What Alaska’s First Use of Ranked-Choice Voting Has to Teach the Rest of Us About How We Vote
A Reminder That Representative Democracy is About Representing the Majority
Welcome to my newsletter: Foreword to Civilization. You can subscribe to get my writing direct to your inbox, and there’s never a paywall thanks to my monthly supporters on Patreon.
Something big recently happened in Alaska that reveals a way forward for the rest of the country. And to be clear, it’s not the outcome itself that’s important. It should not be viewed through a partisan lens where all that matters is who won. What matters is the details of the outcome and how it came to be. What happened was that Alaska used ranked-choice voting for the first time to select a member of Congress, and the result was that a “red” state chose a Democrat to represent them. Let’s dig into the details and what those details have to teach us about the way we run elections and the subsequent impacts on both politicians and voters alike.
First, what is ranked-choice voting? Ranked-choice voting is a way of determining a majority winner of an election through a series of instant-runoffs made possible by ranking candidates in order of preference. If a candidate gets more than 50% in the first round, that candidate wins. If no candidate has more than 50% in the first round, then the candidate with the least votes is removed from the race and their votes are redistributed to the remaining candidates based on voter preference. This process of removing candidates from the race and redistributing their votes to the remaining candidates in successive runoffs continues until a candidate surpasses 50%.
The key point of ranked-choice voting is finding the candidate that a full majority of voters prefer. In standard elections, the most common way of crowning a winner is simply who gets the most votes, which tends to involve just winning a plurality of the vote, not a majority of the vote. That means someone who only 30% of voters voted for can end up winning despite 70% of the people that person will represent, not wanting that candidate to represent them. Because this tends to also be combined with closed primaries where only a small percentage of voters participate, and gerrymandered districts where all that really matters is winning the primary, this can mean representatives winning with extremely low degrees of support. Suffice to say, plurality voting leads to far from democratic outcomes, where the majority of voters often actively dislike their own representatives, and can even conclude there’s no point in voting.
Alaska recognized all of this and made some changes recently. Instead of utilizing standard primaries where Republicans choose a Republican, and Democrats choose a Democrat (and independents get to help pay for the primaries but not vote in them), Alaska chose to utilize a final-four nonpartisan primary system where the top four vote-getters on a single shared ballot listing all candidates from all parties move on to the general election. In this first case, in a special election to fill a spot in Congress opened by the passing of former Representative Don Young, that meant two Republicans, one Democrat, and one independent moving on to the general election where the winner would be determined by the ranking of candidates to determine the majority winner.
Here’s what happened: In the first round of voting, after waiting 15 days for all overseas votes to arrive, the Democrat (Mary Peltola) was in the lead with 40% of the vote, followed by a Republican (Sarah Palin) with 31% of the vote, and another Republican (Nick Begich) with 29% of the vote. There was no fourth candidate like there should have been because the independent (Al Gross) dropped out after the primary and encouraged his voters to vote for Peltola instead. With only three people left, this meant there would only be a single redistribution of votes from Begich to Peltola and Palin. What no one knew was just how many Begich voters would rank Peltola over Palin.
Although half of Begich’s voters ranked Palin as their second choice, three out of ten of them ranked Peltola as their second choice, and two out of ten of them decided not to rank Palin at all. As a result, after a 30-second tabulation of second choice votes, Peltola won with 51.5% of the vote to Palin’s 48.5% of the vote. That was the outcome. For the first time in five decades, Alaska has sent a Democrat to Congress, who also happens to be the first Native Alaskan to go to Congress. It’s all pretty historic.
Again, though, it’s not the outcome of a Democrat winning that is what everyone should be focusing on. It’s how and why Peltola won. Here’s how and why she won:
The key to winning a ranked-choice election is to have both strong support and wide support. The more people rank you as their first choice, the better, and the more people who rank you as their second choice, the better. In a typical plurality election with no runoff, because only strong support is needed, that means an effective strategy is going negative on opponents. If you can demonize your opponents, it makes you look better and can result in you getting a plurality of the vote. But gaining a majority of the vote means reaching out to all those voters who don’t actually most prefer you, but could be okay with you winning. That means an effective strategy is showing a willingness to collaborate with others, and going negative becomes an ineffective strategy.
Sarah Palin went negative, as did Begich. Instead of encouraging Begich voters to rank her second, she attacked Begich and ranked-choice voting itself, while Begich attacked Palin in return. Instead of looking at the election as two Republicans against a Democrat where either Republican is better than the Democrat, both Republicans went after each other as they would in a typical primary election instead of a ranked choice general election. In contrast, Peltola showed a willingness to cross the aisle, expressing support for policies like more oil and gas exploration that typical Democrats don’t express. She also focused on issues like Alaskan fisheries and ran a mostly positive campaign that didn’t demonize her opponents.
