The Impact of Instant Runoff Voting on Representation for Women and People of Color

Introduction

In order to increase the representativeness of elections, candidates should be required to earn a majority of the vote to win. Instant runoff voting (IRV) — also referred to as ranked choice voting (RCV) — allows voters to rank multiple candidates in order of preference on their ballot. Winning candidates are required to receive more than 50% of the vote, ensuring they receive majority support from their constituency.  

Over 50 jurisdictions have implemented IRV; this includes two states (Alaska and Maine), three counties, and 45 municipalities. Another nine jurisdictions are awaiting implementation of IRV in upcoming elections.

One common criticism is that the system can disadvantage voters and candidates from historically underrepresented groups. However, research suggests the opposite is true: underrepresented groups, including first-time, young, female, BIPOC, and/or low-income candidates and voters, benefit from IRV elections.

This explainer first outlines how plurality and traditional runoff election systems disproportionately disadvantage voters and candidates from underrepresented groups. It then introduces IRV as a remedy to this problem and presents evidence from recent research demonstrating how IRV is a fairer system for all voters.  

The Problem

Plurality elections — in which the candidate with the greatest number of votes wins even if they do not have majority support — pose a number of systemic barriers to both candidates and voters from underrepresented groups. Evidence suggests plurality elections suppress certain candidacies disproportionately: Often, those with the least political experience or access to the political establishment forgo running for office, including potential first-time candidates, young people, women, Black, Indigenous, or People of Color (BIPOC) individuals, and/or members of the working class.

Additionally, runoff elections, held after the general election to elect a majority winner candidate, see a significant drop-off in voter turnout during the second election (particularly amongst underrepresented groups). Historically, runoffs were also implemented as a pernicious practice meant to dilute the voting power of Black voters. For example, turnout in primary runoffs decreased by an average of 43.5% amongst people of color in 2020, compared to a 38.3% decrease in white turnout. 

The Solution

Instant runoff voting gives all voters more voice and choice in elections with the ability to rank candidates in order of preference.

Here’s how it works:

1. Voters rank candidates on their ballot in order of preference. They rank their favorite candidate first, their second candidate next, and so on. If they prefer, voters can choose to still rank only one candidate;
2. First choices are counted. If no candidate earns a majority of votes (50%+1), the candidate with the fewest votes is eliminated;
3. Ballots in support of the eliminated candidate are transferred, in accordance with voters’ second choices. These votes are added to the remaining candidates’ totals. This process continues until one candidate earns majority support (or in multi-winner cases, until each seat is filled).

Instant runoff voting ensures full voter sentiment is captured in one decisive election, which eliminates the need for costly runoffs that draw low turnout. IRV elections also help prevent vote-splitting amongst similar candidates, mitigate against the spoiler effect, incentivize civil campaigning and coalition building, and lead to more fair and representative outcomes. 

Research on IRV and Representation

Below is an overview of existing research into how IRV impacts underrepresented candidates and voters.

Research on Candidate Experience

1. A 2018 study conducted by three scholars found California cities that adopted IRV saw an increase in the percentage of candidates of color running for office, as well as increases in the probability of female candidates and women of color winning office.
2. A 2021 report by FairVote that analyzed over 398 IRV contests and 1,422 candidates found that (1) candidates of color benefit from the round-by-round instant runoff process and (2) candidates of color pay no penalty when they run against opponents of the same race or ethnicity.
   
   The report noted the following: “Winning candidates of color, particularly those who are Black or Hispanic/Latino, grew their vote totals between the first and final ballot rounds at a higher rate than winning White candidates. Black and Hispanic/Latino candidates who went on to win grew their vote totals by 36% by the final round, compared to 28% among winning White candidates … This indicates that candidates of color can build strong support among voters outside their traditional political bases…”
3. A 2020 report from RepresentWomen concluded that women, people of color, and women of color win at higher rates in IRV elections than in plurality elections. Between 2010-2020, women won 48% of all municipal IRV elections. As a result, in 2020, municipalities with IRV had achieved near gender parity: Half of mayors and 49% of city councilors in these jurisdictions were women. Just 23% of mayors in non-IRV cities were women.

