Electoral Reform Research Group

With growing national interest in reforming American political institutions, the Electoral Reform Research Group—a collaboration between New America, the Scholars Strategy Network, Stanford University’s Center on Democracy, Development, and the Rule of Law, the American Enterprise Institute, and the Unite America Institute—are organizing emerging research into how changes in electoral rules impact political participation, processes, partisanship, power, and policy outcomes.

Last year the ERRG released a call for proposals for research into the impacts of ranked-choice voting and related reforms. With the help of our advisory board we selected 14 of the proposals to develop into full-fledged research. Their findings will ultimately be published by New America, academic journals, and other outlets.

The projects are summarized below. Topics covered include minority representation, voter satisfaction, information and complexity, moderation and ideological representation, and polarization.

Current Projects

Minority Representation

Does switching from at-large plurality to ranked-choice voting structurally promote minority representation in general?
Tufts Metric Geometry and Gerrymandering Group Redistricting Lab

The MGGG Redistricting Lab, led by Dr. Moon Duchin, has already examined ranked-choice voting (RCV) as a way to empower minority representation in a number of specific cases, including Lowell MA, Chicago IL and Santa Clara, CA. Motivated by the growing interest in RCV as a way to improve minority representation in local government, they now seek to answer the following question: When does switching from at-large plurality to RCV make sense for minority representation and how much does it affect minority representation? To answer these questions, they will adapt and apply statistical models developed in their previous work to real world cities where a change to RCV could increase minority representation. Practitioners have been particularly enthusiastic about this project, agreeing that the results would be extremely helpful to them in the field.

Does ranked-choice voting (RCV) lower the barrier to entry for candidates from underrepresented groups?
Jamil Scott (Georgetown University) and Jack Santucci (Drexel University)

Despite electoral gains for women and people of color across all levels of office in 2018, these groups remain under-represented. Why? Interest in running for office and the decision to run matter: Political interest can vary by race and gender, but how electoral systems impact candidate diversity in office is understudied. Does ranked-choice voting (RCV) encourage more candidate entry, among women and people of color in particular? Does the type of representation people want vary with whether politicians “look like” the people they represent? Scott and Santucci will study these and other questions via a survey experiment in Philadelphia and include their representation-oriented questions in a national survey that oversamples people of color. This project is unique in its focus on descriptive representation and candidate emergence, and has the potential to transform the field.

Lessons for Ranked-Choice Voting and Minority Representation from around the World
Michael Latner (California Polytechnic State University, Union of Concerned Scientists) and Matthew Shugart (University of California-Davis)

Two of the most experienced and widely-published scholars of electoral systems, Latner and Shugart are extending recent advances in comparative research to analyze the relationship between electoral rules and minority representation using a cross-section of municipal elections across the globe. They are building a comparative database of municipal election results from the US, Australia, Ireland, Northern Ireland, Scotland, Malta and New Zealand, including cities using districts with varying magnitudes, and categorical and ordinal (transferable) electoral formulas across those magnitudes, within both partisan and nonpartisan election contexts. Because the Union of Concerned Scientists also seeks to expand representation in the field of election research, they are assembling a network of early-career researchers (ECRs) from around the world to help in the data collection and analysis effort, and will mentor those ECRs toward journal publication. This project, perhaps more than any other here, reflects the community-building, inclusive spirit of the ERRG.

Does Ranked-Choice Voting Reduce Racial Polarization?
Yuki Atsusaka (Rice University)

Statistician and political scientist Yuki Atsusaka is using election and demographic data to test whether RCV promotes moderation and whether RCV decreases racially polarized voting in American elections. He identifies a number of obstacles that researchers face in studying RCV’s effects on racial polarization, including non-standard data formatting across localities, lack of individual-level data on race and vote choice, and the self-selecting nature of RCV-using municipalities. To overcome these obstacles, he is creating a comprehensive dataset of RCV and non-RCV elections in Bay Area mayoral elections; developing a novel method to analyze RCV data, as well as a compatible measure of racial polarization across two types of elections; offering a more reliable ecological inference technique and its validation; and applying state-of-the-art causal inference tools to the data set. Atsusaka’s highly mathematical approach is unlike anything else in our ERRG portfolio thus far and could pave the way for multiple future studies.

