Veto-Proof Majorities,Legislative Procedures,and Presidential Decisions,1981

Abstract

Super-majorities have occurred frequently in Congress but have escaped scholarly attention. This paper employs new measures of positive agenda control and a unique data set of 3,407 nontrivial bills from 1981 to 2008 to answer two questions: how did legislative leaders construct veto-proof coalitions, and what did presidents do with them? Legislative leaders, we argue, deployed procedures to expand and sustain veto-proof coalitions, despite increasing polarization. The resulting history, which signaled members’ commitment to a bill, provided information to the president that reduced uncertainty about possibilities for interbranch bargaining and the likely success of a veto. We find that positive agenda control increased the probability of vote tallies of two-thirds or more, especially after the 1994 election. In addition, we demonstrate that presidents concentrated veto activity on bills with outcomes of less than two-thirds, rejected some veto-proof bills for reputational gains, and deployed signing statements strategically. The analysis suggests that congressional leaders paradoxically gained capacity for nurturing large, bipartisan alliances as the institution became more polarized. Moreover, it demonstrates that strategic activity by legislative leaders is critical to explaining variation in presidential options for veto bargaining and signing statements.

Keywords

Congress legislative coalitions veto-proof majorities agenda control

In his classic study of divided government, David Mayhew (1991) noted that broad, bipartisan coalitions occurred frequently on major legislation in the U.S. Congress. Super-majorities, he observed, result from a “Capitol Hill mindset of problem-solving and a logic of aggregation” in the fragmented American system (Mayhew 1991, 134). Indeed, between 1981 and 2008, roughly 18 percent of bills with a recorded vote in at least one chamber passed Congress with veto-proof margins. The continuation of “a logic of aggregation,” despite increasing levels of polarization, raises questions about the internal workings of Congress and the institution’s relationship with the White House. How did legislative leaders create supersized coalitions during a period of rapid political change, and how did presidents respond to bills with such widespread backing? Procedural maneuvers provided the means for expanding support for a bill, we argue, and the resulting veto-proof tally enabled presidents to assess reputational costs and benefits of a veto. Paradoxically, as Congress became more polarized, its leaders exercised positive agenda controls to sustain bipartisan super-majorities and signal lawmakers’ commitment for a bill to the Oval Office.

In theory and practice, we might expect veto-proof bills to be rare events in Congress. Some theories stress the median lawmaker as decisive in legislative outcomes (cf. Krehbiel 1998), while others emphasize the majority party’s control of the agenda to satisfy its median member (cf. Aldrich and Rohde 2001; Cox and McCubbins 2005). Oversized coalitions not only are larger than necessary but also muddy the majority’s efforts to establish a favorable reputation with voters. In addition, empirical work on Congress emphasizes the ability of minorities to obstruct legislation (cf. Bach and Smith 1988; Koger 2010; Sinclair 2006) and the negative effects of polarization (cf. Binder 2003, 2014; Mann and Ornstein 2006; Theriault 2008, 2013). Although veto-proof majorities are not the norm in Congress, the fact that they occur at all in an institution prone to gridlock warrants investigation.

The president gains enormous leverage over Congress through the decision to sign or reject a bill (Cameron 2000; Groseclose and McCarty 2001; Martin 2012; McCarty and Poole 1995). A veto-proof tally, however, undermines the president’s influence by signaling lawmakers’ unwillingness to negotiate. It also poses a costly choice for the executive between a veto with a low probability of success and a signature on a bill the president dislikes. Given the prevailing view of presidential dominance in Washington (Howell 2013), the passage of a significant percentage of nontrivial bills that constrain White House options raises issues about bargaining between the two branches.

We examine the mechanisms leaders adopted to pass nontrivial bills with veto-proof tallies and the president’s decision to sign or veto them. The key to these behaviors was positive agenda control to expand and sustain a coalition (Finocchiaro and Rohde 2008; Riker 1986; Rohde et al. 2013). By broadening support and deterring defection, legislative leaders generated a legislative history that signaled members’ commitment to a bill. By evaluating both the maneuvers and the veto-proof outcome, presidents reduced uncertainty about lawmakers’ likely reaction to the choice among signing or veto options.

The Department of Defense appropriation bill, H.R. 2863, approved in late 2005, illustrates the impact of positive agenda control on a bill’s fate. The Senate combined it with a measure prohibiting torture in defiance of President Bush’s threatened veto, and a conference committee then added the Senate proposal to the House version, which had lacked the antitorture provisions. The bill passed both chambers with overwhelming majorities. Bush did not follow through with a veto in the face of congressional unity, although he appended a signing statement to the bill objecting to its encroachment on executive powers.

The study uses new measures to capture positive agenda control and a unique data set of 3,407 substantively important bills. The results demonstrate that procedural maneuvers increased the probability of a veto-proof tally, thus complementing research about the influence of party leaders in shaping roll-call outcomes (cf. Den Hartog and Monroe 2011; Harbridge 2015; Lee 2009; Sinclair 2012; Smith 2007). They confirm an old but still relevant bit of Capitol Hill wisdom that “the tools are there,” if members are willing to use them (Sinclair 1983, 240).

The analysis treats presidential decisions as a choice among four options (sign only, rhetorical signing statement [RSS], constitutional signing statement [CSS], and veto) and reveals the impact of a bill’s legislative history on executive decisions. Presidents concentrated veto activity on bills with vote outcomes of less than two-thirds, rejected some veto-proof bills for reputational gains, and used signing statements strategically. The study thus clarifies conditions when veto bargaining and blame game politics prevailed in the Oval Office (Cameron 2000; Gilmour 1995, 2011; Groseclose and McCarty 2001) and when the president engaged in more nuanced forms of communication (cf. Francis and Sulkin 2013; Ostrander and Sievert 2012, 2013). Coalition builders adopted methods of positive agenda control to create veto-proof tallies, thus influencing the executive’s veto power, even if they could not fully constrain it.

Super-Majorities, Procedures, and Presidential Options: A Closer Look

Scholars have overlooked veto-proof bills in Congress and their consequences for the White House. Congressional analysts have focused on negative agenda control to attain majority party goals rather than positive inducements to create bipartisan alliances. Presidential researchers have produced separate analyses of presidential choice, either as binary decisions to sign or veto a bill, or as simple signatures versus signing statements, but they have not evaluated executive choice among all four options. Moreover, researchers have not looked specifically at congressional procedures and veto-proof tallies as informative signals to the White House.

Political scientists have always regarded legislative procedures as important to the passage of bills, but began paying closer attention to rules as the minority party turned to obstruction (Theriault 2008) and the majority adopted negative agenda controls that restricted floor access and deliberation (Aldrich and Rohde 2001; Bach and Smith 1988; Cox and McCubbins 2005; Den Hartog and Monroe 2011; Dougherty et al. 2014; Gailmard and Jenkins 2007; Harbridge 2015; Sinclair 1983, 2006, 2012; Smith 2007). Procedural votes consequently became more partisan than substantive roll calls (Lee 2009; Oleszek 2011), while the fate of bills became highly correlated with “unorthodox measures” (Sinclair 2012).

