Will you be at next week’s YIMBYtown conference in Raleigh? If you’re attending, check out the panel I’ll be on with some of my favorite Substackers: Jeff Fong, Stephanie Nakhleh, Jenn Truman, and Jeremy Levine. Hope to see you there!
Last Tuesday, Texas Democrats swept statewide races by five to seven points, with nearly 10 million Texans voting, the most in a Texas midterm in at least 50 years. Austin voters who woke up the next morning to a blue Texas found the results at City Hall less decisive: three out of five City Council races ended without a winner. The top two candidates in each race will now face off in a runoff election on December 12th. In the seventeen runoffs Austin has conducted since 2014, voter turnout dropped by an average of nearly 60% from the November general election.1 This exercise in democracy is not free, nor is it particularly democratic.
It doesn’t have to be this way. Five years ago, 58% of Austin voters approved ranked choice voting, which would have settled all three races on Election Day—if only the state had allowed it.
City of Yes is a reader-supported publication. If you find this work valuable, please consider becoming a paid subscriber. Thank you for your support!
Why would a majority of Austinites vote to scrap a system that promises every winner a majority? Because, as Austinites have experienced, runoffs don’t deliver what elections are supposed to. A good election should be broadly representative and decisive, and it shouldn’t cost too much—not only for the city that runs it and the candidates who contest it, but for the voters who have to show up for it. Runoffs fail on all three measures.
They aren’t decisive. This year, two incumbents nearly doubled their closest rivals’ vote but fell two or three points short of a majority. In December, that lead counts for nothing. In the third undecided race, which happens to be my council district, the top two candidates each received about 34% of the vote in a six-way race. Several candidates were running in similar lanes and split each other’s votes, so the two headed to the runoff don’t necessarily reflect the viewpoints that voters preferred (though they do present two differing points of view).
They aren’t representative. Turnout routinely collapses between November and December, and it’s getting worse: the most recent runoff, in 2024, lost 75% of November’s voters. In six of those seventeen runoffs, the November leader lost. Runoff defenders argue that those who return to the polls in December are the most engaged and knowledgeable, and therefore make better choices. The inherent elitism of that perspective aside, a democratic election is one that gives the widest possible swathe of the electorate a say. A second election, held after most voters have returned their attention to daily life, rewards whoever is best at organizing and turning out supporters: unions, neighborhood associations, advocacy groups, and other political insiders. A sliver of the electorate ends up deciding representation for the entire district.
And they aren’t cheap. The 2016 runoff cost the city about $470,000 for just one district. In 2022, Travis County estimated that runoffs for mayor and three council seats would cost $866,000. On top of that, candidates, donors, and volunteers, already wiped out by November, have to do it all over again a month later. So do voters, many of whom may not know their district is in a runoff or can’t take time off. This year, early voting starts the Monday after Thanksgiving weekend, and Election Day falls on Saturday, December 12th, in the middle of holiday travel and on a UT-Austin final exam day.
Might there be a better way to settle these races in a single election, while achieving a result that is broadly representative, decisive, and cheap?
Plurality voting. In a plurality or “first past the post” voting system, whoever gets the most votes wins, majority or not. Plurality voting is the norm across America: nearly every state, including Texas, elects statewide officials that way. Rick Perry won reelection as governor in 2006 with just 39% of the vote. Yet the state requires its largest cities to elect by majority.
The main problem with plurality voting is that the winner can take office even when most voters chose someone else. It’s hard to call that representative. Under plurality, the November leader in my district would already be headed to City Council with a margin of only 70 votes. She’s the best candidate in the race, and I’d be delighted to see her win, but she’d take office without a majority behind her. Further, in crowded fields, similar candidates often split votes—the so-called spoiler effect—so the platform most voters support can lose to a candidate who represents a minority view (imagine two candidates who support more housing each get 30% of the vote, but the candidate who doesn’t wins with only 40%). The fear of spoiling a race also pushes candidates to drop out before voters have a say. Plurality is decisive and cheap, but it is not representative.
Approval voting. In an approval voting system, voters select every candidate they find acceptable, and the candidate with the most approvals wins. St. Louis is now the only major U.S. city that uses approval voting, and it uses it oddly. Its nonpartisan primary is an approval vote, but instead of electing the most-approved candidate, the city sends the top two to a second election a month later. That throws away approval voting’s chief advantage, which is settling the race in one round, and keeps all the costs of a runoff. Last year, the second round did not add much: in the March 2025 mayoral primary, a four-way race, Cara Spencer was approved on 68% of ballots and incumbent Tishaura Jones on 33%; Spencer went on to win in April. During the primary, voters approved only 1.4 candidates per ballot, so even the primary behaved much like plurality.