Evaluating Voter Experiences and the Traits of Supported Candidates in Ranked-Choice Elections
Shana Kushner Gadarian (Syracuse University), Jessica Trounstine (University of California–Merced), and Melody Crowder-Meyer (Davidson College)

Recognizing the challenges of studying RCV in the real world (cities with RCV are not representative of American cities more generally and each election may have several confounding factors, etc.), this trio is investigating the effects of RCV on election outcomes and voter perceptions of representation using survey experiments with a large, representative sample of respondents. The team has significant experience running survey experiments on electoral institutions, and both broad and deep knowledge of American city politics, representation, and campaigns. With this project, they are harnessing their joint expertise to test a range of hypotheses which could be of use to advocates, including the comparative impact of RCV and majoritarian settings on minority candidate success, across multiple dimensions.

Ideological Representation and Moderation

Impact of Electoral Context on Affective Polarization
Yphtach Lelkes (University of Pennsylvania)

Affective polarization poses an existential threat to society: it increases congressional gridlock, it leads to a breakdown in democratic norms, and has been linked to increased prejudice. Lelkes is among the foremost experts on affective polarization in the United States. With ERRG’s help, he’s working to answer the question, “Is America more prone to affective polarization because of its winner-take-all system?” by running experiments that are designed to mimic the rules of electoral institutions in a simple enough way that we can test their implications in a laboratory setting. His sequential game-based research design will test, among other things, whether voting under RCV (and proportional representation) rules will reduce affective polarization, and whether the effect of winning or losing an election on affective polarization will be smaller under RCV.

Representation and Ranked-Choice Voting
Arjun Vishwanath (Harvard University)

Vishwanath’s project examines the effects of ranked-choice voting on the caliber of substantive policy representation in the United States. Specifically, he will use observational data in constituencies that have implemented ranked-choice voting in the United States to assess whether the behavior of legislators in these places more closely aligns to the preferences of their constituents. He will measure representation within races in RCV jurisdictions to identify causal changes based on the design of winner-take-all and RCV systems, collecting and analyzing data (i.e. election results, interest group endorsements, candidate surveys) from San Francisco, CA; Berkeley, CA; Oakland, CA; San Leandro, CA; Burlington, VT; Minneapolis, MN; Pierce County, WA; and Santa Fe, NM.

Ranked-Choice Voting and Candidate Extremity
Melissa Baker and Kathryn Herzog (University of California–Merced)

Compared to plurality voting, does RCV affect the electability of extreme candidates? What about voters’ ideological self-identification when faced with extreme candidates? Per Social Comparison Theory, this team hypothesizes that moderate candidates would be more electable in an RCV setting—which should offer more options across the ideological spectrum—than in a plurality context. This project will test respondents in both electoral settings on their reactions to the presence of extreme candidates, as well as their perceptions of how well those candidates represent the ideological position of the electorate. With their results, Baker and Herzog aim to understand how strategic voting works differently under RCV, and thus inform how candidates and parties can be strategic during ranked-choice elections.

Voter Satisfaction, Choice, and Complexity

How Do Citizens Evaluate Ranking?
Andre Blais (University of Montreal), Carolina Plescia (University of Vienna), and Semra Sevi (University of Montreal)

This team, headed by eminent electoral systems scholar Andre Blais, aims to understand voters’ preferences for different types of voting rules. They conducted two independent but complementary survey experiments around Super Tuesday, and their analysis will shed light on the three main questions: do voters prefer RCV (as well as point and approval voting) to the single vote; is RCV preference a function of education level; and is RCV preferred by those with more extreme views about candidates?

Ranked-Choice Voting, Runoff, and Democracy: Insights from Maine and other U.S. States
Cynthia McClintock and Joseph Cerrone (George Washington University)

Taking advantage of the implementation of RCV in Maine, this research will compare the impact of RCV, runoff, and plurality in U.S. states. It asks whether runoffs and RCV encourage ideological moderation, open the electoral arena to new parties, and, ultimately, increase citizens’ satisfaction with democracy. And if so, which approach appears more favorable? This team, led by distinguished professor of political science and Latin American affairs Cynthia McClintock, is leveraging a multi-method design to explore these questions, including primary field research in Maine, with secondary research for comparable areas, and a conjoint survey experiment.

Testing Public Support for Ranked-Choice Voting
David Kimball (University of Missouri–St. Louis) and Joseph Anthony (Oklahoma State University)

How do voters in the United States evaluate reforms like RCV? Renowned political science expert David Kimball and organizer-turned-political scientist Joseph Anthony have teamed up to answer this question through a pair of survey experiments: The first is a module on the 2020 CCES survey and the second is a module on the 2020 Collaborative Multiracial Post-Election Survey (CMPS), which is a very large sample (N=20,000) and includes large subsamples of Latinos, African Americans, and Asian Americans (N=4,000 for each group). For the second they will join forces with another ERRG team, Jamil Scott and Jack Santucci, of Georgetown and Drexel, respectively. In practical terms, this study tests whether a fuller description of voting rules influences voter assessments of those rules.