The narratives of majority rule, however, miss the noticeable proportion of substantively important bills Congress passed with super-majorities. Even in a partisan era, members can benefit from oversized coalitions: to satisfy constituency interests, to avoid blame for inaction on important issues, or to seek bipartisan political cover. Positive agenda controls, which expand and sustain coalitions (Finocchiaro and Rohde 2008; Riker 1986; Rohde et al. 2013), can attract minority members of Congress (MCs) and deter majority lawmakers from defecting. If successful, these tactics send a strong signal to the president about congressional commitment to a bill and counter the president’s veto leverage over legislation, particularly during divided control. The origins of veto-proof bills thus speak to important debates about congressional parties and governance.

Two separate literatures address the president’s disposition of legislation but have limited applicability to the impact of procedural maneuvers and veto-proof tallies on White House decisions. Beginning with studies of veto behavior, analysts treated the president’s actions as a simple binary choice between accepting or rejecting a bill, but excluded various signing options (Cameron 2000; Cameron and McCarty 2004; Martin 2012; McCarty and Poole 1995). Yielding diverse expectations, which Cameron (2012) characterized as contradictory, theorists predicted vetoes would not occur (Krehbiel 1998), or would result from congressional uncertainty (Cameron 2000) or error (Martin 2012) about the president’s ideal point, or would reflect “blame game” tactics to maneuver the president into revealing unpopular positions (Groseclose and McCarty 2001). None of these accounts, however, considers cases in which presidents refrain from deploying their most powerful legislative tool in favor of signing statements.

Second, empirical studies of vetoes indicate a variety of influences on their use but have not looked at the impact of veto-proof tallies. Researchers have established that vetoes occurred more frequently when divided government prevailed (Cameron 2000), when the president’s party coalition in Congress diminished (Rohde and Simon 1984), or when party reputations, blame games, and bicameral differences were involved (Conley and Kreppel 2001; Gilmour 1995, 2002, 2011). None of these analyses dealt explicitly with the size of a bill’s majority as a factor in presidential decisions. Most relied on data predating the evolution of a party-centered Congress (Rohde et al. 2013).

Turning to presidential signing statements, scholars have debated whether or not they represent an expansion of executive power but have not directly compared them with vetoes. Some stressed the influence of the executive branch in having the “last word” (Kelley and Marshall 2008); others viewed statements as a defensive reaction to congressional encroachment, particularly on foreign affairs (Evans 2011; Ostrander and Sievert 2012); and still others focused on presidential efforts to educate the public, build political capital, and communicate with Congress (Francis and Sulkin 2013; Kelley, Marshall, and Watts 2013; Korzi 2011; Ostrander and Sievert 2013). A fair test of the executive’s intent with respect to legislation, however, requires comparison with the president’s veto power.

Students of the presidency have mined a bill’s legislative history for insights into the president’s use of signing statements but have not examined the final vote tally and the procedures that produced it. Berry (2009) looked at signing statements as a presidential reaction to Congress’s use of legislative vetoes in the struggle for policy-making discretion over laws, while Evans (2011) explored how modern presidents used signing statements to protect executive prerogatives from congressional challenges in the realm of foreign policy. Similarly, Ainsworth et al. (2014) suggested that signing statements, by highlighting policy disagreements between the two branches, triggered greater levels of congressional oversight. Finally, Francis and Sulkin (2013) identified supportive coalitions in Congress, along with features of the political environment, as influences on the president’s choice of rhetorical and constitutional statements.

Other scholars have linked signing statements with veto behavior. Rice (2010) compared statements of administration policy (SAPs) and signing statements to argue that presidents relied on constitutional statements to influence discrete provisions in legislation when a veto was too costly. Kelley and Marshall (2009) demonstrated that veto threats were a key predictor of CSSs.

Clearly, both signing statements and vetoes figure in presidents’ approach to bills, but congressional mechanisms behind oversized coalitions and the president’s reaction to veto-proof bills remain unsettled. The legislative history of a bill, which includes procedural maneuvers and the size of the vote total, offers leverage on these questions.

Theoretical Expectations about Super-Majorities

The sequential nature of the legislative process gives lawmakers the advantage of determining a bill’s provisions and presidents the leverage of signing a measure or returning it with a veto. The shared responsibility of the two branches puts a premium on their ability to communicate their intentions, because getting it wrong can be costly for both branches. Through a bill’s legislative history, lawmakers convey their level of commitment to a bill, as well as the degree of difficulty in changing it. Presidents then use this information to determine the likelihood that a veto will improve a bill, lead to stalemate, or result in an embarrassing defeat. Veto-proof bills, along with the maneuvers that eased their passage, clarify the stakes and reduce the probability of an outcome that leaves one or both branches worse off.

A framework of information sharing between the two branches leads to testable propositions about influences on their behavior. We first consider the motivations of legislators and presidents with respect to veto-proof tallies and then examine expectations regarding the influence of procedural maneuvers, veto threats, divided government, and the 1994 election.

Veto-Proof Margins as Informative Signals

Super-majorities occur in Congress for many reasons. Lawmakers with shared constituency interests, for example, unite around pork barrel projects. Leaders come together on salient issues to promote party reputations for problem solving or avoid blame for obstruction. Members demand political cover on tough, unavoidable votes, or seek protection for a favored measure from a veto threat. With a final vote tally of more than two-thirds, lawmakers signal broad bipartisan support for a bill and an unwillingness to negotiate. Margins of less than two-thirds in one or both chambers, however, convey the possibility that a president can do better. These messages about the costs of a veto and its likely success reduce the likelihood of executive miscalculation.

Presidents bring multiple concerns to legislation on their desks. Individually accountable for outcomes in the minds of the public, they face pressure to deliver on campaign promises and avoid blame for legislative failures. Presidents also worry about appearing weak to Washington elites if they accept a bill that they previously opposed or fail to sustain a veto. Finally, presidential actions on a bill have consequences for future dealings with Congress. Francis and Sulkin (2013, 231) noted about signing statements: “Savvy presidents will anticipate . . . legislators’ reactions when deciding whether to praise or criticize a measure.” A similar logic applies to vetoes. Given the downside of misjudging congressional and elite reactions to a veto, presidents have strong motivations to pay attention to information contained in a bill’s final vote tally. The role of information sharing between the two branches is captured in the following hypotheses:

Hypothesis 1a: Congress conveys unwillingness to negotiate changes to a bill with a veto-proof tally but acknowledges the possibility of bargaining with a veto-chance margin.

Hypothesis 1b: Presidents are likely to accept bills and issue signing statements with veto-proof tallies but target bills with margins of less than two-thirds for vetoes.

Procedural Maneuvers

Oversized legislative coalitions are unstable, as individual members demand additional benefits or threaten to defect (Riker 1986; Shepsle and Weingast 1987). Legislative leaders address this volatility through agenda control. With negative controls, they keep divisive bills in committee, control the number and content of amendments, and determine terms of floor debate (Cox and McCubbins 2005). With positive controls, leaders offer inducements to MCs that make the bill more attractive than the status quo either by manipulating the issue space or altering the perceived meaning of a bill (Finocchiaro and Rohde 2008). Riker (1986) viewed such “heresthetic” maneuvers as vital to shore up support for measures vulnerable to defection and voting cycles. The presence of procedural tactics consequently signals a coalition’s potential instability and leaders’ commitment to protect it.

