Support The Bulwark and subscribe today.
  Join Now

How ‘Final Five Voting’ Can Help Save Democracy

By eliminating party primaries and using an instant runoff process, this reform would help reduce the scourge of negative partisanship.
March 1, 2023
How ‘Final Five Voting’ Can Help Save Democracy
(Composite / Midjourney)

The 2022 midterms were heralded as defying history, replete with “historic firsts” and “shocking results”—yet the vast majority of congressional races were essentially decided long before Election Day.

Just before the election, the Cook Political Report rated only 40 House and Senate seats as “toss-ups”—that is, as truly competitive. It was a good call. Of the other 430 races last November, only one had an unexpected result. Just four Senate seats out of 35 were forecast to be close. In the end, only 17 percent of the House races and 26 percent of Senate races were decided by a margin of less than 10 percentage points.

To put this fact another way: The winners of about four in five congressional races were determined no later than September 13, the last primary day. The general election in these cases was just an afterthought. Most seats are essentially owned by one party or the other, so the only races of consequence in those districts and states are party primaries, where hardly anyone shows up. As the democracy-reform group Unite America calculated, “In 2022, 83% of the U.S. House was elected by just 8% of Americans.”

For example, Andy Ogles, who described himself as the “most conservative mayor in Tennessee,” won the August 4 Republican primary for the state’s safe red 4th Congressional District. In a district of 850,000 people, just 21,000 were enough to all but guarantee him a seat in Congress. As expected, he won in November by 14 percentage points and became one of the Republicans voting against Rep. Kevin McCarthy for House Speaker. Rep. Alexandria Ocasio-Cortez provides a similar example on the Democratic side. One of the farthest-left members of the House, she first won her safe-blue seat in a 2018 primary with 16,891 votes out of about 30,000 cast (out of 215,000 registered Democrats) to represent a district of more than 670,000 people.

Primary electorates are not only sparse but highly partisan. Harvard’s Todd Washburn recently diagnosed the problem: “Primary elections are by far the most visible way we have turned power over to our most partisan, polarized citizens.” On both the left and the right, primary voters are motivated by what political scientists call “negative partisanship.” Alan Abramowitz and Steven Webster of Emory University observed in 2017 that “Negative partisanship explains nearly everything in American politics today. . . . American politics has become like a bitter sports rivalry, in which the parties hang together mainly out of sheer hatred of the other team, rather than a shared sense of purpose.” That’s a big problem for democracy.

We all respond to incentives. Legislators who win their seats in primaries have little reason to craft solutions to the great challenges of our time because that’s not what primary voters on either side want. They certainly don’t want to see their representatives consorting with the enemy to get things done. Instead, members of Congress who are elected in low-turnout primaries have strong incentives to bash the opposing party, promise results that will never be achieved, and put gridlock and grandstanding ahead of legislative substance. They mirror their primary voters’ negative partisanship.

The way to change these incentives is to re-enfranchise every general election voter in the country. Have them choose members of Congress and state officials in competitive general elections with a majority vote in November, rather than having tiny cohorts of primary voters make the selection in party primaries spread out from March through September. The proposal I helped develop, Final Five Voting (FFV), requires two changes:

(1) Eliminate party primaries. Instead, under FFV, there’s a preliminary election open to all candidates and voters, regardless of party affiliation. The top five finishers, not just one Republican and one Democrat, qualify for the general election.

(2) Eliminate plurality winners. Instead, FFV uses an instant runoff process to determine who among the five candidates has the support of a majority of voters.

An instant runoff is exactly like a series of runoffs, but instead of having to return to the polls for another round, as they often do in Georgia, voters cast all their votes at once using a ranked ballot. (This process is often called “ranked choice voting,” but instant runoff is a more accurate term.) After the polls close, three rounds of runoffs narrow the field of five candidates to the final two. Between those two candidates, one will always have majority support, and that candidate will be the winner.


