s partisanship.

Figure 5 provides evidence that Senators adjust their procedural preferences ahead of election time depending on the partisan composition of their states (H2). For majority and minority Senators, there are significant differences in the predicted obstruction preference between Senators holding seats that lean toward the opposing party and their colleagues representing politically safe seats. The same holds between minority party Senators who represent marginal seats and their safer seat colleagues. There are no significant differences in obstruction preference between Senators in out-partisan seats and marginal seats for both parties. Majority Senators up for election, and from out-partisan (minority-leaning) seats, have significantly more pro-obstruction preferences than their safer seat colleagues. Indeed, safer seat majority Senators have approximately 9% more pro-obstruction scores than their colleagues representing states that lean toward the opposing party.²⁰ This indicates that majority Senators from states that lean toward the minority and those in marginal seats anticipate a cost if they toe the party line on obstruction, like their safer seat colleagues appear to do.

The same story holds true for minority party Senators. Minority Senators up for election and from out-partisan (majority-leaning) or marginal seats have significantly more anti-obstruction preferences than their safer seat colleagues. To that end, safe-seat minority Senators have approximately 13% and 17% more anti-obstruction procedural preferences than their marginal and out-partisan co-partisans, respectively.²¹ This indicates that both majority and minority Senators, to a certain degree, are responsive to potential state preferences on obstruction.

It is important to note that the difference in obstruction preference between out-partisan seats and safer seats is greater in the minority party ( $\approx 0.34$ ≈ 0.34) than the majority party ( $\approx 0.17$ ≈ 0.17). This is not to suggest that minority party Senators are more responsive to state preferences on obstruction or that they anticipate an electoral cost over being out of step with respect to obstruction. We contend that these differences are part of institutional differences between the majority and minority party in the Senate (Cox & McCubbins, 2005). Since the majority party sets the legislative agenda, party leadership brings forth procedural votes in response to minority party obstruction. Defeat in procedural votes for the majority party is a far costlier proposition than for the minority party, given that it directly reduces the legislative achievements their party can run on. Thus, it is more critical for majority party leadership to secure obstruction loyalty among their members than the minority party. Although majority party Senators in politically unfriendly states can shirk party leadership with respect to obstruction preferences, they may be less able to compared with their minority party colleagues in similar, unfriendly, political circumstances.

Modeling the Electoral Consequence of Senator Obstruction

The previous analysis provides evidence that Senators anticipate an electoral cost for obstruction and, conditional on the partisan preferences of their constituents, change their obstruction in the Congress preceding their reelection bids. However, many Senators may fail to distinguish themselves from party leadership adequately enough on procedural grounds. These Senators who fail to change their obstruction preference and are out of step with constituent preferences on obstruction stand to pay an electoral cost.

This expectation is of particular concern for majority party Senators, given the partisan pressures from party leadership to toe the procedural line rather than aid in the minority’s efforts to defeat the majority’s collective agenda (Campbell et al., 2002). One of the expectations of the theoretical framework we present posits that the effect of Senator obstruction preference on electoral support is conditioned by majority status. Given that the majority party leadership sets the legislative agenda and the successful implementation of this agenda is predicated on securing enough support for procedural votes to bring agenda items to the floor (Cox & McCubbins, 2005), procedural votes provide a salient opportunity for majority party Senators to break from their party. In some cases, the salient opportunity presented allows them to be the pivotal vote (Fong & Krehbiel, 2018). Defection from leadership by majority party Senators signals to voters back home that they are willing to buck party leadership in favor of being responsive to state preferences. Likewise, for minority party Senators, they may break from their party and join the majority on procedural votes to advance legislation favorable to their constituents. Thus, majority party Senators defect from their party by having a greater preference for obstruction while minority party Senators defect from their party by having a lesser preference for obstruction. These defections pay off at election time, as differentiated obstruction preferences relative to party leadership help Senators garner support from partisans of the opposing party during reelection bids. Conversely, Senators with obstruction preferences more congruent with party leadership do not garner the same crossover support. Stated formally,

Hypothesis 3 (H3): Majority party Senators with a greater preference for obstruction should garner more electoral support than majority party Senators with less of a preference for obstruction. Congruently, minority party Senators with less of a preference for obstruction should garner more electoral support than minority party Senators with more of a preference for obstruction.