Research on Voter Experience

1. Exit polling from Alaska’s first IRV election in 2022 found that 85% of voters found the voting system simple, including at least 80% of Alaska Native, Black, Latino, and Asian/Hawaiian voters. Another report found that voters of color in Alaska were more likely than white voters to believe that their vote held more power under the new election system; 54% of Alaska Natives and 47% of non-Native voters of color felt this way compared to 34% of white voters.
2. Exit polling from NYC’s first IRV election in 2021 revealed that 95% of voters found their ballot simple to complete, including 93% of Black voters, 95% of Hispanic voters, and 97% of Asian voters. Across racial lines, about 80% of voters said they understood IRV with just nominal differences between groups.
3. A 2019 scholarly study of voters from cities with IRV found little evidence of racial/ethnic differences in voters’ self-reported understanding of IRV. Academics suggest voter education campaigns play a key role in increasing understanding of IRV and reducing potential voting errors.
4. A 2020 academic study found alternative ballot types (like ranking) were associated with smaller discrepancies in error proneness across age, race, and gender: The results question the conventional wisdom that single-choice voting is inherently easier for all voters. In a related study on 2020 presidential primaries, states that utilized IRV in their presidential nominating contests experienced more representative outcomes and increased voter turnout. 
5. Critics of IRV often claim that voters of color are more likely to see their ballot become “exhausted” because they tend to rank fewer candidates. However, evidence suggests that this is not the case. For example, a 2021 report that analyzed IRV elections found voters of color rank more candidates compared to white voters.
   
   Other research that analyzed New York City’s IRV elections concluded that voters’ use of rankings is more dependent on the context of an individual election than voters’ race or ethnicity. There were large variations in the number of rankings by voters of different racial and ethnic backgrounds in different city council districts. In some districts, voters of color were more likely to rank more candidates, while the same was true of white voters in other districts. This indicates that the context of the race — for instance, the level of competition, the number of candidates appearing on the ballot, and the race and gender of the candidates — played a far more significant role in influencing the number of candidates voters ranked than voters’ ethnic backgrounds. Thus, the conclusion is that no demographic group is disadvantaged by the ability to rank candidates.
   
   Further, a City University of New York study that also analyzed NYC’s 2021 IRV elections replicated these findings: All voters’ ranking patterns differed substantially depending on the context of a given race. While white voters were more likely than Black voters to rank more candidates in the mayoral race, the opposite was true in the comptroller race and competitive borough president and city council races. A similar pattern emerges among higher income and lower income voters, as well.
   
   In summary, the evidence does not support the theory that voters of color are less likely to rank candidates. Rather, many other factors about individual election influence voters’ decisions to rank candidates.

Case Studies on IRV and Improved Representation

• In New York City’s first IRV elections held in 2021, 31 women were elected to the city council, including 25 women of color. This marked the first time that women comprised a majority of the NYC city council.
• As of April 2020, nearly half of all mayors (46%) and 49% of all city council seats decided by IRV are held by women.
• Alaska’s first use of its top-four nonpartisan primary system, which includes instant runoffs in general elections, saw the election of Mary Peltola — the first woman and Alaska Native to represent the state in the U.S. House. That same year, women also won a majority (eight of 15) of state legislative elections that did not feature an incumbent (i.e., “open seats”).
• In the four Bay Area cities that use IRV (San Francisco, San Leandro, Oakland, and Berkeley), candidates of color have won 62% of races, compared to 38% prior to IRV’s implementation (despite only modest increases in their share of voters, and a decline in Black voters). 

Conclusion

IRV is certainly not a panacea for improved representation, and it must be complemented by other well-designed policies (voter and candidate education, clear ballot instructions/design, etc.). However, evidence suggests IRV can help underrepresented candidates and voters overcome barriers inherent under the status quo of plurality or runoff elections.