Expanding Representation: Reinventing Congress for the 21st Century
Introduction
Every two years, American voters in districts across the country cast a ballot for a single congressional candidate, understanding that the candidate with the most votes will win their district’s seat. Many take this basic design of our system as a given—that each district is represented by a single official and that the winner is the candidate who gets more votes than the next closest competitor.
In reality, there are a variety of ways to elect representatives. States have long experimented with other methods, and outside the United States, most democracies use a different system altogether.
As dissatisfaction with our politics has deepened, scholars and reformers have encouraged policymakers to consider America’s electoral system, or the system by which votes are translated into legislative seats, as one important source of dysfunction. Scholarship has linked our current American system, known as winner-take-all, to a broad array of issues affecting the practice of American constitutional democracy, such as escalating polarization and extremism, the underrepresentation of women and racial and ethnic minorities, a decrease in competitive legislative races, and even an increase in political violence. The methods we use to elect our representatives—often taken for granted—play one important role in the challenges confronting our politics.
A growing body of scholars and civil society groups across the country advocates replacing our system for electing lawmakers with a proportional system in which multiple representatives instead of one are elected from each district and seats are won in proportion to votes. Proponents of this change point to research that finds a variety of benefits associated with more proportional electoral systems, from tempering polarization and improving governance to ensuring that more voters are fairly represented.
ELECTORAL SYSTEM: the method by which votes are translated into legislative seats.
WINNER-TAKE-ALL: a system in which a single candidate with more votes than any competitor wins the entire district (that is, “takes all”).
PROPORTIONAL REPRESENTATION: a system in which multiple representatives instead of one are elected from each district and seats are won in proportion to votes.
The American Academy of Arts and Sciences’ report Our Common Purpose: Reinventing American Democracy for the 21st Century (OCP) recommends a specific form of proportional representation known variously as single transferable vote, proportional ranked-choice voting, or emergent proportionality.
This paper aims to help policymakers better understand this OCP recommendation, the backdrop of alternative systems, and the reasons why scholars and advocates believe that reforming our electoral system can spur positive change in the United States.
The American Academy of Arts and Sciences and Our Common Purpose
Founded in 1780 by John Adams, John Hancock, and other scholar-patriots of their day, the American Academy of Arts and Sciences has a long history of convening leaders from a wide variety of fields and backgrounds “to cultivate every art and science which may tend to advance the interest, honor, dignity, and happiness of a free, independent, and virtuous people.” Today, the Academy is both an honorary society and an independent research center that brings together Academy members and other experts in cross-disciplinary efforts to inform public policy.
The Academy’s current work on electoral system reform began in 2018, when it convened its cross-ideological and interdisciplinary Commission on the Practice of Democratic Citizenship to explore the state of our constitutional democracy. The Commission looked to the voices of experts and existing research but also conducted nearly fifty listening sessions across the country with a broad cross-section of Americans. That work led to the release of the OCP report in June 2020.
OCP is grounded in the theory that, in a healthy democracy, strong civic institutions and civic culture reinforce and are reinforced by responsive government institutions, leading to a “virtuous cycle.” In America today, however, we are caught in a “vicious cycle”: our institutions are not sufficiently responsive, which causes individuals to disengage, which further reduces the responsiveness of the institutions, and so on. To remedy this, OCP offers six broad strategies and thirty-one specific recommendations intended to address all parts of this cycle.
The first set of recommendations focuses on improving representation and giving all Americans a greater sense of agency over the government decisions that impact their lives. One of those—Recommendation 1.3—calls for reimagining the system currently used to select members of the U.S. House of Representatives. Since 1967, federal law has mandated that states use single-member congressional districts. This arrangement produces a winner-take-all electoral system, wherein a single representative wins the entire district (“takes all”). The Commission called for replacing this system with a specific type of proportional system, termed by political scientists single transferable vote (STV). This system is also commonly known in the United States as proportional ranked-choice voting, and some advocates refer to it as emergent proportionality.3
Winner-take-all and proportional systems like STV are distinguished by a variety of important features and effects. But the principal distinguishing factor, as the name suggests, is that proportional systems result in greater proportionality in outcomes. For instance, in a proportional system with a district electing three or four congressional seats, a minority group or party supported by about one-quarter of the voters will typically receive at least one seat. The share of seats won by the group thus would be roughly in proportion to the share of voters it represents. By contrast, under winner-take-all, a minority group of that size would almost certainly receive no seats at all, leaving the group’s voters without representation of their choice. As the OCP report explains, enacting Recommendation 1.3 would “signal a victory for equal voice and representation.” Implementation of STV is framed in the report as a logical next step that could build on other recommended reforms, including the single-seat version of ranked-choice voting (Recommendation 1.2 in the report).