The difference in electoral strategy is a key reason that so many Begich voters decided to rank Peltola as their second choice or not rank Palin at all and effectively choose “none of the above” instead of Palin who 60% of Alaska’s voters had a negative view of according to a poll in July. It’s extremely difficult to win a ranked-choice election, or any election for that matter, when 60% of voters actively dislike you.
One argument being used against these results, is that 60% of the first round voters were for Republican candidates and therefore a Republican obviously should have won, but just because more people most preferred someone from the Republican party, doesn’t mean they most preferred anyone from the Republican party, and would have voted that way in a standard election. There’s even an argument to be made that Peltola may have won by an even larger margin in a standard election.
In a standard closed primary, Palin and Begich would have been tearing each other down in order to claw their way to the plurality win. With all the negative things said about both candidates - which would have likely been even more numerous - had Palin gotten the primary win, she would have been bloodied by it. Then with an election between Palin and Peltola as the two main-party candidates, Palin may have found even less support among former Begich voters. Many of them would have stayed home rather than vote for Palin, and many would have chosen Peltola instead. Again, polls showed 60% of polled Alaskans actively disliked Palin, and that was without the impact that a closed primary would have had on all voters that very well could have driven that number even higher.
Additionally, less than 25% of Alaska voters are registered as Republican. 60% of Alaskans consider themselves to not be Democrat or Republican, so to assume that 60% of Alaskans prefer any Republican, is just not accurate. Alaskans are perfectly willing to vote for the person instead of the party, and in this case, they demonstrated that by reaching a majority consensus upon Peltola as their candidate.
A point being raised by democracy reform advocates is that Begich should have won despite having the weakest first choice support, based on the belief that more people may have preferred Begich instead of Peltola if Palin hadn’t run, which is kind of an alternative universe argument to make, and I personally find it strange to say that a more popular candidate spoiled an election for a less popular candidate. However, for those who believe Begich should have won based on being the candidate with potentially the widest support, I’d like to challenge that assumption based on the political realities of the here and now.
We now live in a post-Roe America where abortion rights have become a much more salient issue for many voters, especially women. We also recently experienced another horrific mass shooting at a school that made gun control more salient, and we live in a time where a litmus test is whether a candidate believes Joe Biden legitimately won the 2020 election. These three issues have become of greater importance to Democrats, independents, and even voters who traditionally vote Republican.
According to a recent survey by Democracy Corps of battleground states, failing to raise the age for assault weapons, along with making abortion illegal, are two of the biggest things currently animating voters. For some key demographics, restoring the enhanced child tax credit is right up there too. With this in mind, with such strong feelings on these issues, gaining majority support is aided by a moderate stance. A moderate stance on abortion is legal abortion with some restrictions, and a moderate stance on gun control is legal guns with some restrictions. Senator Murkowski is a great example of having a moderate stance on these issues. She has proposed legislation to codify Roe v. Wade, and she was one of the 15 senators who voted for recent bipartisan gun legislation. In contrast, Begich is on record as saying he supports the Supreme Court’s decision to reverse Roe v. Wade, and that he does not support gun control legislation. He has also declined to answer if Biden legitimately won in 2020. Peltola however holds the opposite view on these issues. At this time, her views better reflect the majority of Alaska’s voters on the issues most salient to them.
I would argue that in the present political environment, it simply isn’t probable to gain majority support in a heavily independent state like Alaska holding the stances both Palin and Begich hold, instead of the views Murkowski and Peltola hold. It’s possible that if Begich moderated his stances on these issues before November, or if these issues become less salient, then he could gain more support, but as long as these issues remain as salient as they are, winning a majority of the vote is an extremely uphill battle for any candidate who actively disagrees with the majority of voters on the issues most important to them at the time of voting. This is why I think requiring majority support with ranked-choice voting is so important to a functioning democracy. Politicians need to listen to voters more and represent their views better.
As with any voting method, there is a degree of personal preference as to what qualifies as the best kind of outcome. With ranked-choice voting, “best” is determined by strong first choice support combined with strong wide support. With a method like approval voting, where voters are asked to simply approve of everyone they like with a check mark, the preference is for the widest support possible but not strong support. In this way, it’s possible for someone to win who no voter likes, but no voter hates. This could lead to very position-averse candidates who just try to avoid being unappealing to anyone. That may be exactly what some voters would prefer instead. It really is up to each person’s preference to determine what kind of voting system they think would be best. Personally, I prefer that strong support be part of the mix.
One question you may have now is just how happy Alaskans are with the new voting system they did choose. Well first, turnout was high for a special election. Also 85% of Alaskans reported finding ranked-choice voting to be simple, and 95% of Alaskans have said they received instructions on how to go about it. Additionally, only 0.2% of ballots were determined as invalid. These data points help counter talking points about how Begich voters who didn’t rank Palin didn’t understand what they were doing. It seems likely they knew exactly what they were doing.
