The Future of Ranked-Choice Voting

The November elections of 2016 and 2020 and the chaotic, norms-shattering events that have followed illustrate just how strongly nearly half of all Americans oppose their only other choice for President. The harmful effects of our polarized political life will certainly intensify unless we open our political system to independent candidates by making a fundamental change in the way we vote.

A poll of fifty thousand Americans we commissioned last year showed that nearly 70 percent, and 80 percent of those under the age of forty, said they want to see independents on the ballot to offer real competition to Democratic and Republican candidates.

Today, voters almost never elect independent candidates, because, with few exceptions, our elections are decided by first-past-the-post plurality voting (FPPV).

We mistakenly rely on a system in which a voter’s expression of preference is restricted to a single candidate. Each voter has just one choice, and if there are more than two candidates in the race, then winning by a plurality rather than a majority is quite possible. Consequently, no matter how attractive an independent candidate may seem in the spring, summer, and early fall of an election year, he or she will be tarnished as a “spoiler” on Election Day and will almost certainly lose. This unfortunate situation reduces the supply of independent candidates willing to compete and perpetually forces Americans into one of two warring factions.

In contrast, ranked-choice voting (RCV) lets voters express an opinion about every candidate by ranking all candidates in order of true preference. If no candidate wins an outright majority, the candidate with the lowest number of first-place votes is eliminated, and the second-preference votes of his or her supporters are redistributed to the remaining candidates. This “instant runoff” process continues until there is a majority winner. Not only does RCV give voters “more voice” in elections, it has the potential to stop our political system from tearing us apart into two camps.

Wouldn’t it be a vast improvement to be able to vote for the person you truly like the most without being afraid that you will end up electing the person you like the least? If all voters could do this, then,

·      Voters in most elections would have more than two choices.
·      Voters would not feel they had to choose between the lesser of two evils.
·      If a voter’s first choice did not win, at least the voter’s next-favorite choice could win, which would be more satisfying.

The shift to RCV would also change the incentives for candidates:

·      Because independents cannot be dismissed as spoilers, candidates who are not career politicians may be more likely to run for office.
·      Winning candidates will always assume office with the support of the majority, giving them increased legitimacy.
·      Incumbents not needing major party support for re-election will be able to prioritize constituents’ interests first.

On November 6, 2018, RCV was employed statewide for the first time in Maine’s congressional elections, after a multiyear effort led by the Committee for Ranked Choice Voting and the Chamberlain Project, organizations we co-founded with former state senator Dick Woodbury (ME) and state representative Kyle Bailey (ME). Starting in late 2014, our campaign organized volunteers across Maine and collected more than 61,123 valid signatures to put RCV on the November 2016 ballot. Yet, even after the measure was approved in a free and fair election, bipartisan political opposition to RCV gathered in Augusta. In 2017 the measure was repealed by both Houses of the State Legislature.

Then, as Larry Diamond has described in his recent book, Ill Winds (2020), we collected another 61,123 signatures—this time in just 88 days, during the worst winter weather in decades—and won a series of lawsuits to put the measure on the ballot again. Unwilling to implement the law on his own, Maine’s secretary of state required a judge to order him to print the ballots that allowed voters to use RCV for the first time in statewide primaries. On June 12, 2018, Maine voters picked their candidates using RCV ballots and once again voted in favor of the continued use of RCV in future elections. In November of 2018 Maine chose the country’s first congressman to be elected by RCV. In 2020 Maine voters, for the first time in U.S. history, used RCV ballots to choose a President.

RCV was never expected to win the first time, let alone the second. But its surprising victories helped spark a surge in RCV activity in other parts of the country:

·      In 2019 New York City adopted RCV with 73 percent of the vote. In June of 2021 New York City Democrats used RCV to elect candidates for mayor, comptroller, and public advocate, in an election with the largest number of ranked-choice votes ever cast.
·      In 2020 Alaska passed a referendum measure that combined RCV with “final four” voting, an expansion of open primaries, and restricted use of “dark money.”
·      In 2021 Virginia Republicans used RCV to choose their candidates for governor, lieutenant governor, and attorney general.
·      Also in 2021, voters in Austin, Texas, adopted RCV by 58 percent to take the place of run-offs in the city.

On the other hand, a 2020 Massachusetts referendum rejected RCV. Even there, however, the campaign had its bright spots. The measure got more than 1.5 million “yes” votes. If Governor Charlie Baker hadn’t made his opposition public in a press conference on the eve of Election Day, it might well have passed.

Still, RCV has a long way to go before it will supplant FPPV as the dominant method of voting in America. Supporters of RCV elections should consider five critical objectives.

First, commit to a national strategy with a ten-year implementation timeline.

A sequential, one-by-one, state-by-state approach will almost certainly fail because bad behavior by the two major parties will be forgotten between election cycles. A strategy that operates in multiple states at once stands a better chance of revealing common tactics of obstruction and prompting national outcry.