Our theory also posits that the effect of obstruction on Senator reelection vote-shares is not only conditioned by majority party status, but also by the partisan nature of the state a given Senator represents. If one assumes that, like Senators, the pivotal median voter possesses an ideal point on obstruction, such positioning by this class of Senators will represent positioning toward the median voter. By contrast, Senators representing seats politically safe for their party gain nothing from defecting their party on obstruction and aligning themselves with the other party, given that such positioning would result in moving further from the pivotal median voter. Stated formally:

Hypothesis 4a (H4a): Senators representing states that lean decisively toward the opposing party or marginal states should be rewarded for differentiating themselves from their party on obstruction preferences. Majority party Senators with more of a preference for obstruction and from states that lean toward the minority party (and marginal states) should see an increase in electoral support compared with majority party Senators in the same circumstance and a greater preference for obstruction.

Hypothesis 4b (H4b): Conversely, minority party Senators with less of a preference for obstruction and from states that lean toward the majority party (and marginal states) should see an increase in electoral support compared with minority party Senators in the same circumstance and a greater preference for obstruction.

Using data on Senate reelection bids from 1974 to 2016, we specify the following OLS regression model below that tests H3.²²

\begin{matrix} {\hat{IncVote}}_{t} = α + β_{1 t} \times {Obstruct}_{t - 1} + β_{2 t} \times {MinoritySen}_{t - 1} + \\ β_{3 t} \times ({Obstruct}_{t - 1} \times {MinoritySen}_{t - 1}) + β_{i t} \times C_{t - 1} + ε_{i t} . \end{matrix}

(2)

The dependent variable of the model is the Senator’s two-party vote-share during their reelection at time $t$ . The main explanatory variables of interest are minority status of the Senator and the Senator’s estimated obstruction preference. The variable capturing minority status, $MinoritySen$ , is coded as 1 if the Senator was in the minority caucus during the Congress in which their reelection bid takes place. For example, a Democratic Senator seeking reelection during the 2016 elections is coded 1, indicating minority party status, while Republican Senators seeking reelection are coded as 0, indicating majority party status. The variable measuring obstruction preference is the previously discussed Senator-level obstruction ideal point for the Congress in which the Senator is seeking reelection. For example, a Senator seeking reelection in 2016 will have an obstruction preference score accumulating during her service in the 114th Congress (2015-2017). Our model includes controls for salient predictors of election outcomes, such as (articulated as $C_{t - 1}$ in the model) presidential approval, state partisanship, midterm election year, incumbent candidate quality advantage, seat partisanship, lagged quality advantage, and lagged Senate vote-share. We also control for national congressional approval, a known indicator of public perceptions about congressional conflict (e.g., Durr, Gilmour, & Wolbrecht, 1997; Ramirez, 2009) and congruence between congressional policy outputs and national public mood (Ramirez, 2012).²³ Detailed coding descriptions of these control variables can be found in the appendix.²⁴

The model includes an interaction between minority status and obstruction preference, ${Obstruct}_{t - 1} \times {MinoritySen}_{t - 1}$ , that evaluates the effect of obstruction, conditional on minority status, on vote-shares as articulated in H1. The coefficient on this multiplicative term should be negative and significant and the coefficient on the minority status constituent term, ${MinoritySen}_{t - 1}$ should be positive and significant.²⁵ If both coefficients are significant and in the hypothesized direction, this would indicate a loss in support as minority party Senators become more congruent with party leadership in obstruction preference (i.e., more obstructionist). The constituent term on obstruction, ${Obstruct}_{t - 1}$ , should be positive and significant, indicating a gain in support as majority party Senators defect from party leadership with respect to obstruction (i.e., less obstructionist).

The results of the quantities of interest are shown in Figure 6 and full interactive model results are presented in Table A2 of the appendix. Figure 6a shows that the effect of obstruction on a Senator’s reelection vote-share is significantly conditioned by majority and minority status of the Senator. In evaluating H3, we find support that majority party Senators are rewarded as they move their obstruction preference from anti-obstruction to pro-obstruction.²⁶ Such a shift from least obstructionist to most obstructionist for majority party Senators signifies a significant break from their party’s leadership. However, we do not find support that minority party Senators gain electoral support if they defect from their party and move their obstruction preference from pro-obstruction to anti-obstruction.²⁷ The discrete marginal effect on Figure 6b articulates the same story. The average marginal effect of a one-unit change in the pro-obstruction direction on Senator vote-share for majority party Senators is significant and approximately 5%.²⁸ The average marginal effect of obstruction on Senator vote-share for minority party Senators is insignificant.

