This November, while the rest of the country decides the future of American democracy, the people of the District of Columbia (who are excluded from much of that democracy) will vote on whether to make our own elections more confusing and less fair.
By Marc Fisher
But this fall’s ballot includes an initiative that would replace that time-tested method with a complex system known as ranked-choice voting (RCV), in which voters list candidates for each office in order of preference. Low finishers are eliminated, and their votes are divvied up among second choices, and so on in successive rounds until one candidate ends up with a majority.
The idea is honorable and well-meant.
“We want to make politicians in D.C. more accountable,” said the official proponent of Initiative 83, Lisa D.T. Rice, an advisory neighborhood commissioner in Ward 7. Many D.C. elected officials win their Democratic primaries (and thus, in this overwhelmingly Democratic town, the general election) with far less than majority support. “They are elected by a sliver of a sliver of the electorate,” she said, “and they govern for that small base. RCV elects more moderate candidates.”
Rice and other RCV proponents argue, reasonably, that candidates who know they must win a majority of the votes will broaden their message. A ranked-choice system, proponents say, also pushes voters toward a broader view, as they find themselves more invested in the success of elected officials whom they helped elect, even as a second or third choice.
RCV’s fans — and there are many, bolstered by adoption of the system in places such as Maine, New York City and right across the Potomac in Arlington County — make a lot of exciting promises. Ranked-choice voting, its campaign material claims, “is natural and easy to understand,” “elects more women and people of color,” “preserves Native Washingtonian voting power” and has all the power of Ozempic without any nasty side effects. Okay, maybe they don’t go quite that far. But they also don’t concede that RCV can be a recipe for electing extremists, confusing voters and empowering the people who least deserve to determine the winner — namely, the supporters of the least-popular candidate in each race.
Here’s how ranked-choice voting can go wrong:
Let’s say there are four candidates for mayor: Genya Genius, Barbara Boring, Larry Lunatic and Ed Extreme. Voters are asked to rank them — first choice, second choice and so on. On election night, the four hopefuls end up closely bunched together, with Genius in first place with just more than a quarter of the vote.
Instead of crowning Genius, who won the most votes, computers start reassigning votes. The least-popular candidate, Ed Extreme, is eliminated from the running and his votes go to his supporters’ second choices. In a sense, the candidate once known as the loser becomes a kingmaker. And because Extreme is a bit of a maniac and attracted some less-responsible voters, in most cases, their second choice is Lunatic.
Lifted by Extreme’s supporters, Lunatic jumps from third place to second; still, no candidate reaches a majority, so the computer launches a third round of counting. This time, it’s the original second-place finisher, Boring, whose votes are reallocated. Guess what: Most of Boring’s supporters can’t stand her longtime rival Genius. They made Lunatic their second choice out of spite.
Thanks to the second and third choices of the Extreme and Boring voters, Lunatic passes Genius and gets over the 50 percent hurdle, “winning” the election. This scenario — and many others imaginable — demonstrates how RCV potentially gives decisive power to supporters of unpopular candidates.
I’m not seeing a boatload of fairness to this. Ranked-choice voting is so complicated that the text of the proposed law runs to 2,326 words — much longer than this column — including 11 definitions of problems the system can create. There’s undervoting, in which voters fail to rank the candidates as instructed. There’s overvoting, when a voter picks more than one candidate without distinguishing them by rank. Some votes might wind up not counting at all, because voters’ ballots are “exhausted,” meaning that the voter ranked fewer candidates than there are rounds of counting.
In some D.C. elections, voters would be asked to go through this ranking game about a dozen times for races ranging from mayor to school board to advisory neighborhood commissioner.
The D.C. Democratic Party went to court to stop RCV from being put to the voters, and an October hearing could yet bounce the question from the ballot. The party argued that the system could “ultimately suppress the voice and influence of voters of color for decades to come.”
How’s that? The Democrats say past elections prove that the “under and over vote in predominantly Black wards (7 and 8) is significantly higher than other wards.” In other words, in at-large council elections, in which D.C. voters have long been asked to pick two candidates to fill two seats, people in Black neighborhoods have been far more likely to choose only one candidate or too many candidates.
Charles Wilson, chairman of the D.C. Democratic Party, says RCV would “shift voting power to more affluent areas like Ward 3,” the largely White area west of Rock Creek Park, because lots of “people in Wards 7 and 8” — the largely Black area mostly east of the Anacostia River — would not rank their choices, just as they often don’t vote for two candidates in those at-large contests.
I asked Wilson, who is Black, why that might be. “It could be the high illiteracy rate in Wards 7 and 8,” he said.
Rice, who is Black and lives in Ward 7, was not having this. “It’s racist and insulting to say that people east of the river can’t learn a new system because they are Black or less educated,” she told me. Party discipline, not ignorance, explains the prevalence of undervoting, she said. “People in Wards 7 and 8 have been schooled by the party to vote Democratic, so they are being intentional about voting for only one candidate.”
In some places around the country, ranked-choice voting has resulted in high participation in more affluent neighborhoods and lower participation in poorer areas. In 2021, when New York City used the system for the first time in a mayoral primary, voters in Whiter, wealthier parts of the city were more likely to rank their choices than voters in less-affluent areas where Blacks and Hispanics make up far greater portions of the electorate.
But no group has a monopoly on finding this system confusing. “One of the top reasons people don’t vote is they say they don’t know enough about the candidates,” said Trent England, co-chair of the Stop Ranked Choice Voting coalition, an Oklahoma-based group. “With RCV, you’re asking voters to pick a fourth or fifth choice. Real voters want to cast an informed vote, but they get frustrated with what this system demands.”
England agrees with RCV proponents that it would be great to elect people who win a majority of the vote. He likes traditional runoffs, in which the top two finishers face off in a second election days or weeks after the general election. That gives candidates time to reach out for broader support. (But runoff elections have notoriously low turnout, RCV advocates say.)
With RCV, voters lose trust because they get confused about “what your vote’s going to mean” after several rounds of reallocations, England said.
A pioneer in the ranked-choice movement, Rob Richie, co-founder of FairVote, says the concept is neither new nor radical. It is similar to how both Democrats and Republicans choose their party chairs, and many parents are familiar with the concept from the process of choosing their children’s public schools.
Richie, who lives in Takoma Park, the crunchy Maryland suburb that has had RCV for almost two decades, says the system “is kind of fun. People see your lawn sign for a candidate and they say, ‘I see who you’re supporting, but how about considering my candidate for second choice?’”
He says my scenario in which a loser prevails after several rounds of RCV is far-fetched because candidates are rarely so closely bunched together. And it’s not so bad to let who came in last determine the outcome: “It doesn’t mean that the people who supported them are bad people.”
Each side in this debate has studies to point to, showing that voter participation declines once the ranked-choice process is introduced, or that people of color win office more often under RCV systems. And each side says the trend is moving its way. FairVote points to RCV resulting in majority-minority city councils in Minneapolis and Salt Lake City; opponents focus on efforts such as this fall’s Alaska vote on repealing the state’s RCV law.
In the end, your view on ranked-choice voting might come down to how you define fairness.
RCV creates more buy-in among voters, Richie says, a deeper sense of belonging — and that makes people perceive the system as fairer. Also, he says, “usually the person who finishes first in the first round wins. So why do it? It incentivizes candidates to go out and engage with more people and earn those second- and third-choice votes.”
On the other side, fairness seems a simpler matter: She who gets the most votes should be the winner.
It’s okay to differ about what constitutes fairness. But in D.C. this fall, one version will prevail — decided by the voters, on the first time around.
