Note 43. Evans on Senate Countermajoritarianism
Representational and Procedural Sources of Partisan Bias
I’m taking the unusual step of devoting a Note to a recently published article. The article, reporting an important study by Professor Larry Evans of the College of William and Mary, addresses the dual questions that everyone recognizes as potentially important. How often do (a) the equal representation of states in the Senate and (b) the Senate’s supermajority requirement for cloture generate outcomes that represent a minority of Americans? He adds a third question: How often do equal representation and supermajority cloture reinforce or counteract each other. I encourage you to read his paper, which is available here.[1]
Background: Equal Suffrage of the States
The Framers of the Constitution made critical decisions about the Senate. Article I provides for “equal suffrage” of the states—each state, regardless of population, selects two senators. It also provides that the legislature of every state choose its senators, a provision replaced by the 17th Amendment, which provides for popular election of senators. Article V provides for ways to amend the Constitution but includes a critical last clause providing that “no State, without its Consent, shall be deprived of its equal Suffrage in the Senate.” This is interpreted to mean that equal representation of the states in the Senate cannot be changed without the approval of all states. That last clause was added in haste at the last minute to firm up support from small states, whose delegates to the convention expressed concern that large states would eventually jettison equal representation of the states in the Senate.[ii]
(Shamefully, the Constitution also provided that slaves were to be counted as three-fifths of a person when determining state populations for the purpose of seats in the House of Representatives. This granted southern states with large slave populations additional House seats but not as many as they would have if slaves, who had no civil rights at the time, had been counted as whole persons.)
From the start, the equal representation of the states in the Senate meant that, individually, states with small populations had an influence on Senate legislative outcomes as any large-population state. Collectively, Delaware, Rhode Island, Georgia, South Carolina had a smaller population than Virginia at the time of the constitutional convention. Concern about the disparities in Senate representation, judged by population, have intensified in recent decades. California now is about 69 times the population size of Wyoming and both have two senators. The disparity is going to increase for decades.
The current concern is intensified by virtue of the fact that small states are represented disproportionately by Republicans. Even when in the Senate majority, Republicans can win votes even though they represent state that have considerably less than half of the American population.
Background: The Filibuster and Supermajority Cloture
The Constitution gave to the House and Senate the power to determine their rules of procedure, although a few procedural provisions were included. Until 1917, the Senate had no way to limit debate. Just before and after the turn of the 20th century, minorities frequently filibustered to prevent debate from ending and holding a vote on the issue at hand. Provision for a cloture motion in Rule XXII was adopted in 1917 to allow a two-thirds majority of senators voting to end debate after a period of debate and consideration of amendments, a rule that was changed in 1975 to allow a three-fifths majority of all elected senators to impose cloture. That is, 60 votes are required (when all 100 or just 99 seats are filled) to force a vote on an issue when a minority refuses to allow a debate to end and a vote to occur.
In recent decades, the number of filibusters and the number of cloture votes has skyrocketed as minority parties have become more obstructionist. Many measures die in the Senate for lack of 60votes. As a result, filibuster reform—really, cloture reform—has again become a salient, controversial issue. Note 5 provides essential detail.
The Senate’s Rule XXII is not the only place where a supermajority is required to agree to motion. The Constitution provides for two-thirds majorities to ratify treaties or approve constitutional amendments, convict in an impeachment trial, override a presidential veto, and expel a member. The Senate also requires a two-thirds majority to suspend a rule and to waive budget limits or overturn a budget ruling by the presiding officer. The Senate can, and often has in recent decades, imposed by unanimous consent a 60-vote threshold for certain votes, such as amendments or final passage. Evans has performed the onerous task of coding the appropriate threshold for all Senate votes since 1789.
The important feature of current practice is that, at least in principle, a minority of senators, representing a minority of Americans, can prevent a Senate majority from getting a vote on a measure.
More Background: Measuring Outcomes
The ”outcomes” we would like to measure is the content of policies that Senate action yields. This is a messy problem. For one thing, the Senate does not act alone. The House must approve legislation, of course, and the ultimate outcome can depend on how legislation is implemented and its effects shaped by administrative and court action. For another, characterizing the policy approved by the Senate is no easy task. Legislation can be remarkably complex and involve dozens, if not hundreds, of distinguishable dimensions of policy preferences. And the Senate often must approve of a measure two or three times as versions of the measure bounce back and forth between the House and Senate. Which version, each of which may represent a bargaining position with the House or president, should be taken as the Senate’s most preferred version often is not clear.
A common practice is to examine roll-call votes—votes on which the yea or nay vote of senators are made a part of the public record of Senate proceedings (the Senate Journal, Congressional Record, and website). Political scientists have created accessible databases of congressional roll-call votes, although some data management skills are required to use them properly. A common source is voteview.com.