Procedural interventions offer useful intelligence to presidents in managing reputational risks associated with signing and vetoing bills. Leaders might be unable or unwilling to reassemble a new coalition if the president disrupts it with a veto, or members of the president’s party might escalate their demands to exploit interbranch conflict. The potential volatility of a bill’s coalition, therefore, raises the likelihood that a veto would produce stalemate, which is likely to be more costly for the president than for the legislature. The calculations of each branch lead to the following expectations:

Hypothesis 2a: Procedural maneuvers, by reinforcing a bill’s support, exert a positive effect on the probability that a bill will have a veto-proof margin.

Hypothesis 2b: Procedural maneuvers, by keeping a coalition together, decrease the probability of vetoes and increase the likelihood that the president will issue signing statements.

Veto Threats

As Congress formulates legislation, presidents sometimes intervene with veto threats. Lawmakers have several responses to the president’s challenge: modify the bill to meet his objections, call his bluff, trap him into a blame game, or neutralize the threat with a vote tally greater than two-thirds. Presidents also have multiple choices having issued a threat: sign the bill with a rhetorical statement praising lawmakers for their cooperation, sign it with a constitutional statement raising objections, or veto it outright. If the president does not get what he wants from Congress, he faces reputational pressure to stand his ground with a veto. In general, threats raise the reputational stakes for both branches, motivating the two sides to save face by digging in:

Hypothesis 3a: Lawmakers will increase their efforts to generate veto-proof bills after a threat.

Hypothesis 3b: Presidents will follow through on threats with a veto or issue criticism in the form of a CSS.

Divided Government

Divided government generates interbranch conflict, but its effects are debatable. Major legislation has passed under split control (Mayhew 1991), yet important agenda items have languished (Binder 2003, 2014). Whatever the overall pattern, presidents have been more likely to veto measures under conditions of divided government (Cameron 2000; Rohde and Simon 1984). Lawmakers can protect bills through veto-proof tallies, if enough members of both parties prefer the bill to nothing, but presidents may still opt for projecting toughness to Washington elites. These motivations lead to the following hypothesis:

Hypothesis 4: Divided government increases the probability of veto-proof votes in Congress and presidential vetoes.

Election of 1994

The election of 1994 ushered in the Gingrich Congress with its confrontational approach to the president and minority party (Theriault 2013). By then, the pace of polarization had accelerated (Theriault 2008). Once the parties became internally homogeneous and ideologically far apart, lawmakers ceded greater authority to their leaders to control the party agenda (Aldrich and Rohde 2001; Cox and McCubbins 2005). Indeed, Rohde et al. (2013) identified the 1994 election as the point of transition from a decentralized, postreform Congress to a party-centered institution. Although MCs undoubtedly had fewer opportunities for bipartisan cooperation, their leaders enjoyed greater capacity to make binding agreements. Greater cohesion within the parties since the 1994 election, however, created conflicting possibilities for presidents. On one hand, they could expect party leaders to keep a coalition together; on the other hand, they could anticipate fewer defectors from the minority to sustain a veto. These possibilities lead to the final hypothesis:

Hypothesis 5: The election of 1994 increased the probability of veto-proof bills, but its effect on the president’s choice among options is unclear.

In summary, both branches had reason to view veto-proof bills, and the maneuvers that produced them, as communication devices regarding legislative intent and reputational consequences from a presidential veto. Veto threats, divided government, and the watershed election of 1994 also were likely influences on congressional and presidential decisions.

Data, Variables, and Trends

Several characteristics of the data, measures, and trends warrant discussion.

Period of Study

We begin with the Reagan administration, which was the first to articulate a theory of signing statements in which the executive’s views were relevant to a bill’s legislative history (Kelley and Marshall 2008), and end with the administration of George W. Bush. Bush’s successor, Barack Obama, issued so few vetoes and signing statements that little variation occurred to examine statistically.

Although our findings may not generalize to subsequent administrations, they are interesting in their own right. The years between 1981 and 2008 included a transition in the organizational structure of Congress from a postreform pattern of relative decentralization, to a party-centered institution after the 1994 Gingrich Revolution (cf. Rohde et al. 2013). The period thus represents a fascinating paradox as Congress became more partisan and more capable of sustaining bipartisan super-majorities.

Veto-Proof Bills

The great majority of legislation passes under unanimous consent or voice vote, but lawmakers decide whether to record the final vote (Madonna 2011). The presence of a roll call thus conveys useful cues about a bill’s salience to one or both chambers. We define veto-proof outcomes for bills if both chambers supported the bill by a recorded vote or division of at least 66 percent, or one chamber passed it with a vote of two-thirds or more and the other chamber employed unanimous consent or voice vote. Veto-chance outcomes, in contrast, occurred if one or both chambers passed the bill with a roll-call vote or division of less than 66 percent. In the study’s twenty-eight-year period, the House recorded 1,640 votes, the Senate recorded 540 votes, and both chambers had roll calls on 418 bills.¹ Overall, the number of veto-proof bills was 1,059, and the number of veto-chance bills was 213.

Figure 1 depicts the proportion of bills with different roll-call outcomes for the final vote. Veto-proof bills grew rapidly after the 1994 election, peaking in 2001 with a ratio of three out of ten bills, and remained relatively high in the Bush (43) years compared with previous Republican administrations. In contrast, the proportion of veto-chance bills, those with a recorded margin of less than 66 percent in one or both chambers, remained relatively stable over the period of study, hovering close to one in a hundred in multiple years, but with a high of one in ten in 1995.

Figure 1.

Proportion of bills with veto-proof or veto-chance majorities.

Measures of Positive Agenda Control

Party leaders deploy a variety of means to exercise positive agenda control. We selected three with potential to expand a coalition and deter defections: combining multiple bills into a single package, filing cloture motions, and sending bills to conference. Although this list does not exhaust all possibilities, the three maneuvers had several advantages for empirical analysis. First, they differed in function and stage in the legislative process. Second, they were readily available to coalition builders without formal rule changes. Third, they were part of a bill’s public record and, consequently, visible to the president.

Combined bills

Packaging separate bills into a single proposal has been commonplace in Congress. The tactic has been used with pork barrel legislation to create “omnibus” measures dealing with agriculture, transportation, or public works, and it also has involved major fiscal and budget reconciliation measures (Krutz 2001; Oleszek 2011; Sinclair 2012). Especially at the end of a session, combined bills have cleared the calendar and dealt with time-sensitive appropriation bills. The relationship between legislative packages and veto-proof tallies should be positive as additional provisions generated more stakeholders and provided political cover for members who could point out sections favorable to their constituents. For the president, bill packages raise the costs of a veto and make signing statements more likely.

Cloture

Scholars have demonstrated a link between cloture and the size of legislative coalitions prior to the 1950s, although they have disagreed about reasons for the connection (Carson, Lynch, and Madonna 2011; Madonna 2011; Wawro and Schickler 2004). Cloture votes seek to end minority obstruction for a bill (Koger 2010; Sinclair 2012), but filing cloture motions also initiates bargaining with respect to the status quo. Senators often do not realize why a favored measure has not come to the floor because filibuster threats are not necessarily public. Filing a cloture motion calls attention to a bill’s plight; in effect, Senate leaders announce a readiness to negotiate with the filibuster pivot. Cloture operates only in the Senate, so its impact on veto-proof margins and presidential choices is likely to be modest.

