Just six years after adopting ranked-choice voting for primaries and special elections, New York City may be headed for another round of electoral reform—this time sparked by a tumultuous mayoral race.

The fragmented Democratic primary means tomorrow’s winner will likely be determined by how voters rank candidates. If the result fractures the party, the general election could be similarly splintered—this time under a single-vote system. That outcome could prompt another push for reform, timed to coincide with the forthcoming revision of the city’s charter.

Finally, a reason to check your email.

Sign up for our free newsletter today.

In November 2019, New York City voters adopted RCV for closed party primaries. The goal was to select the nominee who best unifies a party’s primary electorate. Democratic voters first used the system in 2021 to choose Eric Adams, who went on to win the general election.

Two major research findings suggest that ranked-choice voting does not live up to advocates’ promises. First, RCV often fails to produce a winner who earned a majority of all votes cast. Two political scientists warned of this possibility in a 2014 scholarly article on “ballot exhaustion.” This occurs when voters truncate their rankings—leaving some choices blank—or rank a frontrunner below a candidate eliminated early in the count. When enough ballots are exhausted in this way, the eventual winner may secure a majority of remaining ballots, but not a majority of total ballots cast. The most comprehensive study to date finds that 97 of 185 U.S. RCV elections from 2004 to 2022 suffered from this kind of “majoritarian failure.”

The second key finding is that the average voter is not inclined to rank many choices. A 2009 book, How Voters Decide, explains that it takes more cognitive effort to judge candidates in relation to each other (that is, to rank or score them) than just to choose among them. This has been widely cited in studies of voter error and confusion. A recent extension of the theory asked whether voters actively seek out the information needed to rank candidates—and across a range of scenarios, most do not. These scenarios plausibly cover both New York City’s version of RCV (in closed primaries) and Alaska’s version (without party primaries).

Recent polling in the current mayoral race suggests both problems could resurface. Two early June polls show former governor Andrew Cuomo beating socialist assemblyman Zohran Mamdani by roughly ten points in the final round of counting—one putting the margin at 54.4 percent to 45.6 percent in round ten, the other at 56 percent to 44 percent in round eight. A third poll, from Data For Progress (DFP), showed a tighter race, with Cuomo’s round-eight lead falling within the margin of sampling error.

Of the three, only DFP reported how many ballots might drop out during the count and how exhaustion could vary across social groups. While DFP’s sample is the least representative of the city overall, it implied an 8.6 percent exhaustion rate. By group: Asian voters were least likely to exhaust their ballots (0 percent), while black voters were most likely (9.5 percent). Among non–college-educated voters, 11.1 percent exhausted their ballots, compared with 5.6 percent of college-educated voters. Manhattan voters were most prone to exhaustion (12.5 percent), Queens voters the least (5.3 percent).

These results shouldn’t be over-interpreted. About 5 percent of respondents were still undecided on their first choice—a figure likely to shift as early voting continues. And the survey excluded Staten Island, whose demographics differ markedly from the rest of the city.

Yet, the numbers are good enough to underscore the problems with an RCV election. DFP’s implied ballot-exhaustion rate approaches or exceeds the expected final-round margin in all three polls. That could leave Democrats with a nominee whom a majority of voters might not have wanted, even if they could not unite behind a different candidate.

Meantime, some wonder if the Democratic Party could split in advance of the general election. Should that happen, New York City could face its most fragmented general election since 1969. Mayor Eric Adams is running as an independent. Cuomo says he would do the same, regardless of whether he wins the Democratic primary. Two more candidates are Republican Curtis Sliwa and attorney Jim Walden, yet another independent.

Such a situation might build support for the current proposal from the city’s charter revision commission, which would overhaul the current system of mayoral elections. The new rules would retain RCV but only for the qualifying round. All candidates would run against each other in an RCV round one—an “open” primary, to use the term loosely. The result of that first round would then set up November’s race, in which the top two candidates would face off. Control of the party label remains to be determined.

The allure of this “top two” approach is that it might give important players—candidates, party leaders, and interest groups—a reason to reduce fragmentation. One could argue they’d face new pressure to discourage marginal candidates from entering the race. They might also feel compelled to polarize the field earlier in the count, ironically through negative campaigning. But given how voters actually use RCV, it would likely take both strategies—and even then, a final contest between two broadly supported candidates is far from guaranteed.

Political fragmentation is a hard problem to solve. So far, the response has been to outsource it to voters’ rankings—whether through an RCV primary or an Alaska-style “final four” election. This time, New York City’s experience with RCV could lead to a system that forces an actual runoff. If so, the experiment in “instant runoff” may end up being the teacher.

Jack Santucci is a professorial lecturer at the George Washington University and an adjunct fellow at the Manhattan Institute.

Photo by Michael M. Santiago/Getty Images

Donate

City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).