Moneybox

Maine Just Voted for a Better Way to Vote

And if we’d had it nationwide, it might have won Hillary Clinton the presidency.

Turnout was strong for the final day of absentee voting at City Hall on Nov. 3.
For Maine, it’s a shift that could make third-party voting more viable overnight.

Ben McCanna/Portland Press Herald via Getty Images

On Tuesday, Maine became the first state to challenge America’s first-past-the-post voting system, as voters approved, by a margin of 52 percent to 48 percent, a referendum instituting ranked-choice voting for state and federal elections. It’s by far the biggest victory for a reform movement that has attracted high-profile endorsements from politicians like John McCain and Howard Dean but had so far failed to gain traction beyond a few progressive American cities. For Maine, it’s a shift that could make third-party voting more viable overnight—by eliminating the ability of third-parties to play spoiler.

It’s something that should resonate this morning: Florida, Pennsylvania, Michigan, and Wisconsin, the states that flipped from Barack Obama to Donald Trump, were all decided by a margin of victory smaller than the third-party vote in those states. Pennsylvania’s margin was 70,000 votes; 220,000 third-party votes were cast there. In none of those states did Donald Trump win a majority.

This has been a problem for a while in Maine, where strong third-party showings have routinely produced winners who are unable to get a majority. Nine of the past 11 governors have been elected with less than 50 percent of the vote. Five won with less than 40 percent. Gov. Paul LePage, a Republican and ardent Trump supporter, won the 2014 election with 38 percent of the vote after independent and Democratic rivals split the remainder.

Ranked-choice voting, sometimes called instant runoff voting, has a smoothing function. Under the system just approved in Maine, voters rank five candidates from 1 to 5 on their ballots. Election officials then eliminate the candidate with the fewest first-choice votes and immediately re-assign those voters to their second-choice candidate. The process repeats until one candidate is favored over the remaining alternatives by a majority of voters. It gives candidates with lots of second-choice votes a chance. Gone is the three-way election where the winning candidate gets only 38 percent of the vote.

Here’s an example of how this type of system would play out in your daily life: Say you had to decide what beer to bring to a party. A hoppy IPA might win a standard vote with the dedicated support of the beer snobs, while other less polarizing beers split the remainder of the electorate into small factions. A ranked-choice vote, however, would eventually coalesce the votes of the IPA-hating majority around a more palatable choice. Drinkable beers to the top!

Ranked-choice voting isn’t just a political science experiment. The system is used for national elections in Australia and Ireland, as well as smaller-scale elections around the world. Among American cities, San Francisco has voted numerically since 2002, Minneapolis since 2006. A handful of other cities do it too, including Portland, Maine’s largest city.

The statewide movement for ranked-choice voting has been underway in Maine for more than a decade. It also comes up in a handful of other statehouses every year, only to be condemned to further study. But Maine was the first state in more than a decade to vote on the system. Its passage could set a national example.

Ranked-choice voting, proponents say, makes it easier to vote third party knowing you won’t split the major-party vote. They often cite, as an example of the kind of problem it would solve, the Florida presidential vote in 2000, when nearly 100,000 Floridians voted for Ralph Nader and the final margin between George W. Bush and Al Gore was just 537 votes. In a ranked-choice system, Nader might have nabbed first-place votes from Gore voters wary under the current system (and rightly so) of pushing the state to Bush. At the same time, Naderites’ second-place votes for Gore might have ultimately pushed the vice president over the top.

Or, to take a more current example, Donald Trump received 13.3 million votes in the Republican primaries—about 2 million votes fewer than Ted Cruz, John Kasich, and Marco Rubio, who scrapped for the anti-Trump vote before a final, belated, and misguided compromise. With his sky-high unfavorables, Trump would have likely fared much worse in a ranked-choice system, especially during the early primaries.

Other states impose runoffs if candidates don’t meet a certain vote threshold. But those secondary elections tend to be both expensive to hold and poorly attended. In several states that require low-bar pluralities, instant runoff voting is already in use for military and overseas voters, who fill out primary and runoff ballots at once.

There are theoretical downsides to the system. The winner of a ranked-choice election might be everyone’s second-choice candidate but few people’s first.* In other contexts, that produces results that tend to reward a kind of bland centrism. Oscar voters use ranked-choice voting to pick the best picture, which handicaps films with broad, middling appeal. Maine voters may have just doomed themselves to a future of The King’s Speech–grade politicians. (In this allegory, Paul LePage is a raunchy grindhouse flick beloved by some but baffling and offensive to everyone else.)

There are practical concerns, too, with making America’s low-turnout elections more complicated. Jason McDaniel, an associate professor of political science at San Francisco State University, analyzed 16 years of election data in the city before and after ranked-choice came into effect. “The results show that participation decreased among specific groups of voters after the adoption of ranked-choice voting, especially younger voters, African-Americans and those with low levels of education,” McDaniel wrote in the Bangor Daily News. “My conclusion is that, in San Francisco, ranked-choice voting widens the gulf between the haves and the have-nots when it comes to participating in the democratic process.” In McDaniel’s telling, the complexity exacerbates existing voter-turnout discrepancies. It also raises the standard for what constitutes a complete ballot, resulting in lots of disqualifying ballot errors. In one election, they outnumbered the margin of victory.

Rob Richie, the director of Fair Vote, a Maryland nonprofit that advocates for better voting systems, says McDaniel is using a very small sample size—1 million votes over eight years—to make his point. Better evidence, he says, comes from University of Missouri professor David Kimball, who compared same-day votes of ranked-choice cities with peer cities using pluralities. Kimball concludes that ranked-choice voting’s only effect on turnout is to reduce the big drop in voter participation that occurs between primaries and runoffs.

One problem with analyzing ranked-choice voting in practice, Kimball says, is that the sample size remains very small. There are scarcely more than a dozen cities using ranked-choice voting in the U.S. With Maine’s passage of the system, however, researchers like Kimball will have a fair bit more data to work with.

Given its virtues, why hasn’t ranked-choice voting been tried in more places? For a long time, one challenge was that it was considerably harder to count ranked ballots. Now we have computers for that. Meanwhile, though, advocates’ frequent invocation of the Bush-Gore-Nader split has left ranked-choice voting with a bit of a branding problem: Some Republicans see it as a Democratic initiative.

In Alaska, it was the state GOP that pushed to get ranked choice on the ballot in 2002 after a long run of defeats in governors’ races. John McCain even recorded a robocall in favor of the measure. It failed, in part due to the sense among GOP voters, says Richie, that it would favor Democrats.

That’s not necessarily true, of course. In 2008, Vermont’s Democratic legislature passed ranked-choice voting only to see it vetoed by the state’s Republican governor. Ironically, six years later, Democratic Gov. Peter Shumlin won the office with just 46 percent of the vote. A Libertarian candidate took more than 4 percent.

A bigger problem, Richie suggests, is that state lawmakers tend to be risk-averse and are favorably disposed toward the system that got them into office. “My sense of having been involved in the cities is that there’s a first-in-the-nation kind of breakthrough that’s important,” he says. “San Francisco’s passage paved the way for the cities; Maine passing it would pave the way for a different kind of conversation in the states.”

*Correction, Nov. 9, 2016: This article originally misstated that a candidate with no first-place votes could win a ranked-choice election. If you have no first-place votes, you cannot win a ranked-choice election. (Return.)