“Amend or repeal and replace the 1967 law that mandates single-member districts for the House, so that states have the option to use multi-member districts on the condition that they adopt a non-winner-take-all election model.”
—Our Common Purpose, Recommendation 1.3
Most democracies throughout the world use proportional electoral systems, and those systems are as different as the nations that created them. Within the United States, a handful of cities also uses proportional systems for local elections. The STV-type system recommended in OCP (which is also the type used in Portland, Oregon, and Cambridge, Massachusetts) is different from the proportional systems used in most other countries in that the improvements in representation it engenders need not be mediated through the party structure. Internationally, versions of STV are used in Ireland, Australia, and Malta, as well as by various subnational governments.
The Working Group
As attention to proportional representation has increased, the need for a clear and thorough resource that policymakers can use to understand the ins and outs of electoral system design has become apparent. To address this need, in March 2024 the Academy convened the Electoral Design Working Group, which is composed of political scientists, election law experts, advocates, practitioners, and other leaders with diverse backgrounds and ideologies. In addition to relying on their own expertise and the available academic research, the group conducted outreach to state and federal policymakers and activists to better understand their perspectives on the issues at hand. That outreach occurred alongside a related public opinion research project, for which the working group served in a consultative capacity.
All the members of this working group agree, at a minimum, that the STV-type electoral system recommended in OCP would be an improvement over the current winner-take-all scheme. Opinions among the group are more varied with respect to other forms of proportional representation: some of the working group members prefer STV over other systems. These members argue that alternatives (like the party list systems common in the rest of the world) have drawbacks that make them less viable and effective in our American constitutional system and political culture. Other members of the working group believe that these concerns may be overstated and that other proportional representation systems may in fact work better than STV in the United States. Still other members believe that it is impossible at this stage to know what specific system would work best, and states should thus have the option to experiment with a variety of approaches. Finally, some are hesitant to move forward with any specific system in the absence of a more comprehensive process that meaningfully provides opportunities for agency to historically marginalized groups and other stakeholders.
This paper aims to lay out the policy choices required to implement any new electoral system and to outline the benefits most likely to result from discarding winner-take-all. In so doing, we hope to make the case for abandoning winner-take-all, as recommended in OCP, and to facilitate greater understanding of both the specific system proposed in that report and the full spectrum of possibilities. Where possible, the group has made actionable recommendations; elsewhere it has laid out possibilities and articulated open questions. We enthusiastically invite further research and debate, and we hope that this work will spark an informed discussion of how better electoral system design could make American constitutional democracy more representative, responsive, and resilient.
Research Methods
A principal challenge of predicting the outcomes from any given proportional system of representation in the United States is the relative lack of empirical evidence in the U.S. context. Only one state—Illinois—has ever used a non-winner-take-all electoral system (in that case, a semiproportional system), and never has one been used for a congressional election. While the study of electoral systems is buoyed by substantial research today, extrapolating lessons for U.S. elections given the dominance of the winner-take-all model throughout American history can be challenging.
This paper draws from three principal categories of scholarship to offer insights into the potential implications of adopting the OCP reform or any other form of proportional representation. First, it draws from available domestic research, even if limited. For example, city councils and school boards across the country use semiproportional systems to elect their members, and researchers have studied their results on dimensions like the representation of women and racial and ethnic groups. Second, it draws from cross-national or comparative research, or research that compares experiences among various countries. For example, cross-national evidence indicates that proportional systems are correlated with lower levels of affective polarization. This means that, in countries that use proportional systems, voters are less likely to think negatively about members of opposing parties. Third, it features certain simulation exercises of proportional representation in the United States. For example, modeling shows mathematically the near impossibility of gerrymandering under certain proportional rules. Many members of this working group have helped to produce this research.
While none of these data points can themselves predict how a different electoral system might perform for, say, the U.S. House, they do offer directional guidance. For instance, domestic evidence shows improved racial and ethnic minority representation under semiproportional rules across local U.S. jurisdictions; comparative research generally finds the same in proportional systems across other democracies and finds that proportional systems can be one factor in increasing the number of women who run for and win elected office; and simulation exercises produce similarly improved racial and ethnic representation for U.S. congressional elections. While predicting actual results remains impossible, this evidence does offer directionally useful information. And although general insights can obscure important exceptions and nuances, our hope is that distilling multiple sources of scholarship can, as a starting point, offer meaningful suggestions about potential implications.
Overview of the Paper
Section I of this paper provides a summary of electoral systems and the two broad classes (proportional and winner-take-all) into which they may be divided. It also describes the main takeaways from the working group’s stakeholder interviews and public opinion research to provide insight into how both the grassroots and local and national civic and political leaders may react to electoral reform. Section II discusses the design choices that would be required to implement any change to our electoral system, including a change to the type of proportional system recommended in OCP. Finally, Section III outlines impacts that might be expected from moving to a proportional system.