Conference

Conference committees typically favor members of the authorizing committee by giving them a final opportunity to modify a bill’s provisions (Shepsle and Weingast 1987). Conference bills benefit from special rules that protect them from floor amendments, so conferees are able to pull legislation away from the median lawmaker (Vander Wielen 2010). Increasingly, party leaders have used their discretion to appoint conferees to favor the majority party (Lazarus and Monroe 2007; Sinclair 2012). Moreover, the frequency of conferences declined in favor of amendment trading or adoption of one chamber’s bill in the other chamber (Ryan 2011). Moving a bill to conference thus signified (1) the majority party intended to protect the conferees’ handiwork on the floor and (2) the minority assumed it could not do better through amendment trading and substitution. The “take it or leave it” nature of conference bills not only cast the choice between a given proposal and the status quo into stark relief but also signaled MCs’ commitment in both parties to protect a bill. Together, these calculations suggest that conference bills would increase the likelihood of veto-proof coalitions and motivate the president to issue signing statements, unless a prior veto threat raised reputational stakes.

Figure 2 depicts proportions of bills for which leaders deployed mechanisms of positive agenda control, including packages of more than one bill, cloture motions, or conference committees. The three maneuvers were not mutually exclusive; lawmakers frequently deployed more than one tactic on a bill during the period of study for a total of 2,861 maneuvers. Overall, 1,541 bills had one procedure, 561 had two procedures, and sixty-six had three procedures.

Figure 2.

Proportion of bills with procedural maneuvers, 1981–2008.

The most common maneuver was combining bills into a single measure, with the lowest proportion at .43 and a high of .80 for all bills. The ratio for combined bills tended to spike in the second session of a Congress as lawmakers endeavored to clear the calendar, with the exception of 1995. Motions filed to cut off debate rose from 1981 to 2008, with higher ratios in 1995 and 2007. The proportion of formal House–Senate conferences climbed slightly in 1995 but declined during the George W. Bush Administration. Bills lacking any of the procedures yielded ratios between .15 and .20 throughout the period, with exceptions in 1998 and 2007.

Presidential Choice

Figure 3 presents the proportion of bills for four different types of presidential decisions: sign the bill without comment (Sign), append an RSS or CSS, or issue a veto. The trends for the three decades indicate presidents signed most bills without comment or veto but differed in their use of options.² Reagan and Bush were more likely to use the veto than their successors, but the proportions mask the extent of difference. Under divided government, for example, Reagan vetoed seventy-four bills and Bush rejected forty-six, while Clinton returned seventeen bills in the 104th Congress and twenty-one bills during his second term. George W. Bush did not veto a single bill during the 107th Congress, despite his party’s loss of Senate control in 2001. He vetoed eleven bills in 2007 to 2008 and one veto during his unified government in 2006. With respect to signing statements, Reagan and Bush (41) followed similar patterns, while Clinton favored rhetorical statements (RSSs), and Bush (43) deployed constitutional statements (CSSs) until the last two years of his term.

Figure 3.

Proportion of presidential responses to bills (%), 1981–2008.

Bivariate relationships show significant relationships among procedures, roll-call outcomes, and presidential decisions among the 7,445 bills submitted to the White House. Veto-proof bills were more likely after the sweeping 1994 GOP victory, which marked a shift to party-based government in Congress (Rohde et al. 2013; Theriault 2013). Strong associations arose between veto-proof tallies and procedures, especially for conference. Finally, presidential decisions were related to the type of vote. Indeed, three-quarters of all vetoes were for bills with a recorded vote in at least one chamber of less than two-thirds.

Nontrivial Bills

Many bills entail such low stakes that neither branch engages in strategic behavior, but such trivial measures can swamp statistical analyses. Scholars have used various coding schemes to eliminate minor bills, but none is completely satisfactory. Most studies used legislative content to identify salient legislation through Congressional Quarterly’s Key Votes, media coverage, historical significance, or policy domains. Each approach, however, produced major inconsistencies in how bills were classified (Lapinski 2013).³ In addition, not all disputes between Congress and the president centered on major policy proposals. Gilmour (2011), for example, found evidence of presidential vetoes on minor bills. Moreover, some researchers excluded bills as unimportant if they passed under unanimous consent or with 90 percent or more of the vote. Yet, the Senate relied heavily on unanimous consent to move the majority party agenda (Den Hartog and Monroe 2011), and even a lopsided roll call conveyed useful political information about lawmakers’ commitment to a bill.

Our approach differs from previous studies by combining various indicators of a bill’s policy content and its page length. We classified bills as potentially consequential for interbranch relations if they entailed Key Votes in Congressional Quarterly; omnibus, appropriations, or foreign or defense policy bills; or veto threats. We added bills not covered by these categories that exceeded the median length of two pages for a total of 3,407 bills.⁴ The mean number of nontrivial bills per year was 144, with a low of fifty-four in 1995 and a high of 253 in 1988.

The bills classified as nontrivial ranged from specialized topics, such as legal liability for computer manufacturers under the 1999 Y2K Act, or omnibus budget measures, such as PL-66 in August 1993 or PL-77 in November 2001. Others broke new ground in policy terms, such as the Oceans Act of 1992, but were not significant enough to make headlines or appear in accounts of major legislation. Some involved contentious foreign policy issues, such as granting presidential authority to negotiate trade deals under the Uruguay Round (PL-49 in 1993).

Statistical Models and Results

Statistical results confirm expectations about informative signaling between Congress and the president between 1981 and 2008. Legislative leaders engaged in positive agenda control to create veto-proof tallies, and presidents used a bill’s procedural history and final vote to manage the reputational consequences from signing statements and vetoes. Results for both institutions prove robust under a variety of tests, including controls for a bill’s page length and policy content.⁵

Procedural Effects on Veto-Proof Outcomes

A logistic regression model with robust standard errors clustered on Congress estimates the effect of procedural maneuvers on the probability of a veto-proof margin. The dependent variable takes the value one if the bill had a recorded vote of two-thirds in both chambers or a recorded vote in one chamber with unanimous consent or voice vote in the other, and zero otherwise. The independent variables are the three procedural maneuvers (multiple bills, filing cloture motions, and conference committees) and dummy variables indicating the presence of a veto threat, divided government, and post-1994 election years. A control variable for the second session of Congress captures pressures on lawmakers to clear the legislative agenda before adjournment.

The results for individual procedures appear in Table 1. In Model 1, the three procedural variables are significant with the expected positive signs and large coefficients for bills with cloture motions or conferences.⁶ In Model 2, which introduces variables for veto threat, divided government, and heightened party control after 1994, the coefficients indicate the presence of systematic inclination in Congress toward veto-proof bills following the Gingrich Revolution. The coefficients and signs for procedural variables remain similar to those in Model 1, with a slight reduction in the effect of the variable, multiple bills. Divided government and veto threats did not exert a positive influence on the probability of super-majorities, as expected, although the latter is almost significant (p =.072).⁷

Table 1.

Procedural and Policy Influences on Veto-Proof Votes.