This national strategy should assume that in any single state, it will take an average of four to five years of sustained activity to implement RCV. The math is clear: Around ten RCV campaigns would need to be launched each year for the next five years to get most states to use RCV instead of FPPV within a decade.

This ambitious timeline would require a national organization of talented professionals experienced in successful campaigns, and financial supporters prepared to contribute hundreds of millions of dollars. This may seem a daunting number, but it is only a fraction of the aggregate spending by presidential candidates in a single election cycle.

Second, support the most effective approach to winning RCV in every state.

There are many different, though potentially complementary, perspectives on the value of RCV.

Some favor it because they think it will increase the chances of electing officials who will not necessarily obey the dictates of parties currently dominated by the extremes. For example, Democrat Jared Golden of Maine was the first congressman elected through RCV. In the first round of counting, Golden trailed incumbent Republican Bruce Poliquin. Neither candidate earned a majority. But when the second choices of the independent candidates were tallied, Golden emerged as the winner with 142,440 votes. This means a candidate winning a first-round plurality can lose if he only has support from his partisan base. Moreover, Rep. Golden’s re-election strategy is different from that of a typical Democrat: He has voted against his party more times than any other sitting Democrat.

Others say RCV should be implemented in every state to level the playing field between unaffiliated candidates versus Democratic and Republican candidates. The expectation is that over time a coterie of independents will emerge in Congress to challenge the dysfunctional duopoly.¹ Without RCV the two major parties will continue to frustrate competition while intensifying their mutual antagonism over every significant policy issue.² Americans will increasingly be forced to choose between extreme policy alternatives while resenting the lack of innovative compromises from their elected leaders.

The final-five voting approach—a powerful combination of open, top-five primaries and RCV in the general election—was pioneered by Institute for Political Innovation founder Katherine Gehl. In the past, unfortunately, states that have adopted open primaries have also eliminated the ability of candidates to petition onto general election ballots through grassroots signature drives. All candidates, no matter their affiliation, should have their access to general election ballots preserved through signature drives, even if it means more than five candidates wind up on the November ballot.

Third, put to rest the major criticisms of RCV.

Criticism number one: RCV is too complex and expensive to administer.

As more successful RCV elections take place, these criticisms will fade. Maine’s secretary of state initially predicted the cost of implementation at $11 million. Later, he testified it would cost $1.5 million. In fact, the incremental cost of the first RCV election was $79,965. The election and subsequent tabulation, by all accounts, went off without incident, despite his prediction that there would be “cars burning in the streets.”

True, the recent New York City election initially created unnecessary confusion and anxiety. Because the NYC Board of Elections literally forgot to reset its machines, it reported inaccurate preliminary election results. Once the machines were reset, the results were accurate. The election was close, but the runners-up conceded immediately and there were no lawsuits challenging the results.

It should not take days to announce an RCV winner. Indeed, there is no reason why RCV winners should take any longer to determine than FPPV winners. The high-speed scanner used to tabulate results in Maine is capable of reading three hundred paper ballots per minute. The final tabulations have been conducted before news cameras and the public within some ten seconds.

Criticism number two: RCV is too complicated for voters to understand.

In fact, people do not find ranked ballots confusing. Exit polling conducted by Edison Research in New York City showed that 95 percent of voters found the ballot simple to complete. As of this writing we are unaware of any RCV elections marred by voter confusion.

Criticism number three: RCV produces “ballot exhaustion.”

Ballot exhaustion—an evocative but misleading term—occurs when all the candidates a voter ranked have lost even though two or more other candidates remain in the race.

By that standard, any voters who choose independent candidates in traditional FPPV elections have “exhausted” their ballots because it is so rare for an independent candidate to win.

Criticism number four: RCV undermines the principle of “one man one vote.”

This argument is made by respected political scientist Harvey Mansfield. He writes,

Ranked-choice voting rewards extremism in the electorate. Voters who make extreme choices should be punished via exclusion from the majority. Ranked-choice voting rescues them from the penalty they deserve for throwing away their ballot on an extreme first choice.

Mansfield’s harsh judgment reflects a political science assumption dating back to E. E. Schattschneider in the 1930s and Clinton Rossiter in the 1960s: Coalitions of diverse interest groups are first formed within each party before conflicts are resolved by compromise between the two parties. Yet that model is the one that has created our modern dysfunction, because it assumes that there are incentives to compromise. In fact, such incentives have nearly vanished.

Mansfield expects that a voter’s first choice in an RCV election will be “extreme.” But what if the first choice is an independent who is moderate? Without RCV, the candidate would likely not have considered running, fearing the label of “spoiler.” Should a voter be “punished” for ranking this candidate first?