We respecify the model to test the potential that state partisanship, in addition to majority status, conditions the effect of obstruction preference on Senators’ electoral support. In the previous model specification, the state partisanship was a continuous measure centered at zero and coded in the incumbent party’s direction. To evaluate H4a and H4b, we must include a triple interaction into the model. We again factor out state partisanship into two binary dummy variables, one for out-partisan states and other for politically safe seats with the omitted category being marginal states. We then interact the two state partisanship dummy variables with our interaction of obstruction and majority states ( ${Obstruct}_{t - 1} \times {MinoritySen}_{t - 1}$ ), resulting in two triple interactions in the model. We keep the same control variables in this model specification.²⁹

Figure 7 shows the linear predictions of each quantity of interest in the triple interaction model. The first panel articulating the effect of obstruction, conditioned by majority status and representing an out-partisan state, on Senator vote-share tells an interesting story.

Majority party Senators representing out-partisan seats gain electoral support the more obstructionist their preference. For example, the predicted vote-share for a majority party Senator representing an out-partisan state and with the maximum pro-obstructionist preference (1 on the scale) is 0.72. By contrast, a majority party Senator representing an out-partisan seat, but with a maximum anti-obstruction score (–1 on the scale), receives a predicted vote-share of 0.53. The results are insignificant for minority Senators representing states that lean toward the majority. Turning our attention to the marginal states panel, it is clear there are no significant relationships of interest. Senators do not gain extra votes if they change their obstruction preference if they represent marginal states, independent of direction of moderation and majority status. This provides partial evidence for H4a, majority party Senators gain support if they become less polarized on their procedural preferences, but only if they represent states that lean toward the minority party as opposed to representing marginal states. We find no support for H4b, minority party Senators do not gain an electoral advantage if they adjust their obstruction preference closer to the majority party if they represent states that lean toward the majority party or marginal seats.

As expected, the same story holds when analyzing Senators who represent politically safe seats. This indicates that majority party Senators representing states that lean toward the minority party gain if they change their obstruction preferences while minority party Senators representing states that lean toward the majority party do not. Indeed, this null finding may originate from the possibility that obstruction defection from majority party Senators is more salient than defection from minority party Senators. However, it could be the case that minority-party Senators anticipate a cost to obstruction, given the stigma associated with obstruction and partisan gridlock among the mass public (Durr et al., 1997; Hibbing & Theiss-Morse, 1995).

Discussion: Partisan and Constituent Pressures

Our results show that procedural votes impact the election prospects of Senators and that those Senators are aware of the impact that procedural votes have on their reelection. We first find that procedural votes in the Senate make up for a higher percentage of overall roll call votes and that media coverage of procedural votes has increased over time. We believe that voters use the highly salient procedural votes in the Senate to assess how loyal a Senator is to her party or her constituency. Senators, especially those in vulnerable reelection positions, who are too loyal to their party face punishment for their partisan procedural votes, similar to how members of the House are punished for being too partisan (Carson et al., 2010; Lebo et al., 2007; Patty, 2008).

These findings bode well for the awareness of the American voter. They suggest that sitting minority party Senators are aware of the influence their procedural votes will have on reelection. Indeed, we find that the effect of Senator obstruction on electoral outcomes is conditioned by potential electoral vulnerability; Senators representing politically hostile states, from both parties, are the most likely to moderate their obstruction preferences ahead of a reelection bid. Since Senators representing marginal states and states that favor the other party cannot afford a loss in support, particularly among voters of the other party, it is not surprising that they are the most likely to defect on procedural votes. If Senators are concerned about how their procedural votes will be used against them in a reelection campaign, then citizens must possess the ability to understand the implications of procedural votes. Although Americans may not like to look at the process (Hibbing & Theiss-Morse, 2002), when made aware, our results suggest they are capable of punishing lawmakers for procedural votes. In addition, our analysis finds empirical support of the theory proposed by Ashworth and Bueno De Mesquita (2014). We find that some lawmakers do anticipate public reactions to their votes, and those who do not face punishment. Those findings compliment their theory that no study of civic competency can be complete without thinking about the interactive nature between the public and lawmakers.