Variables	Model 1	Model 2	Model 3
Multiple Bills	0.230*** (0.071)	0.189** (0.073)	0.127 (0.083)
Cloture	0.990*** (0.241)	0.783*** (0.211)	0.650*** (0.223)
Conference	1.119*** (0.127)	1.167*** (0.137)	0.869*** (0.107)
Major Legislation			0.718*** (0.161)
Appropriations			0.618*** (0.101)
Foreign Policy			0.283** (0.112)
Omnibus			0.017(0.289)
Veto Threat		0.375(0.213)	0.224(0.155)
Divided		0.025(0.103)	0.019(0.099)
Post-1994		0.578*** (0.117)	0.593*** (0.083)
Second Session	−0.125(0.122)	−0.120(0.121)	−0.029(0.086)
Constant	−1.094*** (0.164)	−1.467*** (0.083)	−1.649*** (0.129)

N	3,407	3,407	3,407
Pseudo-R²	.059	.073	.089

Dependent variable is 1 if bill passed with veto-proof vote, 0 otherwise. Robust standard errors clustered on Congress are given in parentheses.

p < .05. **p < .01. ***p < .001.

The substantive influence of key variables on the probability of veto-proof tallies appears in Figure 4. The panel presents the difference in predicted probability of a super-majority with 95 percent confidence intervals as the value of a variable changes from zero to one, holding other variables at their means. The shift to a combined bill increases the likelihood of a veto-proof margin by 3.9 percent, filing a cloture motion raises the predicted probability by 18.3 percent, and sending the measure to conference augments the probability by 26.7 percent. In addition, bills adopted after 1994 increase the probability of a two-thirds tally by 12.2 percent. All the differences in predicted probability are statistically significant except veto threat. We include veto-threat values because the coefficient is almost significant, but the substantive difference of 8.3 percent is not significant.

Figure 4.

Change in predicted probability of veto-proof coalition by procedural maneuver: Difference in predicted probabilities as values shift from zero to one.

Several robustness checks confirm legislative leaders’ capacity for positive agenda control to produce an oversized vote tally. A potential problem is that an underlying factor not accounted for in the model drives the results; political conditions perhaps stimulated MCs to join bipartisan coalitions, for example. A variety of tests for fixed effects, random effects, and individual Congresses indicate, however, that unobserved, time-related variables are not a problem in the analysis. A second possibility is that our measures of positive agenda control are an artifact of other, unspecified tactics by party leaders. If this alternative explanation is correct, the number of procedures per bill should not matter for the likelihood of veto-proof votes. Yet, statistical results for the number of procedures, which are included in the online appendix as Table 1A, suggest the opposite conclusion. With one procedure, the difference in predicted probability of an outcome of two-thirds or more rises by 6.5 percent, with two procedures by 33.8 percent, and with three procedures by 36.7 percent. Thus, legislative leaders appear to tailor procedural measures to a particular bill.

A third consideration is that a bill’s policy content dictates leadership decisions to engage in procedural maneuvers to generate a super-majority. To test this proposition, Model 3 of Table 1 includes four dummy variables: major legislation, appropriations, foreign and defense policy, and omnibus measures. Clearly, policy substance exerted a positive influence on the likelihood that a bill passed with a recorded two-thirds vote in one or both chambers, with the exception of omnibus bills. In addition, tests comparing the fit of Model 2 and Model 3 favor the latter. A bill is 16.5 percent more likely to garner a veto-proof vote if it involves major legislation, while those involving appropriations or foreign and defense policy are, respectively, 13.7 and 6.1 percent more likely to yield a super-majority. The predicted differences are statistically significant.

Nevertheless, the coefficients of two of the three procedural variables, filing cloture motions and conference, remain highly significant, while the third, multiple bills, is near significance (p =.137). Substantively, the predicted difference between a bill for which cloture was filed and one for which it was not, is 14.9 percent, while the predicted difference in probability for a bill sent to conference remains high at 19.6 percent, and these effects are statistically significant. Thus, positive agenda control through procedural maneuvers has a clear effect on the outcomes of measures that garner high levels of bipartisan support in Congress.⁸

Presidential Signing Options

Multinomial logistic regression predicts the president’s choice of options with robust standard errors and fixed effects for each president’s first and second administrations. The dependent variable takes the values: zero for Sign, one for RSS, two for CSS, and three for veto.⁹ The key independent variables include veto-proof and veto-chance dummy variables for the final vote tally, a dummy variable for whether or not the president issued a veto threat, the three procedural dummy variables, and dummy variables for divided government and years post-1994. A control for the president’s year in office varies from one to eight. Separate analyses of a model that excludes the procedural variables and a model that incorporates a bill’s policy content appear in the online appendix as Tables 2A and 3A, respectively.

Table 2 presents statistical results with columns for rhetorical statements (RSSs), constitutional statements (CSSs), and vetoes compared with the omitted baseline of Sign. Overall, the results indicate the value of examining all the president’s options together. Each choice has different patterns of significant coefficients, and several independent variables effectively distinguish the three alternatives from each other. Bills with a coalition of less than two-thirds, for example, were more likely to generate rhetorical statements or vetoes but not constitutional statements, and the differences are statistically significant. In other words, presidents declined the opportunity Congress presented with a relatively mild statement or accepted the chance to bargain with a veto, while avoiding confrontational CSS language.

Table 2.

Influence of Congressional Tactics on President’s Choice.

Variables	RSS	CSS	Veto
Veto-Proof Bill	0.834*** (0.117)	0.565*** (0.137)	0.509** (0.223)
Veto-Chance Bill	0.570** (0.249)	0.361(0.254)	2.265*** (0.275)
Veto-Threat Bill	0.879*** (0.216)	1.331*** (0.204)	2.695*** (0.252)
Multiple Bills	0.199(0.111)	0.349** (0.130)	−0.644*** (0.190)
Cloture	0.600* (0.303)	0.911*** (0.259)	0.238(0.461)
Conference	0.768*** (0.136)	1.672*** (0.146)	0.571** (0.223)
President’s Year in Office	−0.208*** (0.053)	−0.003(0.055)	0.078(0.084)
Divided	1.030** (0.388)	−0.581** (0.222)	2.318** (1.045)
Post-1994	0.163(0.456)	1.147** (0.490)	13.634*** (1.080)
Constant	−2.383*** (0.428)	−2.724*** (0.354)	−5.043*** (1.069)
N	3,407
Pseudo-R²	.185

Dependent variable ranges from 0 to 3. Fixed effects for presidential administrations are not shown. Robust standard errors are shown in parenthese. RSS = rhetorical signing statement; CSS = constitutional signing statement.

p < .05. **p < .01. ***p < .001.

A comparable pattern is evident with divided government in which presidents were most likely to veto a bill or accept it with an RSS and least likely to issue a CSS. Again, both differences are statistically significant.¹⁰ These results are consistent with emerging consensus in the literature that signing statements facilitate interbranch communication. For the dummy variable, post-1994, the coefficients vary across various options but are problematic for vetoes.¹¹ Although tests of model fit favor exclusion of the variable, we include it on the grounds of theoretical importance.

