Abstract
The extent to which the policymaking process is majoritarian, rather than controlled by the majority party, is a fundamental question in U.S. congressional politics, and in collective choice institutions broadly. Previous research has examined whether some House rules empower the minority party to propose alternative legislation to the floor and circumvent the majority party’s agenda power. I argue the motion to instruct conferees allows the minority to influence policy during the conference process. Motions to instruct are the prerogative of the minority party, are frequently offered, and are frequently passed by the chamber. They substantially moderate conference bills as compared to those bills without a motion to instruct, suggesting that the procedure weakens majority party agenda power. Further, the minority may offer a motion to instruct intended to fail as a way of demonstrating policy extremity to voters, consistent with blame-game bargaining models.
Whether seemingly majoritarian institutions produce non-majoritarian policy outcomes is a central question in legislative studies. In the U.S. House, the rules require only a simple majority of the chamber for passage, but they also seem to empower the majority party to prevent consideration of legislative alternatives that could defeat its preferred policy.
The extant literature examines procedures, such as the discharge petition and the motion to recommit, which may allow the minority party to force policy alternatives onto the agenda, thereby circumventing or weakening majority party agenda control and producing more centrist policies (Crombez et al., 2006). Similarly, I argue the motion to instruct conferees empowers the minority party to propose policy alternatives to the floor. Conferencing is frequently an important step for passage, especially for salient or complex bills, and evidence for majoritarianism in this process informs questions about the overall ability of the majority party to control the House agenda in an era of polarization (Hare and Poole, 2014) and empowered party leadership (Curry, 2015). I answer three questions: when are motions to instruct offered; when are they successful; what are the effects of successful and unsuccessful motions?
I show they are more likely to be offered and passed when the bill distributes more benefits to the majority party, and conference bills with a successful motion to instruct are more centrist as compared to other conference bills for which the motion failed. Further, consistent with blame-game bargaining theory, the minority is able to convey the majority party’s extreme policy preferences to voters by introducing motions to instruct destined to fail. In recent congresses, these motions are almost always offered, suggesting the minority party sees them as a way to effect legislative change or force embarrassing votes, providing an important indirect form of policy and electoral influence.
Majority party procedural advantages in the house
The House is majoritarian if for all pairwise comparisons of policies x and y a floor vote can be obtained for policy x if it is preferred to y by at least 218 members (Patty, 2007). Policies originating from non-majoritarian rules or procedures will be defeated by a floor vote or rendered moot by other majoritarian institutions. The discharge petition and the motion to recommit, for example, may undermine the gatekeeping power of the majority party. Because House rules seem to codify the rights of the majority party to control the agenda, party-based theories claim the House is not majoritarian.
If the majority party is able to exercise agenda control, it will produce policies that disproportionately favor its members, allowing them to “enjoy most of the spoils that accrue to the legislative process,” [emphasis mine] (Jenkins and Monroe, 2012: 899). Cartel theory argues that the majority party uses standing committees and the Rules Committee to block majoritarian (centrist) bills from reaching the floor, resulting in more partisan bills than would be the case in a true majoritarian chamber (Cox and McCubbins, 2005).
If the majority party is not able to make a take-it-or-leave-it offer to the chamber and alternative proposals must be considered, policy outcomes are those preferred by a chamber majority choosing between the majority party’s proposal, the alternative proposal, and the status quo, rather than between only the majority party’s proposal and the status quo. This will leave members of the majority party with fewer policy benefits because their proposals will sometimes be defeated and more centrist outcomes will be produced. Evidence that the minority party is able to force alternative proposals onto the agenda is weak (see Pearson and Schickler, 2009 on the discharge petition, and Roberts, 2005 on the motion to recommit.)
Motions to instruct conferees in the House
Conference committees remain a critical part of the interchamber resolution process and have met recently on large authorization bills (e.g., the 2020 Defense authorization), and on a number of appropriations bills (e.g., a consolidated appropriations bill in 2019, and additional agency appropriations bills). Conferences offer a “hard” test for majoritarian control of the agenda because the appointment of conferees is controlled by the chamber leadership, and amendments are not allowed post-conference, seemingly giving the majority party control over the process (Park et al., 2017).