We’ll see what happens in November when the election is essentially replayed. Will the Republican candidates learn from their mistakes and run a coalition campaign of “rank the red” where they encourage each other’s supporters to rank the other second? Will they moderate their views at all? Anything is possible, although so far Palin appears to be demanding that Begich drop out, and Begich is refusing that demand and saying Palin is a terrible candidate. Meanwhile, Peltola will have a couple months in Congress under her belt, which could help her or hurt her.
The thing I think that many people don’t immediately understand about ranked-choice voting but need to, is that it’s really about changing incentives, and that can mean all the difference between a highly polarized Congress that refuses to work together on anything, and a Congress that feels crossing the aisle would actually help them win their elections. If a Republican is more able to win if Democrats rank them, and vice versa, then that can make all the difference in the world for actually passing legislation that the majority of voters actually want. More legislation can actually be passed in a bipartisan fashion when more politicians want to appeal to voters outside their base.
As long as states like Alaska and Maine are the only ones creating an incentive structure for bipartisanship (and possibly Nevada if they become the third state in November to adopt ranked-choice voting) then most of the country will continue selecting candidates based on who has the strongest support to win a plurality election, where attacking everyone wins, where crossing the aisle is considered traitorous, and where the majority of voters actually dislike their representative.
Ranked-choice voting is about consensus building and actually getting things done. If you’re a voter who is into that kind of thing, well then you may want to look into your state following Alaska’s lead on ranked-choice voting and nonpartisan primaries. And if you’re into ending gerrymandering too, like 9 out of 10 voters are, look into combining both with multi-member districts for proportional representation. Now that’s how to vote.
Thank you for reading this. If you’d like to support my work, besides sharing it with others online, you can also become a monthly supporter on Patreon.
"A point being raised by democracy reform advocates is that Begich should have won despite having the weakest first choice support, based on the belief that more people may have preferred Begich instead of Peltola if Palin hadn’t run, which is kind of an alternative universe argument to make, and I personally find it strange to say that a more popular candidate spoiled an election for a less popular candidate."
A) It's not an alternative universe argument to make since we actually know the preference rankings and can determine if Begich was preferred by more voters to both Peltola and Palin. Based on the cast vote records from the AK SOS, Begich beats Peltola by 5% and beats Palin by 23% https://twitter.com/cinyc9/status/1568251916810915841
B) Is Palin the "more popular candidate" compared to Begich if Begich was preferred over Palin by a super-majority of voters (23% margin)? I find it hilarious that RCV supporters fall into the same trap as plurality supporters by prioritizing 1st place votes as indicators of popularity or worthy of more weight when any ranking scheme implicitly values all ranking orders as equal because true majority support is built on compromises from voters' preferred candidates. In the 2017 Minneapolis mayoral race, at a minimum Jacob Frey's victory was built on half of his votes being votes from voters ranking him 2-5 (his 57% victory was only 44% composed of 1st place rankings); he wasn't even elected until they got to the 5th round. Whether it's RCV or Condorcet tabulation, no matter how low the preference rankings are on a voter's ballot, it's all the same in the final round/count. So calling Palin "more popular" than Begich is falling into the same mindset that keeps us in the plurality mindset. BTW, the phenomenon of a "more popular candidate" spoiling an election for a "less popular candidate" is called "center squeeze" and is a well documented critique of RCV going back all the way to when it was called IRV.
"The key point of ranked-choice voting is finding the candidate that a full majority of voters prefer."
Obviously this isn't true because as noted above, the AK RCV election only told us one majority: the one where Peltola beat Palin. It ignores that Begich handily beat Palin (which confirms that Palin was the overall loser) but also ignored that Begich was majority preferred over Peltola (by even a larger margin than Peltola over Palin). This is a piece of sophism by RCV supporters; they can claim it because RCV forces through a candidate who gets 50%+ of the vote among a certain set of candidates, but it does so at the expense of at least one head to head matchup: the RCV winner versus the Condorcet winner (when they differ). Basically, the Condorcet winner beats the RCV winner in a head to head contest based on voter preferences, getting more than 50%+ of the votes, but since they don't win following the strict elimination rules of RCV, the Condorcet winner doesn't get elected, despite actually being the majority winner.
"However, for those who believe Begich should have won based on being the candidate with potentially the widest support, I’d like to challenge that assumption based on the political realities of the here and now."
Here you basically spend 400+ words saying that Begich shouldn't have a wider base of support than Peltola based on "political realities of the here and now", yet the election was held with voters living in the "political realities of the here and now" and more voters preferred Begich over Peltola based on actual votes cast. Look, I'm a Democrat and would prefer to live in a world where more people preferred the Democrat to all Republicans in all races, but adherence to democratic principles binds me to say that Begich is the winner based on democratic principles. Maybe one day a Democratic candidate will be the Condorcet candidate in Alaska, but it wasn't this election.
As you noted, proportional representation is the gold standard that we need to strive for, but defending RCV in the face of its obvious failure in electing the correct majority candidate in this race isn't standing up for democratic principles. Bad faith criticisms of RCV from Palin or a Tom Cotton type should be countered, but in this case, it was correct that a Republican should have been elected.