A veto-proof tally is significant in predicting an increased probability for all presidential options. The coefficient for the veto has an incorrect sign, however, and the variable does not differentiate statistically between categories. Most of the forty-six vetoed bills with super-majorities involved reputational calculations. Fifteen were pocket vetoes, so the president did not have to worry about losing face from a successful override. All the cases occurred under divided government, half during Reagan’s presidency, when the president had to balance two unpalatable choices: a near-certain override or the anger of his fellow partisans who joined the coalition. For the remaining thirty-one nonpocket vetoes, presidents had made veto threats in twenty cases that raised reputational stakes from backing down. The president lost all but seven of these battles, but demonstrated resolve.

Looking more closely at veto threats, recall from Table 1 that the variable was not significant predicting super-majorities. Nevertheless, a threat not only increased the likelihood of a veto but also increased the likelihood of either an RSS or CSS.¹² The differences between categories are statistically significant, as presidential follow through on a threat outweighed more conciliatory approaches.¹³ During 1981–2008, then, presidents treated the final vote tally as a signal about the reputational costs and benefits of a veto.

Table 2 indicates that presidents also reacted to procedural maneuvers in a bill’s legislative history. Bill packages increase the chance of the CSS option expressing White House objections, but diminish the likelihood of a veto, producing statistically significant differences. Filing cloture motions raises the probability of constitutional statements, although the variable does not statistically differentiate the RSS, CSS, and veto options from each other. Bills that had been to conference augment the probability of all three choices and statistically distinguish signing statements from each other and the CSS option from the veto.¹⁴ Signing statements, thus, emerge as a nuanced form of communication with Congress, as in Francis and Sulkin’s (2013) study.

Figures 5 and 6 present substantive effects for key independent variables, showing the difference in predicted probabilities as the value changes from zero to one, holding all other variables at the mean. Large differences for rhetorical and constitutional statements are apparent in Figure 5, especially for a veto threat, which increases the probability of a CSS from 8.8 to 23.3 percent, for a change of 14.5 percent. All predictions are significant with the exception of the change in the veto-chance variable for the CSS option. The substantive changes for the procedural variables in Figure 6, however, are modest, with predicted differences of 2 to 3 percent, and the estimated differences are not always significant.

Figure 5.

Change in predicted probability of presidential choice by strategic variables: Difference in predicted probabilities as values shift from zero to one.

Figure 6.

Change in predicted probability of presidential choice by procedural maneuver: Difference in predicted probabilities as values shift from zero to one.

The substantive impact of strategic and procedural variables on vetoes is difficult to interpret because vetoes were so infrequent that the predicted differences were tiny.¹⁵ The main takeaway for vetoes is that differences are all positive and mostly significant. In Figure 5, for example, a veto-chance vote produces a change in predicted probability of a veto of 1.1 percent, which entails an eightfold increase, while a veto threat yields a change of 1.3 percent, which represents a tenfold increase. The impact of divided government is not only less, a predicted difference of only 0.24 percent, but is also significant. In contrast, a veto-proof margin generates a difference in the predicted probability of a veto by 0.10 percent, a small change that is not significant, but makes sense given the president’s reputational calculations noted above. Among the procedural variables, bill packages produce a negative, moderate, and statistically significant difference in the predicted probability of a veto. Neither cloture nor conference, however, generates a significant difference in a veto’s predicted probability.

A separate model, shown as Table 3A in the online appendix, examines whether policy content trumps a bill’s legislative history in predicting presidential choices among signing and veto options. Previous research demonstrated that a bill’s subject matter affected the likelihood of a signing statement compared with the sign only option. The results for all four options indicate that once a bill’s legislative history is taken into account, policy content variables exert only modest influence over the president’s choice and do not alter the impact of the vote outcome and procedural variables. Major legislation, for example, produces a positive effect on the RSS and CSS options, although substantive change for both is modest and not significant for the CSS prediction. The coefficient is not significant with respect to the veto, and the predicted change is negligible, a finding that contrasts with Cameron’s (2000) results. The omnibus measure has a positive impact on the probability of a CSS and is substantively interesting with a positive change of 10 percent. Although significant in predicting a veto, it does not produce a significant predicted difference in probability. A foreign or defense measure increases the probability of constitutional statements and yields a positive shift of about 10 percent.¹⁶ None of the other relationships is significant. Indeed, tests to compare model fit with and without policy content variables favor the simpler version.

These findings are relevant to presidential decisions in several respects. First, they indicate veto bargaining was not necessarily the result of congressional uncertainty or error about the president’s position. Legislative leaders attempted to forestall a veto by creating bipartisan coalitions when they could, and presidents exploited the leverage of the veto when the final vote tally signaled the possibility of success. Second, results for the veto-proof variable indicate the presence of blame game calculations on both sides. Lawmakers scored points by challenging the president to veto a bill with widespread, bipartisan support; presidents asserted their willingness to defend a position, even when it was unpopular on Capital Hill. Third, by examining vetoes and signing statements in the same model, we can see that presidents chose systematically among their four options to serve their political goals, especially with respect to veto-chance bills and veto threats.

Conclusion

The veto-proof bills that emerged in Congress between 1981 and 2008 exemplified Mayhew’s (1991, 134) “mindset of problem-solving and a logic of aggregation.” Super-majorities did not just happen, however: positive agenda control enabled legislative leaders to generate bipartisan coalitions, especially after the watershed election of 1994. Although bipartisan outcomes became more difficult to imagine as the House and Senate polarized, they were more likely to succeed as the postreform Congress transitioned to a party-centered institution that empowered leaders’ use of procedural tools. Strategic maneuvers in the legislature, in turn, provided information that aided presidents in evaluating reputational costs and benefits from signing or vetoing a bill. The president could assess both the effort invested in crafting a bipartisan bill and the likelihood that fellow partisans would sustain a veto. Consequently, the executive could focus the veto power on bills with a chance for successful bargaining or those that required demonstrations of resolve.

Our findings challenge conventional wisdom about the effects of divided government and party polarization. Bipartisan super-majorities occurred more frequently between 1981 and 2008 than most political observers acknowledged, and they were intentional, not accidental, as legislative leaders gained capacity to deploy existing rules. Nevertheless, the narrative of omnipotent presidents confronting hapless lawmakers has persisted, because the public and the press seldom pay attention to institutional processes.

The analysis holds important implications, as well, for the study of Congress. Scholars have long debated the meaning of roll-call votes. One camp asserts MCs express their ideological preferences, and the other sees party leaders frequently influencing lawmakers’ choices by structuring the legislative agenda. Unlike negative agenda control, which operates indirectly by keeping bills that divide the majority from the floor, positive agenda control manipulates bill content directly to attract supporters and deter defectors. The importance of procedural maneuvers in generating veto-proof outcomes confirms Aldrich and Rohde’s (2001) contention that polarized parties gain capacity to fulfill their objectives. Our results suggest that congressional coalitions not only reflect members’ ideological leanings but also depend on leadership activity. More broadly, our findings support Lee’s (2009) view that scholars need to look “beyond ideology” to understand legislative outcomes. Thus, a promising avenue for future research would move from how Congress creates veto-proof bills, which is the goal of this paper, to when polarized parties are motivated to cooperate on veto-proof bills.

Furthermore, the paper raises methodological issues relevant to future analyses of legislative behavior. The findings support claims by Rohde et al. (2013) and Theriault (2013) that Congress became a different institution after 1994. Furthermore, the study offers some new ways of studying legislative activity. It not only highlights the value of examining a bill’s legislative history for markers of positive agenda control but also offers a different approach to capturing bill salience.