“…[A]fter the House decides to go to conference but before the Speaker appoints the House conferees— a motion to instruct the conferees is in order” (Rybicki, 2018: 1). 1 Instructions can be very specific, covering multiple sections or provisions in a bill and direct conferees to accede to a Senate provision, insist on a House provision, or reach a compromise with the Senate that meets certain requirements. (See supplemental Appendix A for language from H.R. 2346 in the 111th Congress as one example of a motion to instruct.) “Under well-established House precedents, recognition to propose this motion to instruct is a prerogative of the minority party,” [emphasis original] Rybicki (2018: 2).
Motions to instruct are largely seen as opportunities for position-taking because the instructions conveyed by the House are not binding (Park et al., 2017). That is, a point of order cannot be sustained if the conferees fail to follow the instructions (Rybicki, 2018). Yet, this does not mean motions to instruct are meaningless. I argue they are a mechanism for the minority party to circumvent majority party agenda control. Enforcement occurs through the possibility of rejection or recommittal of the conference report. Further, the Senate conferees observe the motion, which may change the bargaining process. These enforcement mechanisms cannot be directly observed; rejection or recommittal is rare, but that may be because motions to instruct successfully modify bills.
Motions to instruct are offered on about 15% of conference bills (see supplemental Tables B1 and supplemental Table B2), and are successful about 64% of the time. On average, about 49% of the majority party and 91% of the minority party votes for the motion. These empirical patterns demonstrate motions to instruct are offered selectively by the minority party, attract significant support from the majority party, and are frequently successful.
A theory of motions to instruct
I use insights from the Romer and Rosenthal (1978) and blame-game bargaining models to develop a theory of when the minority party offers a motion to instruct, when motions are successful (or not), and the effects of both outcomes. I characterize the majority party and minority party as represented by their medians (M m and N m , respectively). The existing policy is the status quo (SQ) and the majority party has proposed a policy that is closer to the chamber median (C m ) than the status quo because the chamber median has voted to send the bill to conference (|C m − SQ′| < |C m − SQ|).
If the majority party’s proposal is not at the chamber median’s ideal point (SQ′ ≠ C m ), there is always an alternative policy the minority party can propose, through the motion to instruct, that will be preferred by the chamber median. The House conferees will then advocate for that policy in the final conference bill. 2 The result will be more centrist legislation (i.e., offer a larger policy payoff to the chamber median and members of the minority party) because the majority party previously exercised its gatekeeping power to pass an extreme bill, relative to the chamber median.
The minority party does not always offer a motion to instruct, however. I assume offering the motion has costs greater than zero due to the time and effort required to write the motion, whip legislators, and engage in other related floor activities. If offering the motion were costless, one would be offered on every conference bill.
The benefits the minority party receives from offering a motion to instruct are based on either the policy change achieved by a successful motion, or electoral benefits received from signaling an extreme bill to voters, characterized as the ideal point of the national median voter (V) to which both parties appeal. Holding costs constant, the minority is more likely to offer a motion to instruct when the potential policy gains from the motion are high, or when electoral benefits are high. These situations are more likely as the bill becomes more extreme relative to the chamber median and V.
If the minority party offers a motion to instruct, what conditions result in the passage of a motion? That is, the minority party can propose an alternative policy that induces acceptance because it is closer to the median than the majority party’s proposal (|C m − SQ N | < |C m − SQ′|), or it can propose a policy that will be rejected because the distance from its proposal to the median is greater than that of the majority party’s (|C m − SQ N | > |C m − SQ′|).
The minority party makes an offer that induces acceptance when the policy change achieved by its proposal is large. Because the minority party’s proposal is limited by what the chamber median must accept (it must be the case that |C m − SQ N | < |C m − SQ′|), this also implies that the chamber median is relatively distant from the majority party’s proposal, allowing the minority to significantly moderate legislation while still ensuring the median prefers its proposal.
The blame-game bargaining model, which explores bargaining in front of an audience, suggests that political actors induce rejection in some circumstances because they receive electoral benefits, d, from signaling out-of-step preferences to voters (Groseclose and McCarty, 2001). Electoral benefits increase as the distance between the voter’s ideal point and the minority party median decreases, and as the distance between the voter’s ideal point and the majority party’s proposal, SQ′, increases. The use of motions to instruct as a signal to voters is consistent with the existing literature which claims the motions are a way for the minority party to force members of the majority to take embarrassing votes (Crespin et al., 2011; Park et al., 2017).