As the proposal for Final Five Voting has gained momentum, it has also come in for criticism, much of it based on a flawed understanding of what the proposal is and how it works. Here are a few of the major misapprehensions:

Final Five Voting is the same as Ranked Choice Voting. Not at all. Final Five Voting is the combination of an open preliminary election out of which the top five finishers advance to the general election and instant runoffs in the general election. The result of this combination is, intentionally, quite different than what RCV delivers. FFV results in more competition for—and therefore more accountability to—general election voters. The objective of FFV is to change the incentives of legislators to focus on problem-solving. That won’t happen with RCV because winners in RCV-only systems still need to survive a party primary first, ensuring that their top priority would remain serving the tiny sliver of the electorate motivated by negative partisanship. Only FFV ensures that winners are chosen by a majority of the full electorate in November. Said another way, while proponents of RCV claim that it is more “fair” than existing plurality voting, even if they’re right (and others dispute this), it won’t accomplish what citizens really want and need: a Congress that gets the job done because it won’t change what winners do once they’re in office.

Journalists, commentators, and reform advocates often make the mistake of conflating RCV and FFV. But they are dramatically different—as you can see if you look at real-life examples. While Nevada and Alaska passed referendums for FFV-style systems (with the small difference that in Alaska, only four candidates, not five, compete in the general election), Maine and New York City, which instituted RCV, still have low-turnout conventional party primaries and little competition.

Final Five Voting is the same as open primaries or California’s top-two system. Since 2010, California has held single-ballot primaries open to all registered voters and candidates, with the two leading finishers, regardless of party, going on to the general election. Louisiana has a similar system. The problem with these “top two” contests is that most still send one Democrat and one Republican to the general, so, once again, elections in safe districts (or statewide, when one party outnumbers the other) are mostly still won in the low-turnout primary. These systems don’t re-enfranchise general election voters, allow for more choices, or encourage the airing of diverse views. As a result, they don’t substantively change legislators’ incentives.

Increased competition matters. In 1992, for example, Ross Perot’s presidential campaign brought the issue of fiscal responsibility to prominence. While Perot received only 19 percent of the popular vote, he changed the behavior of the winners. Neither President Bill Clinton nor House Speaker Newt Gingrich wanted to risk the rise of a third party, so they cooperated to deliver balanced budgets for the first time since 1969. Paul Begala, one of Clinton’s top advisers, conceded, “I am not sure we would have ever balanced the budget without the pressure Perot and his voters brought to the issue.”

Final Five Voting will never fly in enough places to make a difference. Not so; it takes only a few states to start improving legislative results for the entire nation. If four states (in addition to Alaska) pass FFV ballot measures in 2024, any of the 10 senators from those states could create pivotal “gangs of six” or “gangs of eight” to break the partisan stranglehold on governing. That’s because incumbents are incentivized more by their next re-election than by the election they already won, so a senator would likely begin answering to general election voters right away. As a result, the Congress sworn in in January 2025 would be in a position to help solve complex problems like immigration by reaching across the aisle, innovating, negotiating, collaborating, making tradeoffs, and cutting deals to reach workable—though less ideologically pure—compromises as it did successfully for generations.

Article I, Section 4 of the Constitution gives states wide latitude in devising voting systems, and FFV-style ballot initiatives have a two-for-two record in ballot measures: 2020 in Alaska and 2022 in Nevada. Last November, Alaska became the first state to use Final Five–style Voting for state and federal general elections.

Final Five Voting is designed to elect Democrats (or Republicans or moderates or centrists). The purpose of the system is not to elect someone from a particular party or ideological group or perspective. It is to change the incentives of legislators, whatever their ideology, by freeing them from the harmful grip of the party primaries on both sides.