For presidential scholars, the evidence shows that the executive enjoyed multiple options rather than a binary choice in deciding a bill’s fate. Presidents maximized leverage by focusing on bills with less than veto-proof majorities, although they were not always deterred by a veto-proof margin when a prior veto threat required a firm stand. These findings suggest that negotiation with Congress was not necessarily the result of uncertainty or mistake and that both branches engaged in blame game politics. Moreover, the analysis reveals a relatively nuanced approach on the part of presidents in using signing statements to communicate with Congress. Finally, the results indicate that divided government, while important, was just one determinant of legislative–executive interactions. Presidents paid attention to the size of their party’s coalition, of course, but also considered the legislative history of individual bills and leadership tactics to pass them.

By highlighting the president’s attention to the internal workings of Congress, the paper raises an interesting question about when presidents opt for unilateral action rather than engage in bargaining with Congress. For the period of study, the legislature and executive engaged in the type of give-and-take the framers envisioned as the two branches pursued strategies to maximize their policy and political benefits through legislation. President Obama, however, eschewed the powerful veto weapon for most of his tenure and minimized communication with Congress through signing statements. Heading into his final year in office, as we completed our analysis, he had deployed only two vetoes. By November 2016, he had issued another ten vetoes and a total of thirty-seven signing statements. His final tally of vetoes was similar to George W. Bush’s record but fell well below his predecessors.

After the Tea Party wave of 2010, which delivered majority control in the House to Republicans, the president had many fewer opportunities to engage lawmakers legislatively. Compared with previous administrations, Obama received fewer than half the number of bills Congress sent to Reagan or Bush (41) and 40 percent fewer bills than made it to the desks of Clinton or Bush (43). To implement its policy agenda, then, the White House relied more heavily upon unilateral tools, such as executive orders. The Administration also used less visible strategies, including memoranda from the Office of Legal Counsel (OLC) and agency directives or waivers, while exerting greater influence over agency rulemaking. Apparently, the White House pivoted away from a legislative presidency to a more muscular regulatory or administrative presidency—one designed to influence public policy while depending less upon bargaining or winning coalitions in Congress. In this respect, the inclination of presidents to operate on their own (Howell 2013) appears contingent on congressional behavior.

Republicans have been highly critical of Mr. Obama for an “imperial” approach to the presidency, but we suspect that the behavior of his administration reflects a careful, and sadly realistic, reading of the dysfunctional nature of the contemporary Congress. GOP lawmakers not only were unproductive legislatively but also showed a distinct preference for confrontation, brinksmanship, and repudiation of their party leadership. Former Speaker John Boehner (R-OH) and former Senate Majority Leader Harry Reid (D-NV) devoted considerable energy to deflecting the excesses of ultraconservative lawmakers, but had less leeway to craft bills that constrained the president’s options or led to bargaining between branches. Scholars may look back on the period of our study, therefore, as a milestone in the way lawmakers dealt with partisan polarization.

Footnotes

Authors’ Note

We acknowledge the assistance of Kahlie Dufresne, who adapted Christopher Kelly and Bryan Marshall’s (2008) original data set for her Honors Thesis at Dartmouth College and who coauthored a paper with us on a different version of this project. We also thank our departmental colleagues and anonymous reviewers for invaluable feedback.

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.

Notes

Supplemental Material

Appendix materials and replication data for this paper can be accessed online.

References

Ainsworth

Scott

Harward

Brian

Moffett

Ken

Rice

Laurie

. 2014. “Congressional Response to Statements of Administration Policy and Presidential Signing Statements.” Congress & the Presidency 41 (3): 312–34.

Aldrich

John J.

Rohde

David W.

2001. “The Logic of Conditional Party Government: Revisiting the Electoral Connection.” In Congress Reconsidered, edited by Dodd

Lawrence C.

Oppenheimer

Bruce I.

, 7th ed., 269–92. Washington, DC: CQ Press.

Bach

Stanley

Smith

Steven S.

1988. Managing Uncertainty in the House of Representatives: Adaptation and Innovation in Special Rules. Washington, DC: Brookings Institution Press.

Berry

Michael J.

2009. “Controversially Executing the Law: George W. Bush and the Constitutional Signing Statement.” Congress & the Presidency 36 (3): 244–71.

Binder

Sarah A.

2003. Stalemate: Causes and Consequences of Legislative Gridlock. Washington, DC: Brookings Institution Press.

Binder

Sarah A.

2014. “Polarized We Govern?” Center for Effective Public Management. Brookings Institution. https://www.brookings.edu/wp-content/uploads/2016/06/BrookingsCEPM_Polarized_figReplacedTextRevTableRev.pdf.

Cameron

Charles M.

2000. Veto Bargaining: Presidents and the Politics of Negative Power. New York: Cambridge University Press.

Cameron

Charles M.

2012. “Vetoes, Bargaining, and Boundary Conditions.” Political Analysis 20 (4): 520–24.

Cameron

Charles M.

McCarty

Nolan

. 2004. “Models of Vetoes and Veto Bargaining.” Annual Review of Political Science 7: 407–35.

10.

Carson

Jamie L.

Lynch

Michael S.

Madonna

Anthony J.

2011. “Coalition Formation in the House and Senate: Examining the Effect of Institutional Change on Major Legislation.” Journal of Politics 73 (4): 1225–38.

11.

Conley

R. S.

Kreppel

Amie

. 2001. “Toward a New Typology of Vetoes and Overrides.” Political Research Quarterly 54 (4): 831–852.

12.

Cox

Gary W.

McCubbins

Mathew D.

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

13.

Den Hartog

Chris

Monroe

Nathan W.

2011. Agenda Setting in the Senate: Costly Consideration and Majority Party Advantage. New York: Cambridge University Press.

14.

Dougherty

Keith L.

Lynch

Michael S.

Madonna

Anthony J.

2014. “Partisan Agenda Control and the Dimensionality of Congress.” American Politics Research 42 (4): 600–27.

15.

Evans

Kevin A.

2011. “Challenging Law: Presidential Signing Statements and the Maintenance of Executive Power.” Congress & the Presidency 38 (2): 217–34.

16.

Finocchiaro

Charles J.

Rohde

David W.

2008. “War for the Floor: Partisan Theory and Agenda Control in the U.S. House of Representatives.” Legislative Studies Quarterly 33 (1): 35–61.

17.

Francis

Katherine

Sulkin

Tracy

.2013. “Legislative Coalitions and Presidential Signing Statements.” Congress & the Presidency 40 (3): 230–54.

18.

Gailmard

Sean

Jenkins

Jeffrey A.

2007. “Negative Agenda Control in the Senate and House: Fingerprints of Majority Party Power.” Journal of Politics 69 (3): 689–700.

19.

Gilmour

John B.

1995. Strategic Disagreement. Stalemate in American Politics. Pittsburgh: University of Pittsburgh Press.

20.

Gilmour

John B.

2002. “Institutional and Individual Influences on the President’s Veto. Journal of Politics 64 (1): 198–218.

21.

Gilmour

John B.

2011. “Political Theater or Bargaining Failure: Why Presidents Veto.” Presidential Studies Quarterly 41 (3): 471–87.