We find evidence that Senators are punished by their constituents in the form of a reduced two-party vote-share. We find that majority members are more likely to be punished for supporting their party on procedural items. Combining this finding, with our earlier finding that minority members change their obstruction heading into elections suggests that there is a dynamic between party influence and electoral consequences. This is similar to the process found by other scholars with respect to the cross-pressures members of Congress face (Carson et al., 2010; Koger & Lebo, 2012; Lebo et al., 2007). Our findings suggest that majority members are less likely to break on procedural votes, but because of their loyalty, they are punished more during reelection. This story falls in line with thermostatic models and existing theories of party influence. The majority party in the Senate needs loyalty from members to accomplish policy goals. Accomplishing these goals leads to poor reelection prospects, especially for electorally vulnerable members. Given that the minority party is not tasked with creating new policy, it does not require the same loyalty on procedural votes and is thus more focused on electioneering to take back control of the chamber (Lee, 2016).

Based on these findings, more theories of organization that have originated from the House should be tested in the Senate. Our work does not uncover the entire process of organization in the Senate. We believe that many of the mechanisms thought to explain party control in the House, work differently in the Senate. The cost of compensating members to vote in favor of moving a status quo further from their ideal point is likely higher in the Senate than it is in the House. This dynamic makes policymaking more difficult, especially given cloture rules in the Senate. Studying a member’s procedural voting behavior as they graduate from a Representative to a Senator could explain some of the differences in agenda-setting power between the House and Senate. If members seem to operate with more autonomy than when they were in the House, it would support our argument that procedural votes are more salient in the Senate. This study contributes to our understanding of legislative behavior by providing evidence that Senators respond to electoral incentives to deviate from their party on procedural matters. As the proportion of procedural votes increases in the Senate, we expect the salience of these votes to grow as well, making research on the implications of procedural votes all the more necessary.

Supplemental Material

Obstruction_Cost_APR_Appendix – Supplemental material for The Member-Level Determinants and Consequences of Party Legislative Obstruction in the U.S. Senate

Supplemental material, Obstruction_Cost_APR_Appendix for The Member-Level Determinants and Consequences of Party Legislative Obstruction in the U.S. Senate by Carlos Algara and Joseph C. Zamadics in American Politics Research

Footnotes

Appendix

Acknowledgements

A previous version of this paper was presented at the 2017 Annual meeting of the Midwest Political Science association. We appreciate Gary Jacobson, E. Scott Adler, Chris Hare, Erik Engstrom, and the UC Davis American Politics Reading Group for their thoughtful feedback on previous iterations. We also wish to thank Geoffrey Skelley & the Center for Politics at the University of Virginia for sharing Senate election data. Lastly, we are grateful for the helpful suggestions provided by the reviewers and Editor.

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.

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Carlos Algara is a doctoral candidate in political science at the University of California, Davis. His research focuses on the nature of legislative representation and behavior, the electoral process, mass public opinion, and latent variable measurement methods. His website can be found at: .

Joseph C. Zamadics is currently a PhD candidate at the University of Colorado Boulder. His current research uses content analysis to measure issue salience at the local level to develop richer theories of lawmaker response to large events. Other areas of research include state legislatures, Congress, and legislative success.

References

Aldrich

J. H.

Rohde

D. W.

(2000). The consequences of party organization in the house: The role of the majority and minority parties in conditional party government. In Bond

J. R.

Fleisher

(Eds.), Polarized politics: Congress and the president in a partisan era (pp. 31-72). Washington, DC: CQ Press.

Althaus

S. L.

Edy

J. A.

Phalen

P. F.

(2001). Using substitutes for full-text news stories in content analysis: Which text is best? American Journal of Political Science, 45, 707-723.

Ansolabehere

Snyder

J. M.

Stewart

III . (2001). The effects of party and preferences on congressional roll-call voting. Legislative Studies Quarterly, 26, 533-572.

Armstrong

D. A.

III Bakker

Carroll

Hare

Poole

K. T.

(2014). Analyzing spatial models of choice and judgment with R. Boca Raton, FL: CRC Press.

Ashworth

Bueno

Mesquita

(2014). Is voter competence good for voters? Information, rationality, and democratic performance. American Political Science Review, 108, 565-587.