Approval voting, used sensibly, lets voters back several candidates, so vote-splitting mostly disappears, and the winner is usually the candidate with the broadest appeal. But voters can’t rank preferences, and the advantage shrinks when, as often happens, voters approve just one candidate. Perhaps the main drawback, though, is that approval voting does not guarantee a majority. The winner is simply the most-approved candidate, even if fewer than half of voters approved them.
Ranked choice voting. Ranked choice or “instant runoff” voting allows voters to express relative preferences by ranking candidates in order. If no candidate receives a majority of the first choices, the last-place candidate is eliminated and those ballots count toward their next choice. This goes on for round after round until someone has a majority of the ballots still active. RCV requires only one election, it reduces spoilers, and it rewards candidates who appeal more broadly to the electorate or supporters of other candidates. It offers a way to meet all three tests while letting voters distinguish their first choice from their fallback options.
It does have some drawbacks. Ballots that rank only eliminated candidates exhaust, so the final “majority” can be less than half of all ballots cast. In New York City’s 2025 Democratic mayoral primary, about 5% of ballots were exhausted before the final round, though the winner still had a majority of all ballots cast. While RCV loses a few percent of ballots along the way, runoffs like Austin’s lose nearly 60% of voters. Another potential drawback of RCV is that it can eliminate a broadly liked candidate who might beat each rival in a one-on-one race (although this can also happen under a runoff system). Voters in some cities have also complained about the complexity of ranking candidates.
RCV is used statewide in Alaska and Maine and locally in New York City, Washington, D.C., Minneapolis, San Francisco, and dozens of other places. Utah, hardly a progressive laboratory, ran a pilot that let its cities opt into ranked choice for local elections, and voters liked it: in a 2023 survey, 94% of voters in participating cities said they were satisfied with it. At its peak, 23 cities opted in, most of them in Republican-leaning areas, before the pilot expired at the start of this year. Unfortunately, after RCV was blamed for the defeat of Trump-favored candidates in the 2022 Alaska House and Senate races, national conservative groups pushed legislation to outlaw it. Today, nineteen states have explicitly banned ranked choice voting. Alaskans, at least, seem to want to keep it: a second repeal attempt on last Tuesday’s ballot appears headed for defeat.
So what’s stopping Austin, and what can be done about it?
Austin’s 2021 vote for RCV came with a condition: the proposition applied only “if such voting is permitted by state law.” Officials read the law to rule it out, based on an opinion of the Secretary of State that was later endorsed in a 2003 opinion by then-Attorney General Greg Abbott. In the anti-RCV zeitgeist, state Republicans attempted to add Texas to the list of states where RCV is banned with SB 310, which passed the state Senate in 2025 on a 24-7 vote but died in the House.
But there’s new leadership at the Texas Capitol and a real opportunity to revisit how we vote. Notably, Attorney General-elect Nathan Johnson voted against SB 310 when he was a state senator. As a state representative, incoming Lieutenant Governor and Senate leader Vikki Goodwin filed a bill in 2023 to let cities use RCV in local elections. Meanwhile, Governor-elect Gina Hinojosa will have the opportunity to appoint a new Secretary of State, the official who oversees elections.
Given all that, there are three concrete steps that new leadership might take: 1) the next Secretary of State could revisit the office’s old guidance and come to a different interpretation; 2) a legislator could ask the incoming Attorney General for a new formal opinion on whether ranked choice voting satisfies the state’s majority requirement; and 3) legislators could refile Goodwin’s local-option bill, amending the Texas Election Code, though this would require support of the Republican-controlled state Senate. None of these steps has to impose RCV on anyone; they’d simply allow those cities that want to change how they elect their local leaders to do so.
On December 12th, far fewer Austinites will return to the polls to cast their ballots for City Council. Ranked choice voting isn’t perfect, and no voting system is. But it would settle races like these in one election, reflecting the preferences of the more representative November electorate. Hopefully, this will be the last time Austin has to vote twice.
How do these systems of voting rank for you?
Further Reading
Texas Turned Blue: What’s Next for Its Cities?
On the walls of the Texas Capitol rotunda hang fifty-three portraits of former governors and presidents of the republic. Whenever a governor leaves office, his or her portrait goes up and every other portrait moves over one spot. Soon, Republican Governor Greg Abbott’s portrait will join them...
If you enjoyed this essay and would like to support more of this work, please consider becoming a paid subscriber. Thank you for reading!
Runoff turnout and outcome figures are my own tally of City of Austin and Travis County results for all seventeen City Council runoffs held since 2014.