It is quite reasonable to ask, as Evans does, how many votes yield wins for senators representing a minority of Americans—what Evans calls “countermajoritarian” outcomes. “External countermajoritarianism” is generated by population inequalities that allow senators representing a minority of Americans to outnumber other senators. “Internal countermajoritarianism” introduces supermajority thresholds by counting votes on which a minority of senators representing a minority of Americans blocked a majority of senators from reaching the appropriate supermajority threshold.
Of course, not all votes are alike. Some concern procedural or minor policy matters, others concern amendments, many concern presidential nominations, and yet others concern the passage of bills or approval of conference reports or House amendments to Senate bills. For the period since 1945, Evans separated the major types of motions subject to votes and calculated the frequency of countermajoritarianism outcomes for each, taking into account the appropriate threshold.
Trends in Countermajoritarianism
Evans provides a detailed analysis of both forms of countermajoritarianism that I will not attempt to summarize here. I emphasize three key points.
First, external countermajoritarian vote outcomes vary widely in frequency but have spiked in recent decades. As the figure below indicates, in a couple recent Congresses with Republican majorities in the Senate have generated a countermajoritarian outcome on nearly 40 percent of Senate roll-call votes. That is, Republicans, while representing smaller states on average than Democrats, carry many Senate votes when they are in the majority
Evans emphasizes several features of this pattern. He points out that close votes are more likely to produce countermajoritarian outcomes—and in recent Congresses Republican majorities have been quite small. He also shows that a majority of countermajoritarian outcomes are on amendments, which are designed to modify the content of a bill, often substantially, and are offered by more minority than majority party members. And he shows that few countermajoritarian outcomes occur on final passage votes. That is, when senators have adjusted a bill to acquire the necessary votes for passage, often by accepting compromises to acquire 60 votes, countermajoritarian are rare.
Second, internal countermajoritarian vote outcomes have become quite common, too. The most common supermajority thresholds involve cloture votes that require three-fifths majorities to be adopted. As several my Notes show, the number of cloture votes has skyrocketed in recent decades. The following figure from Evans illustrates the frequency of countermajoritarian outcomes on cloture votes—that is, the frequency that a simple majority in favor of cloture is defeated by a large minority that represents a minority of Americans. In recent decades, well over a third of cloture votes were defeated by a countermajoritarian minority.[iii] Keep in mind that these (hundreds) of cloture votes that a simple majority supported but lost
Third, Evans observes that internal countermajoritarianism can reinforce or counteract external countermajoritarianism. Supermajority thresholds that underpin internal countermajoritarianism can make it easier for a minority of senators representing a minority of Americans to block action on legislation. At other times, a minority of senators representing a majority of Americans can use supermajority thresholds to block a Senate majority representing a minority of Americans. By examining recent Senate action on major legislation, Evans demonstrates that both possibilities occur with some frequency, although he notes that the two forms of countermajoritarianism reinforce each other far more frequently than counteracting each other.
Concluding Observations
Evans’s splendid analysis should spur additional debate about Senate representation and accountability. The Senate, like the Electoral College for electing a president, is becoming increasingly problematic as democratic institution designed for modern American government.
The Framer’s vision for the Electoral College, which seemed to hope for a deliberate body of political elites choosing a president, was undermined within a few election cycles by the emergence of parties and the self-interest of states. It has yielded a hybrid system that involves both a popular vote in each state and the formalities of an Electoral College. It also has yielded five national popular vote winners who lost the electoral vote count. Two of the five have occurred in recent times—2000 and 2016—when Republican presidents were elected.
In contrast, equal suffrage of the Senate was never based on a democratic vision. Rather, it was grounded in the practical necessity of acquiring the support of small states that were holding out for more influence than their populations justified. It was quickly rationalized as way to represent state governments in a federal system, but that was not the tenor of the discussion at the constitutional convention. Due to the special provision in Article V, it is (debatably) an unchangeable feature of American government.[iv]
The filibuster practice and associated cloture rule were invented by senators themselves decades after the Constitution was implemented. Although it took a few decades, Senate minorities, particularly southern minorities on issues of slavery and race, to exploit the institution’s lack of general rule limiting debate. It took another issue—self-defense of the U.S. merchant marine at the start of the first world war—to motivate the Senate to adopt a debate limit and then only with a supermajority requirement. Stalemate on today’s issues is motivating a new wave of interest in filibuster reform.
Modern democracies, including the American democracy, reflect numerous compromises over competing but deserving values. Representing majority opinion, accountability, and meeting evolving policy challenges are high among them, but representation of the range of opinion, stability, checking the exercise of power, and protecting civil rights and liberties are, too. The Evans paper reminds us that rebalancing these values requires that we reconsider how the Senate how it does its representational and legislative work.
[1] C. Lawrence Evans, “Senate Countermajoritarianism,” American Political Science Review (2024). doi:10.1017/S0003055424000510
[ii] George Mader, “Binding Authority: Unamendability in the United States Constitution: A Textual and Historical Analysis,” Marquette Law Review (2016), 841-891.
[iii] Also see Benjamin Eidelson, “The Majoritarian Filibuster” The Yale Law Journal (2013): 980–1023.
[iv] Mader, op. cit.