The minority party offers a proposal which is farther from the chamber median than the majority party’s proposal |C m − SQ N | > |C m − SQ′|), and induces rejection, when it receives substantial electoral benefits by signaling an out-of-step majority party. This is more likely to occur as the distance between the majority party’s proposal and the voter increases. Again, because the majority party’s offer depends on the location of the chamber median, it must also be the case that the chamber median is far from the voter.
While the minority party median could offer a proposal slightly closer to the chamber median’s ideal point than the majority party proposal (assuming SQ′ ≠ C m ), the difference in utility the minority receives from this policy is smaller than the electoral benefits the minority party median receives from rejection (d > |C m − SQ N | − |C m − SQ′|).
Figure 1 captures these dynamics. In the top panel, the minority party median makes an offer that induces acceptance when the policy change achieved by SQ
N
, defined as being less than the distance from SQ′ to C
m
(θ), is substantial. SQ
N
is located at |θ − (2ɛ)| because it must be closer to the chamber median’s ideal point than the majority party’s offer. Because the new status quo is a function of the location of the chamber median, this implies that the chamber median is also relatively distant from the majority party median. Minority proposals of a motion to instruct showing passage (Top panel) and failure (Bottom Panel).
In the bottom panel, the minority party would propose SQ N at the point defined by |θ − (2ɛ)| to induce acceptance, but it receives greater utility from inducing rejecting because the policy benefits achieved by SQ N are small and the electoral benefits are large. SQ′ is far from the voter’s ideal point, because the chamber median is also far, and therefore the electoral benefits to the minority party outweigh the policy benefits should the chamber median accept the minority’s proposal.
To summarize, proposing and passing a motion to instruct require that the majority party bill is sufficiently far from the chamber median. The chamber median must not be so far from the voter that the minority party prefers to induce rejection to receive electoral benefits. If the chamber median is far from the majority party median, the minority party can propose a motion that results in a substantial policy change. If the chamber median is close to the majority party median and far from the voter, the minority party deems a possible policy change to be insignificant and instead offers a too moderate proposal that will fail in order to signal to voters that the majority party is passing policy far from the voter’s ideal point. Finally, because a successful motion to instruct moves the status quo toward the minority party, these bills should be more moderate (distribute more benefits to minority party members) as compared to those bills for which a motion to instruct was offered but failed.
Empirical implications and approach
The empirical models use conference bill as the unit of analysis and predict whether a motion to instruct is offered, whether it passes, and the effect of passage on bill extremity. It is not possible to measure the ideological location of various actors on individual bills, so I proxy for the extent a bill distributes benefits to the majority party using differences in majority and minority coalitions voting for the bill, and overall House coalition size voting in favor of the bill (Krehbiel, 1998). I use changes in coalition size pre- and post-conference to measure the extent to which the bill moderated as a result of a successful motion to instruct.
The theory first predicts that motions to instruct are most likely to be offered when the policy change from a successful motion is substantial, or when electoral benefits to the minority are high. The majority pursues partisan legislation, limited only by the distance from the chamber median to the status quo (θ in Figure 1). As bills move closer to the preferences of the majority-party median, the bill draws less support from minority party members and members of the majority on the opposite side of the chamber median from the majority party median. 3 That is, more majority-party oriented bills will have near minimal winning coalition sizes made up exclusively of majority party members.
The independent variable is the difference between the percentage of the majority party voting for the bill on initial passage and the percentage of the minority party voting for the bill. As this variable increases, the bill distributes more policy benefits to members of the majority party. I predict that as the difference between majority and minority party support increases, the probability a motion to instruct is offered also increases. As a robustness check, I also estimate models using overall percentage of the chamber voting in favor of the bill on initial passage, with an increase in coalition size expected to reduce the probability a motion to instruct is offered.
Motions to instruct are most likely to be successful when the difference between the chamber median and the party median is high. Again using difference in coalition sizes as an indicator for the extent to which a bill is majority-party oriented, I expect an increase in coalition differences to have a negative effect on the probability a motion to instruct passes, if one is offered. Similarly, an increase in the overall percentage of the chamber voting for the bill on initial passage will increase the probability a motion to instruct passes.