Binder

S. A.

(1997). Minority rights, majority rule. New York, NY: Cambridge University Press.

Binder

S. A.

(2011). Legislative productivity and gridlock. In Edwards

G. C.

Lee

F. E.

Shickler

(Eds.), The Oxford handbook of the American Congress (pp. 641-658). New York, NY: Oxford University Press.

Binder

S. A.

Lawrence

E. D.

Smith

S. S.

(2002). Tracking the filibuster, 1917 to 1996. American Politics Research, 30, 406-422.

Binder

S. A.

Smith

S. S.

(1997). Politics or principle? Filibustering in the United States Senate. Washington, DC: Brookings Institute Press.

10.

Boydstun

A. E.

(2013). Making the news: Politics, the media and agenda setting. Chicago, IL: The University of Chicago Press.

11.

Campbell

A. C.

Cox

G. W.

McCubbins

M. D.

(2002). Agenda power in the U.S. Senate, 1877 to 1986. In Brady

H. E.

McCubbins

M. D.

(Eds.), Parties, procedure and policy: Essays on the history of Congress (pp. 216-247). Palo Alto, CA: Stanford University Press.

12.

Carroll

Lewis

J. B.

Poole

K. T.

Rosenthal

(2013). The structure of utility in spatial models of voting. American Journal of Political Science, 57, 1008-1028.

13.

Carson

J. L.

Koger

Lebo

M. J.

Young

(2010). The electoral costs of party loyalty in Congress. American Journal of Political Science, 54, 598-616.

14.

Cox

G. W.

McCubbins

M. D.

(2005). Setting the agenda: Responsible party government in the U.S. House of Representatives. Cambridge, MA: Cambridge University Press.

15.

Cox

G. W.

McCubbins

M. D.

(2007). Legislative leviathan (2nd ed.). New York, NY: Cambridge University Press.

16.

Den Hartog

Monroe

N. W

. (2011). Agenda setting in the U.S. Senate: Costly consideration and majority party advantage. Cambridge, UK: Cambridge University Press.

17.

Doherty

(2015). How policy and procedure shape citizens’ evaluations of senators. Legislative Studies Quarterly, 40, 241-272.

18.

Durr

R. H.

Gilmour

J. B.

Wolbrecht

(1997). Explaining congressional approval. American Journal of Political Science, 41, 175-207.

19.

Flynn

D. J.

Harbridge

(2015). How partisan conflict in Congress affects public opinion: Strategies, outcomes, and issue differences. American Politics Research, 1, 1-28.

20.

Fong

Krehbiel

(2018). Limited obstruction. American Political Science Review, 112, 1-14.

21.

Gailmard

Jenkins

J. A.

(2008). Minority-party power in the Senate and House of Representatives. In Monroe

N. W.

Roberts

J. M.

Rohde

D. W.

(Eds.), Why not parties? Party effects in the United States Senate (pp. 181-197). Chicago, IL: The University of Chicago Press.

22.

Green

M. N.

(2015). Underdog politics: The minority party in the U.S. House of Representatives. New Haven, CT: Yale University Press.

23.

Grimmer

(2010). A Bayesian hierarchical topic model for political texts: Measuring expressed agendas in Senate press releases. Political Analysis, 18, 1-35.

24.

Groseclose

Milyo

(2010). Sincere versus sophisticated voting in Congress: Theory and evidence. The Journal of Politics, 72, 60-73.

25.

Harbridge

Malhotra

(2011). Electoral incentives and partisan conflict in Congress: Evidence from survey experiments. American Journal of Political Science, 55, 494-510.

26.

Hibbing

J. R.

Theiss-Morse

(1995). Congress as public enemy: Public attitudes toward American political institutions. New York, NY: Cambridge University Press.

27.

Hibbing

J. R.

Theiss-Morse

(2001). Process preferences and American politics: What the people want government to be. American Political Science Review, 95, 145-153.

28.

Hibbing

J. R.

Theiss-Morse

(2002). Stealth democracy: Americans’ beliefs about how government should work. Cambridge, MA: Cambridge University Press.

29.

Highton

(2000). Senate elections in the United States, 1920-94. British Journal of Political Science, 30, 483-506.

30.