Understanding the Support for Ranked-Choice Voting in American Cities
Michael Cobb (North Carolina State University), Richard Engstrom (University of Houston), and Baodong Liu (University of Utah)

The team is collecting opinion data about citizens’ perceptions of ranked-choice voting from election surveys, both online and by mail, from a group of states holding Democratic presidential primaries using ranked-choice voting: Nevada, Wyoming, Hawaii, and Kansas. Building on original exit poll data from two Southern cities, they also aim to collect all extant exit polling data from across the country for meta-analysis. One of the most pressing questions among democracy reformers this primary election season concerns how RCV performed in the aforementioned states given this the extraordinarily large Democratic primary field and the confusion surrounding coronavirus. This project can give us some answers.

Who Ranks Best (and Least) among Us? An Experimental Study of Voter Decision-Making and Political Information in Ranked-Choice Voting Systems
Scott MacKenzie and Cheryl Boudreau (University of California–Davis)

This project seeks to understand whether and under what conditions voters take advantage of the greater opportunity for political expression that San Francisco’s new ranked-choice voting system provides. MacKenzie and Boudreau are two UC–Davis political scientists with a strong track record of studying voter choice in San Francisco municipal elections under RCV. For the 2020 cycle—when voters will be able to rank up to ten candidates on their ballots, up from three—their candidate and voter surveys will assess the effects of political information on several aspects of voters’ decision-making, including the propensity to use all of their allotted rankings and whether these rankings reflect their policy views. The surveys will also provide a first appraisal of San Francisco’s new RCV system, which is likely to serve as a model for RCV reforms elsewhere.

Vote Choice and Voter Error with Ranking Ballots
Jason Maloy (University of Louisiana–Lafayette)

Do voters make different political judgments when given different voting rules for the same candidates and contests? This study gives one ballot to each of several sub-samples and analyzes them statistically to see if there are significant differences across ballot types. The experiment was performed using an online questionnaire and samples from Super Tuesday voters in Tennessee, Colorado, and Texas. With such a wide geographic and ideologically diverse sample, the results of this experiment will contribute significantly to our understanding of whether and how changing voting rules changes election outcomes.

ERRG Organizing Committee
Lee Drutman - Senior Fellow, Political Reform Program, New America
Didi Kuo - Senior Research Scholar, Center on Democracy, Development, and the Rule of Law, Stanford University
Avi Green - Executive Director, Scholars Strategy Network
Kevin Kosar - Resident Scholar, American Enterprise Institute
Nick Troiano - Executive Director, Unite America Institute

ERRG Advisory Board
Heather Balas - Good Government Reform Policy Officer, Thornburg Foundation
George Cheung - Director, More Equitable Democracy
Larry Diamond - Senior Fellow, Hoover Institution and Freeman Spogli Institute for International Studies, Stanford University
Francis Fukuyama - Senior Fellow, Freeman Spogli Institute for International Studies, Stanford University
Keesha Gaskins-Nathan - Director, Democratic Practice–United States program, Rockefeller Brothers Fund
Ruth Greenwood - Co-Director, Voting Rights and Redistricting, Campaign Legal Center
Kristen A. Johnson - Assistant Counsel, NAACP Legal Defense and Educational Fund, Inc.
Yuval Levin - Director, Social, Cultural, and Constitutional Studies, American Enterprise Institute; Editor in Chief, National Affairs
Michael Li - Senior Counsel, Democracy, Brennan Center for Justice
Sam Mar - Vice President, Office of the Co-Chairs, Arnold Ventures
Eric Maskin - Adams University Professor, Harvard University
Jeanne Massey - Executive Director, FairVote Minnesota
Zachery Morris - LDF/Fried Frank Fellow, NAACP Legal Defense and Educational Fund, Inc.
Maria Perez - Principal Consultant, Democracy in Action
Ezra Rosenberg - Co-Director, Voting Rights Project, Lawyers' Committee for Civil Rights Under Law
Daniel Stid - Director, Madison Initiative, William and Flora Hewlett Foundation
Charles Stewart III - Kenan Sahin Distinguished Professor of Political Science, MIT
Jessica Trounstine - Professor of Political Science, University of California, Merced