Ranked Choice Voting: Avoiding a One-Size-Fits-All Approach

In recent years, Ranked Choice Voting (RCV) has been pitched as a panacea, the fix for all that ails American elections.¹ In 2024 alone, RCV will be on the ballot in as many as five states and Washington, D.C, with more proposals likely in future years.

At Responsive Gov, we think that structural reforms must be closely tailored to a problem in a particular state that the reform is trying to solve. In particular, reforms like RCV that change the outcome of who wins elections and require voters to use a new and unfamiliar system must be deployed carefully in the particular political and implementation environment of each state.

In an environment where RCV is seemingly up for consideration everywhere, policymakers and the public should ask themselves if it makes sense in their particular state. Voters and legislators considering RCV should carefully assess the changed or unpredictable electoral outcomes RCV will bring in their state, including for candidates of color, and whether these outcomes further or undermine their goals. Likewise, RCV debates should seriously consider practical issues of implementation, including risks of increased voter confusion, voter error, and lower turnout, with potentially higher effects among low-income and low-education voters. Asking these questions is all the more important in states considering adopting RCV through a constitutional amendment, which is significantly more difficult to change or undo.

In some scenarios, RCV may make sense in a particular state. But legislators, voters, and reformers should avoid treating it as a one-size-fits all solution, and instead demand a strategic case-by-case plan that considers the impact of the change in their political geography.

Background on RCV

RCV is a form of voting in which voters rank candidates in order of preference. Under RCV, rather than casting a vote for one candidate, voters rank candidates in order of preference, with tabulation proceeding in rounds. In a contest with RCV, if no candidate earns more than 50% of the vote among first-choice ballots, the candidate finishing last is eliminated, and the ballots voters cast for the eliminated candidate are reallocated to those voters’ second choices. If any candidate has more than 50% after this reallocation, that candidate is elected. Otherwise, this process of elimination and reallocation continues until one candidate earns more than 50%.

RCV has been used for local elections for decades, but in recent years, there has been a sustained push to adopt the policy for state and federal elections. Thus far, Maine and Alaska have adopted RCV for some state and federal elections.

RCV is often paired with a top-four blanket primary. Alaska uses this system and similar proposals are on the ballot or attempting to qualify for the November 2024 ballot in Nevada, Montana, and Colorado. With a top-four primary, all candidates run in the same primary, regardless of party, without voters ranking candidates. The top four candidates in the primary advance to the general election, which uses RCV. (Nevada is proposing a top-five primary, where the top five candidates advance to the general, which would proceed under RCV rules.)

Notably, in Colorado, Nevada, and Montana, these changes are being proposed as amendments to the state constitutions. By contrast, both Alaska and Maine have adopted RCV by statute. Constitutional amendments on RCV would effectively lock these states into these reforms, limiting the ability of the legislature to revise procedures in response to changed circumstances, implementation challenges, or unintended consequences. These provisions could only be changed through an additional constitutional amendment.

Changed Electoral Outcomes

By design, the combined reforms of RCV and top-four primaries change electoral outcomes in a particular state. In these systems, the general election is no longer a head-to-head contest between the winner of the Democratic primary and the winner of the Republican primary, where the person who wins the most votes prevails. Instead, a top-four primary and a RCV general create rather a more unsettled and unpredictable general election, with a larger number of candidates, options from across the political spectrum, and indeterminate vote shares as tabulation proceeds in rounds.

Accordingly, the first question voters and legislators in states considering RCV reforms should ask themselves is if they actually want these changed electoral outcomes and heightened unpredictability in high-stakes state and federal contests. Consider states like Colorado and Nevada, where Democrats routinely win state and federal elections as a result of winning the Democratic primary and then defeating conservative Republicans in a head-to-head contest. Or consider the past three Senate races in Nevada, where Democrats have narrowly eked out victories over Republicans, with serious consequences for control of the United States Senate. A shift to RCV in these states would dramatically upset this dynamic. Instead of a contest between a conservative Republican and a moderate or Progressive Democrat (depending on the district), the contest would be a four-way (or five-way) battle between candidates all along the political spectrum. Such a system introduces considerable uncertainty and unpredictability, with serious consequences for political and policy outcomes.

RCV has the power to significantly change electoral outcomes. Therefore, voters and legislators considering RCV reforms should be strategic in deciding whether to support these changes in their state. Advocates for RCV should ask themselves if RCV will cause them to lose state and federal elections that they currently win. If voters and legislators in a particular state actually like the results that currently occur, they may not want to upset the apple cart.

Consequences for Candidates of Color

Voters and legislators considering RCV should also carefully assess the consequences for candidates of color and minority representation in their state. RCV may ultimately hurt candidates of color by undermining traditional paths to power for candidates of color in state and federal contests.

To be sure, RCV provides clear structural benefits to minority voters in multi-winner contests (like a city council race where multiple seats are up for election), because it mimics the effects of proportional representation.² But no similar structural benefit exists in single-member contests, where only one seat is up for election. Indeed, in single-member contests, the combination of RCV and a top-four primary may actually undermine the success of candidates of color, depending on the circumstances.