In an ironic way, Final Five Voting has already unified left and right. The Democratic and Republican parties have labored long and hard to develop a system that cements their duopoly, and they seem to like things just the way they are—even though voters have rarely been more dissatisfied. In Alaska and Nevada, both parties opposed the ballot measures, with the Democrats spending in the seven figures in a failed attempt to defeat the Nevada referendum. Yet FFV has also begun to unite voters and donors from both sides in support of the measure. The ballot initiatives in Alaska and Nevada passed with bipartisan voter support, and well-known Republican and Democratic donors contributed to the FFV effort—even as they remained on opposite sides of the candidate races.

Voters using the FFV-style system in Alaska in November chose a centrist Democratic representative, Mary Peltola, a centrist Republican senator, Lisa Murkowski, and a Trump-backed Republican governor, Mike Dunleavy. What do these officials have in common? They are now liberated to work with their colleagues across party lines, cooperating to govern effectively.

FFV is too complicated. Americans are just as capable as the Irish and Australians, who have used ranked ballots for over a century. All voters need to do in the instant runoff is rank their choices from 1 to 5 (or, if they wish, they can pick fewer, even just one). How hard is that? According to a post-election poll of 800 Alaskans, 79 percent said the process was simple. Three quarters of those said it was “very simple.” Only 9 percent said it was “very difficult.”

FFV is anti-party. To the contrary. Under the current system, parties often have a hard time getting the candidates they need to meet their own strategic goals. Senate Majority Leader Mitch McConnell predicted in August that his party would be harmed by the “candidate quality” his party’s primaries were producing. He was right. McCarthy suffered through 15 ballots before his party elected him speaker because he was opposed by his own most extreme members, most of whom sit in safe seats and take their cues from primary voters. Rep. Matt Gaetz, the last holdout, won his party’s nomination in Florida’s 1st Congressional District in 2016 with just 36,000 votes, or little more than one-third of those cast—effectively securing victory in the heavily Republican district with the support of less than 10 percent of the registered voters.

Final Five Voting distorts voters’ wishes. Just the opposite. Our existing system leaves general election voters with a fait accompli in most cases, and a lesser-of-two-evils choice in many of the competitive races. Instant runoffs, by contrast, produce winners backed by a majority of voters, rather than a plurality.  After a special election in Alaska last summer, Sen. Tom Cotton fumed, “60 percent of Alaska voters voted for a Republican, but thanks to a convoluted process and ballot exhaustion—which disenfranchises voters—a Democrat ‘won.’” That’s not exactly right: In the first round, 60 percent of voters did list either of the two Republicans as their top choice, while Democrat Mary Peltola had 40 percent support—the largest plurality of any of the candidates. After Republican Nick Begich III, the smallest vote-getter, was eliminated, his 28 percent of voters had their votes automatically reallocated to their next choices in the final round, which gave Peltola the majority.

Final Five Voting takes too long to deliver a result. Critics point to Alaska, but, by law, the state allows 15 days after election day to receive ballots across an enormous territory. Once all the ballots were collected, the instant runoff was, indeed, instant. In fact, even with this law, in 2022 Alaska was the 26th state to certify its election, meaning 24 other states took longer.


FFV has a special appeal for those of us who want government to meet big challenges. I’m not talking about activism for its own sake, but practical problem-solving of the sort that parties used to do together—like the 1983 Social Security fix, the broad welfare reform of 1996, the balanced budgets produced by Clinton and Gingrich, and the earned-income tax credit, first enacted in 1975 and vastly expanded in the early 1990s through the advocacy of Republican Vice President Dan Quayle and Democrat Rep. Tom Downey.

Polarization and gridlock will continue to dominate—and many critical national challenges will remain unaddressed—until we make serious structural changes. The good news is that FFV is both achievable and powerful. It has proven its popularity at the polls. It potential is immense. As Americans become more frustrated with our system of government, it becomes more urgent that we think creatively about how to make it more representative, more responsive, and more accountable. FFV does all three.

Katherine Gehl

Katherine Gehl, former CEO of Gehl Foods, is the originator of Final Five Voting and a cofounder of the National Campaign for Final Five Voting. Twitter: @katherinegehl.