In fact, two voters who have the same preferences on two issues may vote for two different candidates—if there are only two—because the voters might disagree in the intensity of their preferences. With RCV, a fiscal conservative may be a voter’s first choice, while a socially progressive Democrat may be the second choice (or the reverse). This voter does not fit a party paradigm. Should the voter be punished? Or is the voter providing the winner with useful information about how to govern?

Criticism number five: RCV has been designed by Democrats to make Republicans less competitive.

Mansfield writes the following, “One suspects that progressives like ranked-choice voting because it would allow them to vote twice: once for Bernie Sanders and once for Joe Biden.”

True, in recent years more Democrats than Republicans have blessed RCV. But under RCV, it is also Republicans who would not need to fear a “Ralph Nader equivalent” on the right. While such a candidate could increase turnout among the most conservative voters, these same voters would likely make a moderate Republican their second choice rather than a Democrat. There is no reason to doubt that, over time, advantages and disadvantages would accrue equally to Democrats and Republicans.

Fourth, develop “high touch” political organizations suited to each state.

The Maine and Massachusetts RCV campaigns were blessed with such organizations; thousands of enthusiastic volunteers were backed by campaign professionals and attorneys with abundant know-how. This type of mobilization of volunteers is not easy, but it is the factor most critical to success.

Some observers recommend using polls to gauge a state’s “proclivity” to adopt RCV. We believe this strategy is less reliable than choosing specific people for their leadership, organizational skills, and motivation. These qualities create proclivity, no matter what the state. The national movement for RCV will require significant resources but, more important, many talented Americans who will make RCV their most important work.

Fifth, dedicate resources to a sustained national education effort.

A successful national RCV strategy will require the intense messaging that will lead majorities of the American public to see RCV as a principal antidote to today’s dysfunctional political system. The RCV message, “more voice and more choice,” and the promise—the certainty—of majority, as opposed to plurality, winners succeeded in Maine. At the national level, the superiority of RCV over FPPV must become widely accepted. We know this is possible, since current RCV supporters represent every conceivable segment of the political and ideological spectrum.

In a recent study, the pollster Douglas Schoen, a longtime adviser to our RCV efforts, identified the messages and themes with the most impact in moving independents toward RCV. Schoen’s findings are counterintuitive: It turns out that negative messaging about FPPV moves public opinion more dramatically in the “right” direction than positive messaging about RCV does. Positive messaging about RCV produces a net gain of 21 percent among independents, but negative messaging about FPPV produces a net loss for FPPV of 42 percent. These data confirm the importance of a national education campaign that identifies FPPV as the fundamental cause of political dysfunction in America, which only RCV can successfully address.

Recurrent messaging is needed, especially to show that the status quo is unacceptable. Elections based on FPPV should become generally viewed as being as harmful to our political health as cigarettes are to our physical health. Further, the public needs to understand the way RCV removes the “spoiler” accusation leveled by every Democrat or Republican who runs against an independent. They need to see that without this burden, Americans without a major party affiliation can successfully compete for public office. When this happens, major party candidates will either seek support outside their base or lose to independents.

Voters are now unaware of RCV or ambivalent about it. In the poll showing that nearly 70 percent of voters want more unaffiliated candidates, only 50 percent wanted to rank their votes. Too many still see RCV as a gimmick rather than a solution to the political dysfunction they abhor.

Developing a national electorate informed about the virtues of RCV will be critical. RCV needs to evoke the same types of public passions as other great social movements in our country’s history. For a movement to succeed, its members need shared aspirations. They also need common enemies—which in this case must be FPPV.

The good news is that RCV is increasingly recognized as the “mother of all reforms.” Its success may lead to the adoption of other voting reforms, like those related to gerrymandering and campaign finance. As with other historic social movements, unity among those most committed to RCV may well produce increased resources that, in turn, will lead to greater support by millions of volunteers and voters.

Peter Ackerman is chair of the Chamberlain Project Foundation. In 2010 he founded Americans Elect to create a nonpartisan pathway to the presidency. He currently serves as co-chair of the international advisory council of the U.S. Institute of Peace and on the executive committee of the Atlantic Council.

Cara Brown McCormick is co-founder of the Committee for Ranked Choice Voting and the Chamberlain Project Foundation. She has worked professionally on American political campaigns in every cycle for the past thirty years.

¹For the best analysis of the pernicious effects of a lack of competition in our political system, please read The Politics Industry by Katherine M. Gehl and Michael E. Porter.

²In the 1994 study The minimal group paradigm: Categorization into two versus three groups, Hartstone and Augoustinos suggest why the level of antagonism between politicians in the two major parties is so high:

Comparison of the results with three minimal groups with those of a baseline two-group experiment shows that with a three-group structure there is no significant ingroup bias…A two-group context may be particularly effective in evoking an ‘us versus them’ contrast. Self-categorization as a group member is more likely to occur in the presence of two groups whereas three minimal groups renders an ‘us-them’ contrastive orientation less salient.