Jessee

S. A.

Theriault

S. M.

(2014). The two faces of congressional roll-call voting. Party Politics, 20, 836-848.

31.

Jones

D. R.

(2010). Partisan polarization and congressional accountability in house elections. American Journal of Political Science, 54, 323-337.

32.

Jones

D. R.

(2015). Declining trust in Congress: Effects of polarization and consequences for democracy. The Forum, 13(3), 375-394.

33.

Keele

(2007). Social capital and the dynamics of trust in government. American Journal of Political Science, 51, 241-254.

34.

Koger

(2010). Filibustering: A political history of obstruction in the House and Senate. Chicago, IL: The University of Chicago Press.

35.

Koger

Lebo

M. J.

(2012). Strategic party government and the 2010 elections. American Politics Research, 40, 927-945.

36.

Lebo

M. J.

McGlynn

A. J.

Koger

(2007). Strategic party government: Party influence in Congress, 1789-2000. American Journal of Political Science, 51, 464-481.

37.

Lee

F. E.

(2009). Beyond ideology: Politics, principles, and partisanship in the U.S. Senate. Chicago, IL: The University of Chicago Press.

38.

Lee

F. E.

(2016). Insecure majorities: Congress and the perpetual campaign. Princeton, NJ: Princeton University Press.

39.

Matthews

D. R.

(1959). The folkways of the United States Senate: Conformity to group norms and legislative effectiveness. American Political Science Review, 53, 1064-1089.

40.

McCarty

Poole

K. T.

Rosenthal

(2006). Polarized America: The dance of ideology and unequal riches. Cambridge, MA: MIT Press.

41.

McKelvey

R. D.

Niemi

R. G.

(1978). A multistage game representation of sophisticated voting for binary procedures. Journal of Economic Theory, 18, 1-22.

42.

Park

H. M.

Smith

S. S.

(2016). Partisanship, sophistication, and public attitudes about majority rule and minority rights in Congress. Legislative Studies Quarterly, 41, 841-871.

43.

Patty

J. W.

(2008). Equilibrium party government. American Journal of Political Science, 52, 636-655.

44.

Poole

K. T.

(1985). A spatial model for legislative roll call analysis. American Journal of Political Science, 29, 357-384.

45.

Poole

K. T.

Lewis

J. B.

Carroll

(2007). Scaling roll call votes with wnominate in R. Journal of Statistical Software, 22(1), 1-23.

46.

Poole

K. T.

Rosenthal

(1997). Congress: A political-economic history of roll call voting. New York, NY: Oxford University Press.

47.

Ramirez

M. D.

(2009). The dynamics of partisan conflict on congressional approval. American Journal of Political Science, 53, 681-694.

48.

Ramirez

M. D.

(2012). The policy origins of congressional approval. The Journal of Politics, 75, 198-209.

49.

Roberts

J. M.

Smith

S. S.

(2003). Procedural contexts, party strategy, and conditional party voting in the U.S. House of Representatives, 1971-2000. American Journal of Political Science, 47, 305-317.

50.

Rohde

D. W.

(1991). Parties and leaders in the postreform house. Chicago, IL: The University of Chicago Press.

51.

Seo

Theriault

S. M.

(2012). Moderate caucuses in a polarised US Congress. Journal of Legislative Studies, 18, 203-221.

52.

Smith

S. S.

Ostrander

Pope

C. M.

(2013). Majority party power and procedural motions in the U.S. Senate. Legislative Studies Quarterly, 38, 205-236.

53.

Smith

S. S.

Park

H. M.

(2013). Americans’ attitudes about the Senate filibuster. American Politics Research, 41, 735-760.

54.

Stimson

J. A.

(1998). Public opinion in America: Moods, cycles, and swings (2nd ed.). Boulder, CO: Westview Press.

55.

Theriault

S. M.

(2008). Party polarization in Congress. New York, NY: Cambridge University Press.

56.

Volden

Bergman

(2006). How strong should our party be? Party member preferences over party cohesion. Legislative Studies Quarterly, 31, 71-104.

57.

Wawro

G. J.

Shickler

(2004). Where’s the pivot? Obstruction and lawmaking in the pre-cloture Senate. American Journal of Political Science, 48, 758-774.

58.

Woolley

J. T.

(2000). Using media-based data in studies of politics. American Journal of Political Science, 44, 156-173.

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

4.36 MB