Groseclose and McCarty (2001) use a decline in presidential approval ratings after a veto to measure voter blame. I use a similar strategy by measuring changes in the percentages of Americans who approve of Congress. If motions to instruct signal to voters that the majority is passing an extreme bill, relative to their preferences, I expect congressional approval to decrease after a motion to instruct fails for an enacted conference bill as compared to the passage of conference bills when a motion to instruct is either not offered or is offered and passes. I use data from the Gallup Poll which has asked voters approximately monthly since the mid-1980s whether they approve of Congress.
Partisan conflict decreases approval of Congress as a whole (Ramirez, 2009), which will have negative effects on majority party approval (Jones, 2010; McDermott and Jones, 2005). As Algara (Forthcoming) notes, “more moderate policies may raise citizen evaluations of congressional job performance, which, in turn, could benefit the majority party at the ballot box.“ 4 Voters are not aware of motions to instruct per se, but they are aware of attempts by the majority party to pass partisan bills. 5
The last theoretical claim is that successful motions to instruct have an effect on the legislation produced by the conference committee. I use the difference in minority party support (and overall coalition size) from initial passage in the House to passage of the conference bill to measure the extent to which the bill moves toward the minority party’s (and chamber median’s) position. Coalition size on initial passage of the bill offers a measure of bill extremity than can be compared to the conference report vote to determine whether the bill became more centrist and less majority-party dominated. I predict that conference bills with a motion to instruct will increase coalition size change from initial passage to conference passage as compared to conference bills with a failed motion to instruct.
Motions to instruct data
The sample is all bills (H.R. or S.) which went to a conference committee between the 93rd and 114th Congresses and had a vote on the conference report, or were passed via unanimous consent. There are 1446 conference bills, and in some cases, more than one motion to instruct was offered on a bill, resulting in 328 total motions of which 210 passed (about 64% of those offered). Party voting data on motions to instruct is taken from the PIPC dataset (Crespin and Rohde, 2020), and the author collected voting data on passage of bills before and after conference (see supplemental Table B1). Narrow failure is common, with 106 of the 118 failed motion to instruct votes receiving support from greater than 40% of the chamber, indicating that motions to instruct seem designed to maximize policy or electoral benefits for the minority (see supplemental Figure B1). Motions to instruct are much more common in recent congresses (Supplemental Table B2), providing evidence for the claim that a minority party excluded from the legislative development process is more likely to seek agenda control or signal out-of-step preferences.
Other variables are taken from the Congressional Bills dataset (Adler and Wilkerson, 2008). The models predicting coalition change after a motion measure the difference in coalition sizes between initial passage and conference passage, effectively holding unobserved bill-level characteristics constant. Fixed effects for congressional term control for all relevant variation at the congressional level, including divided government and party size. In some models, due to the limited number of observations per Congress, fixed effects are collinear with other variables and so all results are shown both with and without the fixed effects. See supplemental Appendix B for more details on the variables included. In a separate set of analyses, I use matching to compare bills which are similar on observed characteristics to determine the effect of a motion to instruct on change in minority party and overall coalition size.
Predicting the introduction and passage of motions to instruct
The first models predict whether a vote occurred on a motion to instruct for a bill going to conference and separately, whether the motion passed (see supplemental Table C1). Consistent with expectations, an increase in majority–minority party difference in favor on initial passage increases the probability a motion to instruct is offered, but decreases the probability it passes. The left panel of Figure 2 shows that when the difference in party coalitions is at its minimum, the probability a motion to instruct is offered is 0.13 (95% CI: 0.11 to 0.16), but increases to 0.32 (95% CI: 0.24 to 0.40) at the maximum. The effect of coalition size difference on the probability a motion to instruct is offered and passed. Predicted probabilities from model 2 of Supplemental Table C1 (left panel) and model 4 of Supplemental Table C1 (right panel). Standard errors clustered by bill. All continuous or ordinal variables held at their mean, all dichotomous variables held at their mode.
In the right panel, an increase in coalition size difference decreases the probability a motion to instruct passes (models 3 and 4 in supplemental Table C1). The probability of passage is 0.69 (95% CI: 0.61 to 0.78) at its minimum and decreases to 0.50 (95% CI: 0.26 to 0.64) when difference is at its maximum. The results for overall support are consistent with these (see supplemental Table C2).