For example, consider a circumstance where a minority candidate runs in a Democratic-leaning state legislative district with a large, but not plurality, minority community. Such a candidate could prevail in a Democratic primary, due to strong support from voters of color, and then win the general election due to unified Democratic support. However, in an election with RCV and a top-four primary, this path to victory would be far more difficult and uncertain. Such a candidate would have to face off against a broader field of candidates in the general, and could no longer rely on a victory in the Democratic primary as a springboard to a general election win.

While not all districts or elections will fit this profile, interested parties in a state debating RCV should consider the impact such a change could have on minority representation in state and federal elected offices. Before adopting RCV, states should assess whether legislative or executive offices that currently provide a path for minority representation would be subject to unpredictable or different outcomes, potentially undermining the goal of representation for candidates and communities of color.

Practical Consequences for Voters: Confusion, Error, and Turnout

Decision-makers considering RCV should also give thought to practical consequences. Several studies suggest that RCV is associated with increased voter confusion and error and decreased voter turnout, with heightened effects among low-income voters, voters with lower levels of education, and communities of color. Voter error and confusion not only harm voters, but also raise concerns of tipping outcomes in close contests, and states should consider these consequences when debating RCV.

RCV can increase voter error. A recent study incorporating data from Alaska and Maine found that 1 in 20 voters improperly mark ballots in RCV contests.³ These errors can include overvotes (where a voter selects more than one candidate for a given ranking), overranking (where a voter ranks the same candidate more than once), and skips (where a voter leaves a ranking blank but fills in a subsequent ranking). Other studies similarly find that ballot errors in RCV elections are particularly high in areas with lower levels of education, lower levels of income, higher minority populations, and a higher share of limited English proficient voters.⁴ While not all of these errors result in a ballot rejection, the study cited above found that ballots in RCV elections are ten times more likely to be rejected than ballots in an non-RCV election.

Moreover, even if an error doesn’t result in a rejection, a ballot with an error may be an indication of voter confusion or a failure to accurately capture voter preferences. Recent studies indicate that as many as 16% of voters in RCV elections were somewhat or very confused by the process, with significantly higher rates among Latino voters.⁵ Similarly, in Maine, a study found that RCV “produced significantly lower levels of voter confidence, voter satisfaction, and ease of use.”⁶ Other studies find that, perhaps due to this increased voter confusion and decreased satisfaction, RCV may also be associated with decreased turnout.⁷ This may be true particularly among communities of color, as a study of San Francisco mayoral elections found a significant decrease in Black turnout with RCV.⁸

In any other context, election advocates would raise alarm bells about a policy that results in a 5% error rate, heightens voter confusion, and potentially decreases turnout, particularly with heightened effects among low-income and low-education voters. But thus far, the same level of alarm has seemingly not occurred with RCV, even though the consequences are just as significant. For example, consider a swing state like Nevada, where extremely close contests are the norm for President and Senate elections. If these races switched to RCV, the resulting voter error and confusion could affect tens of thousands of voters and easily tip the balance of power for both the Presidency and control of Congress.

Voters and legislators considering RCV should ask themselves if they are prepared for these implementation challenges and the risks they carry for election outcomes. Before moving forward with RCV, states need a well-funded, well-designed, and sustained implementation plan designed to counteract these challenges in all elections, as well as buy-in from election administrators to address these concerns. Such efforts should be particularly designed to reach low-income, low-education, and minority voters, who may experience the most challenges from implementation. In the absence of such commitment, RCV may reasonably be viewed more cautiously in a state, given the risks it carries for voters as well as election outcomes.

Conclusion

Before states rush headlong to lock themselves into RCV, voters and legislators should first seriously ask if RCV is actually the right fit for their state. This assessment should weigh changed and unpredictable electoral outcomes in the state, particularly for communities of color. It should also consider practical risks for voters and electoral confidence. By asking these questions, states can avoid being swept up in a national trend. Ultimately, voters and legislators must weigh the promises and pitfalls of RCV in the context of their particular state.

^{1 See National Public Radio, Ranked Choice Voting is Being Touted as a Cure-all for U.S. Deep Partisan Divides, Dec. 3, 2023.
2 See, e.g., Consent Decree & Judgment, United States v. Eastpointe, No. 4:17-cv-10079 (E.D. Mich. July 24, 2019).
3 Stephen Pettigrew & Dylan Radley, Ballot Marking Errors in Ranked-Choice Voting (Dec. 20, 2023).
4 See Lindsey Cormack, More Choices, More Problems? Ranked Choice Voting Errors in New York City (Dec. 15, 2023); Corey Corey & David Latterman, Ranked Choice Voting in the 2011 San Francisco Municipal Election: Final Report (2011).
5 See Lonna Rae Atkinson et al, The Impact of Voter Confusion in Ranked Choice Voting (Mar. 19, 2024).
6 Jesse Clark, The Effect of Ranked-Choice Voting in Maine (March 18, 2021).
7 See Jason A. McDaniel, Electoral Rules and Voter Turnout in Mayoral Elections: An Analysis of Ranked Choice Voting (June 28, 2019) (finding a 3-5% decrease in voter turnout with RCV).
8 See Jason McDaniel, Writing the Rules to Rank the Candidates: Examining the Impact of Instant-Runoff Voting on Racial Group Turnout in San Francisco Mayoral Elections (March 22, 2014).}