22.

Groseclose

Tim

McCarty

Nolan

. 2001. “The Politics of Blame: Bargaining before an Audience.” American Journal of Political Science 45 (1): 100–19.

23.

Harbridge

Laurel

. 2015. Is Bipartisanship Dead? Policy Agreement and Agenda-Setting in the U.S. House of Representatives. New York: Cambridge University Press.

24.

Howell

William

. 2013. Thinking about the American Presidency. Princeton: Princeton University Press.

25.

Kelley

Christopher S.

Marshall

Bryan W.

2008. “The Last Word: Presidential Power and the Role of Signing Statements.” Presidential Studies Quarterly 38 (2): 248–67.

26.

Kelley

Christopher S.

Marshall

Bryan W.

2009. “Assessing Presidential Power: Veto Politics and Signing Statements as Coordinated Strategies.” American Politics Research 37 (3): 508–33.

27.

Kelley

Christopher S.

Marshall

Bryan W.

Watts

Deanna J.

2013. “Assessing the Rhetorical Side of Presidential Signing Statements.” Presidential Studies Quarterly 48 (2): 274–98.

28.

Koger

Gregory

. 2010. Filibustering: A Political History of Obstruction in the House and Senate. Chicago: The University of Chicago Press.

29.

Korzi

Michael J.

2011. “‘A Legitimate Function’: Reconsidering Presidential Signing Statements.” Congress & the Presidency 38 (2):195–216.

30.

Krehbiel

Keith

. 1998. Pivotal Politics: A Theory of U.S. Lawmaking. Chicago: The University of Chicago Press.

31.

Krutz

Glen S.

2001. “Tactical Maneuvering on Omnibus Bills in Congress.” American Journal of Political Science 45 (1): 210–23.

32.

Lapinski

John S.

2013. The Substance of Representation: Congress, American Political Development, and Lawmaking. Princeton: Princeton University Press.

33.

Lazarus

Jeffrey

Monroe

Nathan W.

2007. “Conference Committee Appointments in the 97th-106th Congress.”

Political Research Quarterly

60 (4): 593–606.

34.

Lee

Frances S.

2009. Beyond Ideology: Politics, Principles, and Partisanship in the U.S. Senate. Chicago: The University of Chicago Press.

35.

Long

J. Scott

Freese

Jeremy

. 2003. Regression Models for Categorical Outcomes. Rev ed. College Station, TX: STATA Press.

36.

Madonna

Anthony J.

2011. “Winning Coalition Formation in the U.S. Senate: The Effects of Legislative Decision Rules and Agenda Change.” American Journal of Political Science 55 (2): 276–88.

37.

Mann

Thomas E.

Ornstein

Norman J.

2006. The Broken Branch: How Congress Is Failing America and How to Get It Back on Track. New York: Oxford University Press.

38.

Martin

Gregory J.

2012. “Testing Theories of Congressional–Presidential Interaction with Veto Override Rates.” Political Analysis 20 (4): 501–19.

39.

Mayhew

David R.

1991. Divided We Govern. New Haven: Yale University Press.

40.

McCarty

Nolan M.

Poole

Keith T.

1995. “Veto Power and Legislation: An Empirical Analysis of Executive and Legislative Bargaining from 1961 to 1986.” Journal of Law, Economics, & Organization 11 (2): 282–312.

41.

Oleszek

Walter J.

2011. Congressional Procedures and the Policy Process. Washington, DC: Congressional Quarterly Press.

42.

Ornstein

Norman J.

Mann

Thomas E.

Malbin

Michael J.

2008. “Vital Statistics on Congress: 2007-2008.” Brookings Institution, Washington, DC.

43.

Ostrander

Ian

Sievert

Joel

. 2012. “The Logic of Presidential Signing Statements.” Political Research Quarterly 66 (1): 141–53.

44.

Ostrander

Ian

Sievert

Joel

. 2013. “What’s so Sinister about Presidential Signing Statements?” Presidential Studies Quarterly 43 (1): 58–80.

45.

Rice

Laurie L.

2010. “Statements of Power: Presidential Use of Statements of Administration Policy and Signing Statements in the Legislative Process.” Presidential Studies Quarterly 40 (4): 686–707.

46.

Riker

William H.

1986. The Art of Political Manipulation. New Haven: Yale University Press.

47.

Rohde

David W.

Simon

Dennis M.

1984. “Presidential Vetoes and Congressional Response: A Study of Institutional Conflict.” American Journal of Political Science 29 (3): 397–427.

48.

Rohde

David W.

Stiglitz

Edward H.

Weingast

Barry R.

Evans

Larry

Schickler

Eric

. 2013. “Dynamic Theory of Congressional Organization.” Unpublished manuscript. Accessed July 11, 2016. https://web.stanford.edu/group/mcnollgast/cgi-bin/wordpress/wp-content/uploads/2013/10/rsw_dynamics_1302171.pdf

49.

Ryan

Josh M.

2011. “The Disappearing Conference Committee: The Use of Procedures by Minority Coalitions to Prevent Conferencing.” Congress & the Presidency 38 (11): 101–25.

50.

Shepsle

Kenneth A.

Weingast

Barry R.

1987. “The Institutional Foundations of Committee Power.” American Political Science Review 81 (1): 85–104.

51.

Sinclair

Barbara

. 1983. Majority Leadership in the U.S. House. Baltimore: Johns Hopkins University Press.

52.

Sinclair

Barbara

. 2006. Party Wars: Polarization and the Politics of National Policy Making. Norman: University of Oklahoma Press.

53.

Sinclair

Barbara

. 2012. Unorthodox Lawmaking: New Legislative Processes in the U.S. Congress. 4th ed. Los Angeles: CQ Press.

54.

Smith

Steven S.

2007. Party Influence in Congress. New York: Cambridge University Press.

55.

Theriault

Sean M.

2008. Party Polarization in Congress. New York: Cambridge University Press.

56.

Theriault

Sean M.

2013. The Gingrich Senators: The Roots of Partisan Warfare in Congress. New York: Oxford University Press.

57.

Vander Wielen

Ryan J.

2010. “The Influence of Conference Committees on Policy Outcomes.” Legislative Studies Quarterly 35 (4): 487–518.

58.

Wawro

Gregory

Schickler

Eric

. 2004. “Where’s the Pivot? Obstruction and Law-Making in the Pre-cloture Senate.” American Journal of Political Science 48 (4): 758–74.

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.

3.30 MB

0.00 MB

0.26 MB

Veto-Proof Majorities,Legislative Procedures,and Presidential Decisions,1981–2008

Abstract

Keywords

Super-Majorities, Procedures, and Presidential Options: A Closer Look

Theoretical Expectations about Super-Majorities

Veto-Proof Margins as Informative Signals

Procedural Maneuvers

Veto Threats

Divided Government

Election of 1994

Data, Variables, and Trends

Period of Study

Veto-Proof Bills

Measures of Positive Agenda Control

Combined bills

Cloture

Conference

Presidential Choice

Nontrivial Bills

Statistical Models and Results

Procedural Effects on Veto-Proof Outcomes

Presidential Signing Options

Conclusion

Footnotes

Authors’ Note

Declaration of Conflicting Interests

Funding

Notes

Supplemental Material

References

Supplementary Material