Tuesday may be the last traditional election day of my life in Washington, DC, where I’ve been voting for the past 12 years.
Ballot included Initiative 83a measure of adoption of ranked choice voting (RCV); It passed overwhelmingly. Although it is possible that the DC government could refuse to implement the measure (they did Do it first), from now on I’ll be ranking candidates for DC Council and Mayor — not just voting for one candidate per position.
This story first appeared in the Future Perfect Newsletter.
Sign up here to explore the big, complex problems facing the world and the most effective ways to solve them Sent twice a week.
Ranked Choice is an electoral reform that seemed like a pipe dream a few years ago, but has become mainstream over the past decade or so. Alaska, hawaiiAnd MaineCongress adopted it for some elections to statewide offices. Although few are like municipalities San Francisco And Minneapolis Having used it for decades, they recently joined New York City, SeattleAnd Portland, Oregon. Besides DC, states Colorado, Idaho, NevadaAnd Oregon All voted on adopting the system on Tuesday, and Alaska Vote on whether to keep it.
Full disclosure: I voted yes on the DC initiative. I think it probably does more good than harm in the context of our city. First-past-the-post voting is clearly deeply flawed, which is why many places have jumped on the RCV bandwagon. But I also think the benefits of RCV have been oversold and we should examine other ways to make selection more proportionate.
Ranked preference voting, explained
In ranked preference voting (also called “instant runoffs”), voters rank the candidates in order. All first preference ballots are counted. If a candidate does not have a majority of first-preference votes, the candidate with the smallest share is eliminated; Their votes are then redistributed based on second place to their supporters. This continues until no candidate has a majority.
I first encountered this idea after the 2000 election. in Florida, 97,488 voted for Ralph Nader; Only 537 of those would be needed to win the state for Al Gore and thus the presidency. What if those Nader voters – who are overwhelmingly liberal – were able to give Gore second place? Then it would have happened naturally, and the failure to mobilize left-of-center voters would not have led to the presidency of George W. Bush, the war in Iraq, etc.
This argument is also why I support the idea in DC. Here, as in many coastal cities, almost all political contests occur in Democratic primaries, which are often incredibly crowded. Every four years, the well-to-do people here try to vacate the seats Anita BondsOur notoriously invalid And ineligible At-large city councilors, and each time, multiple contenders split the anti-bond vote. Two years agoHe won redistricting with 36 percent of the vote, while the two challengers each received 28 percent. RCV will divide the opposition and make it harder for unpopular incumbents to get re-nominated.
As a narrow tool to avoid the spoiler effect, RCV works quite well. But its supporters also have big ambitions.
Kathryn Gehl, a wealthy former CEO who has bankrolled many recent RCV initiatives, argues that her particular version (called “Final Five” voting) would almost single-handedly force politicians to work together again. Gehl wrote two years ago:
Cooperation is hindered. Senators and Representatives are free from negative bias restrictions. They are free to reach across the aisle, solve complex problems through innovation and negotiation.
The theory is elegant. In Final five votesAll candidates – regardless of party – participate in a primary The top five finishers are then placed on the general election ballot, where voters can rank them.
The hope is that it eliminates the dynamic where partisan primaries push party nominees to ideological extremes, and where fear of such primaries prevents incumbents from compromising or defying their party (see 10 House Republicans Who Voted to Impeach Trump, Four of them lost nominations when challenged by pro-Trump Republicans). Then, ranked-choice voting in general elections means candidates compete for the No. 2 and No. 3 votes, reducing negative campaign incentives.
Case(s) against RCV
Sounds great! So why would anyone oppose RCV?
One possible reason is political scientist Nolan McCarty’s finding that under RCV, Neighborhoods with more ethnic minorities tend to experience more “ballot fatigue.” (Many candidates fail to rank as an affiliate). This means, McCarthy argues, that Reform tends to “reduce the electoral effect of ethnic and racial minority communities.”
Work by Lindsay Cormack, associate professor at the Stevens Institute of Technology, similarly found that “overvoting” (using the same ranking multiple times, meaning ballots cannot be counted accurately) is more common in minority communities, while the University of Pennsylvania Stephen Pettigrew and Dylan Radley Ballot errors are generally higher in ranked choice than in traditional elections.
Anything that raises the specter of diminishing electoral influence for minority communities in the United States is worrisome. That said, I’m not sure this case is meritless either. Ranked choice is a significant change that takes time for a selector to understand and adjust to. I am not convinced that the higher error rates for the newly adopted method of voting indicate that these error rates will continue as the practice normalizes.
A more compelling counterargument to me is that RCV does not seem to do anything to reduce bias and encourage cross-party compromise. The reason has to do with the classic case against instant-runoff voting, which you’ve probably heard if you’re a friend. Social choice theory nerds (like, alas, me).
One of the things you want a voting system to do is select the person who would win a head-to-head race against every other candidate. It is called “Condorcet wins,” and while an election doesn’t always have one, when there is One, it seems that a good electoral system should win them over, because the electorate prefers all the options.
Ranked choice voting doesn’t always pick a Condorcet winner, and we’ve now seen multiple real-world elections where the Condorcet winner (which you can deduce from ranked-choice ballot records) has lost. In Alaska’s 2022 US House special electionWhich used ranked choice, Condorcet’s winner was Republican Nick Begich, but Democrat Mary Peltola won. The same thing happened 2009 Burlington, Vermont, mayoral race.
Importantly, in both cases the Condorcet winner was the most moderate of the three major candidates. Begich was to the right of Peltola, but to the left of third candidate Sarah Palin (!). In Burlington, the left-wing Progressive Party nominee defeated both the Democratic and Republican nominees, although the Democrat (a centrist in Burlington terms) Condorcet was victorious.
RCV advocates note that these two cases and this one are among thousands of RCV selections In practice, Condorcet failure is rare. I’m not so sure about that.
Research from Nathan Atkinson, Edward Foley and Scott Ganz used a nationally ranked preference survey of American voters to simulate what an election would look like under a nationwide system. For each state, they simulated 100,000 elections with four candidates. They found that in 40 percent of cases, the Condorcet winner lost, suggesting that the rarity of Condorcet failure in practice may simply be an artifact of RCV being relatively new, and that such outcomes will become more common as the method spreads.
Worse, the simulation paper found that the system results in far more extreme winners (that is, winners far from the median voter) than choosing a Condorcet winner. Indeed, “states where [the system] The states that do the worst (including Arizona, Nevada and Georgia) are the most polarized [it] The best performers (including Massachusetts, North Dakota, and Vermont) are among the least polarized.” The system actually appears to encourage polarization, not avoid it.
New American political scientist Lee Drutman was once such a big fan of RCV that he Wrote a book It is called for, but in recent years it has been considered Rarely is the remedy for polarization and dysfunction He looked at it once, in part because of findings like Atkinson, et al. A better solution, he argues, is Strengthening parties And encourage them to form more.
States should allow “fusion voting,” where candidates can run on multiple party lines (New York already does this), and for legislatures, seats should be allocated proportionally: if there are 100 seats, and Democrats and Republicans each get 45 percent of the vote and The Greens and Libertarians each get 5, then they would need 45, 45, 5, and 5 seats respectively.
This is a far more radical change than ranked preference voting, and requires a real rethink by politicians. It’s hard to imagine a DC with multiple viable political parties, or one where no one important is a Democrat. But it’s worth trying and testing. We have learned a lot using RCV and we can learn more.