Evidence for electoral costs
Motions to instruct are expected to fail when the minority party uses them to signal out-of-step policy preferences with voters. I estimate the difference in congressional approval rating from Gallup polls taken before and after a bill was enacted, comparing bills with failed motions to other conference bills. The results in supplemental Table D1 limit the sample to enacted laws in which a poll was taken within 45, 60, and 90 days of enactment, and the prior poll was also taken within the same number of days as the subsequent poll. The models include other bill-level factors, election year, and the quarterly change in real personal income. As robustness checks, I conduct a “reverse causality” test to predict the change in congressional approval at specified days before the failure of a minority report, and a placebo test predicting the number of CQ Lines in a bill using a failed motion to instruct (see supplemental Appendix D for more details). Supplemental Figure D1 plots coefficients at different time intervals for both the regression results and the reverse causality tests, while the placebo test with the alternative dependent variable shows an insignificant relationship.
A failed motion to instruct produces a drop of about 1.51% (95% CI: −2.95% to −0.08%) in congressional approval within 45 days, while the results for other time intervals are very similar. This test shows a positive and statistically insignificant effect for all time intervals. This result provides empirical evidence for the common claim that motions to instruct serve as an opportunity for the minority party to embarrass the majority.
The effects of a motion to instruct on bill extremity
If the motions allow for minority party policy influence, the bill will move toward the center of the policy space and the ideal point of the median. I expect conference bills with motions to instruct to have greater support from the minority party and greater overall support on conference report passage as compared to bills for which the motion to instruct failed. The results are shown in supplemental Table E1 and demonstrate that controlling for other factors, including the amount of support the bill received on initial passage, there is an increase of about 5% in overall support and 10% in minority party support for bills with a motion to instruct. 6
Summary of estimates of the effect of a passed motion to instruct on conference bills.
In the non-pruned data, there are 55 conference bills without a passed motion to instruct and 141 with a passed motion to instruct. In the pruned data, there are 19 observations without and 33 observations with a passed motion for CEM with automated coarsening; there are 51 observations without and 103 observations with a passed motion for CEM with explicit coarsening.
The effect sizes for these results range from an increase in overall support of approximately 5%–10% and in minority party support of approximately 8%–12.6%. The substantive increase in minority party coalition size is about six to ten additional legislators voting for the bill after conference, an important effect in an era of polarization and narrow partisan majorities. Legislators who change their vote from oppose to support after a motion to instruct are only those minority party members to the right of the status quo in Figure 1 who are closer to the minority party’s proposal, as minority party members to the left of the existing status quo will oppose any policy movement in the direction of the majority party. These results provide evidence that motions to instruct have an important effect on the policy extremity of conference bills, despite their non-binding nature.
Conclusion
The results here demonstrate “soft” minority power during the conference process as the House minority has procedural tools to change the bill in a way more favorable to its own position, making it more majoritarian. Absent a rules change, these procedures exist independently of parties, and may mitigate at least some partisan effects. While failed motions indicate that the majority party can still receive its preferred bill if it is unified, there may be electoral costs for doing so.
The theory and results raise related questions. Given that the majority knows its proposal will either be rejected by the minority (and policy will collapse to the chamber median’s ideal point), or it will pay significant electoral costs for passing policy inconsistent with voter preferences, why does it seek to pass partisan legislation? One possible answer is that the majority is uncertain about what the chamber median will accept and pushes its offer to the limit, risking rejection but maximizing its policy benefits. It may also value these benefits more than electoral consequences, especially if the party median’s audience is not the national median voter.
A larger question concerns the institutional rules of collective choice institutions. In the House, why does the majority initially agree to adopt rules that limit its future ability to dictate legislative outcomes? If the majority party is truly in charge, why does it continue to allow rules and procedures— like the motion to instruct— which limit its ability to create party-centered legislation? These questions have long been noted in the literature, and have implications not just for the House, but for how all legislatures operate and the policy outcomes they produce.
Supplemental Material
sj-pdf-1-rap-10.1177_20531680221083798 – Supplemental material for Motions to instruct conferees as a majoritarian tool in the U.S. house
Supplemental material, sj-pdf-1-rap-10.1177_20531680221083798 for Motions to instruct conferees as a majoritarian tool in the U.S. house by Josh M Ryan in Research & Politics
Footnotes
Acknowledgements
I thank Carlos Algara and Broen Westberg for helpful suggestions and comments.
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.
Correction (April 2025):
Supplemental material
Notes
References
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.
