This article is part of U.S. Democracy Day, a nationwide collaborative on Sept. 15, the International Day of Democracy, in which news organizations cover how democracy works and the threats it faces. To learn more, visit usdemocracyday.org.
When City of Boulder voters cast their ballots in the Nov. 7, 2023 election, they will directly elect the city’s mayor for the first time.
And instead of selecting just one candidate, voters will rank as many candidates as they want in order of preference. This process is known as ranked-choice voting, or more specifically, instant-runoff voting.
The election change is the result of the overwhelming approval of the 2020 Our Mayor Our Choice ballot measure. Previously, members of the Boulder City Council appointed one of their own to the position.
The new voting method is designed to ensure that the winner has the backing of a larger portion of voters, as opposed to the more typical plurality voting system, where the candidate with the most votes — even if it’s a small percentage of votes — wins.
Four candidates are vying for the post: Mayor Aaron Brockett, Councilmember Nicole Speer, Councilmember Bob Yates, and Paul Tweedlie, a retired software engineer.
The race is historic because it’s a first for Boulder. Legally, however, the mayor will not gain any extra political power. The mayor is still one of nine councilmembers with additional responsibilities of representing the city and managing meetings, among other tasks.
The Boulder County Clerk and Recorder’s Office will oversee the election.
“We now have an upgrade to our voting system to conduct ranked-choice elections and have done thorough testing of its functionality,” Mircalla Wozniak, a spokesperson for the Boulder County Clerk and Recorder’s Office, told Boulder Reporting Lab. “We are excited and confident [about] being the first Colorado county to conduct a ranked-choice election on a modern voting system.”
How to fill out a ranked-choice voting ballot
Ballots will be mailed out on Oct. 16, 2023. In the mayoral race, voters will see a box with rows for each candidate and columns to rank them from first to fourth.
You don’t have to rank them all for your vote to count. But here are some guidelines and scenarios to keep in mind:
- You can rank as many or as few of the four candidates as you want in order of preference.
- If you only support one candidate, you can select that candidate as your first choice and leave the rest blank.
- If you mark the same candidate for your first and subsequent choices, your ballot will be counted as if you only selected that candidate as your first choice and left the rest blank. In other words, there is no advantage to selecting your favorite candidate multiple times.
- If you select multiple candidates for your first choice, your vote will not be counted.
How votes will be counted
The county will count the ballots initially by tallying all the first-choice selections. If a candidate gets more than half the votes, they win.
If no candidate earns more than 50% of votes in that initial round, the candidate with the fewest first-choice selections will be eliminated. Those votes would then be reallocated to the other candidates based on voters’ second-choice selections. This process of redistributing votes will continue until a candidate has an absolute majority.
The winner will serve for one three-year term. Beginning in 2026, when the city is planning to switch its elections to even years, voters will elect the mayor to two-year terms.
Ranked-choice voting is not new
Alaska uses ranked-choice voting in its general elections, but not primaries. Maine uses it in federal elections, but not state elections. Many large cities use some form of ranked-choice voting for mayoral races, including New York City, San Francisco, and Cambridge, Massachusetts.
Cities in Colorado have dabbled with some form of ranked-choice voting, including Aspen, Basalt, Carbondale and Telluride. Broomfield and Fort Collins are moving ahead with the new voting method. Denver and Pueblo have considered it.
Even the City of Boulder once used ranked-choice voting. It was called the “Hare system of preferential voting,” according to the 1917 City Charter. The goal was to ensure “real representation” and prevent “well known evils of parties and of excessive campaign funds,” the charter states.
What makes Boulder’s transition to instant-runoff voting different from most other cities is that councilmembers with two years left to serve can also run for mayor, according to Celeste Landry, a co-leader of the Voting Methods Team for the League of Women Voters of Boulder County.
Councilmember Speer’s term ends in 2025. If she wins her bid for mayor this year, that opens up a fifth seat on the Boulder City Council. However, on this year’s ballot, voters can vote for up to four city council candidates. So in this scenario, the top five vote-getters would win seats.
“We have this interplay between the council and the mayoral contest. They can impact each other,” Landry said.
Landry said the League of Women Voters of Boulder County supports instant-runoff voting but has not established a position on directly electing mayors. It did not take a stance on the 2020 ballot measure for the direct election of city mayor.
How the new method will change campaigning is unclear
Beyond being more democratic, supporters of instant-runoff voting see it as a way to change how candidates campaign, in contrast to the more widely used system of “winner take all,” where a person can win with only a plurality of votes.
Under the City of Boulder’s ranked-choice voting system, candidates who appeal to a broad range of voters, at least theoretically, stand to benefit because they may have a better shot at earning second- and third-choice votes, not just first-choice selections. So there could be an incentive to build coalitions and avoid negative campaign tactics.
But it remains unclear exactly how this will play out in Boulder.
The mayoral race is happening alongside a high-stakes city council election in which a majority of seats are up for grabs. In city council elections, unofficial candidate committees typically raise and spend money on behalf of a slate of candidates. Because there tends to be two sides on key issues, this effectively divides candidates into two separate camps, or slates.
A similar situation has already played out in the race for mayor. Boulder Elevated, a newly formed political organization that has made public safety one of its top issues, for instance, is endorsing Yates for mayor. The Boulder Progressives, an organization that has advocated for more liberal causes, such as housing density, is backing Speer and Brockett. Both organizations have endorsed a slate of city council candidates too. No candidate has won the endorsement of both groups, which, for the most part, are on opposing sides on major issues.
How you can learn more
Boulder County has an explainer on its website.
The city has an interactive ballot to practice voting. It’s available in English and Spanish.
The League of Women Voters of Boulder County is hosting an event at the JCC on Tuesday, Sept. 19 at 6:30 to discuss the election changes. Boulder County Clerk Molly Fitzpatrick and Boulder City Clerk Elesha Johnson will be speaking and answering questions. Spanish interpretation will be provided. The event will be recorded.
Thank you for a great overview of an important election!
But you made a common mistake by saying “This process of redistributing votes will continue until a candidate has an absolute majority.”
No voting method can guarantee that the winner gets a majority. That’s obvious, because of course no one can force half of the electorate to support anyone in particular. The IRV (single-winner RCV) method claims to do so, but does so by ignoring any ballot which doesn’t have a vote for one of the two candidates in the final round, thus completely ignoring the preferences of voters who don’t support one of them.
And IRV sometimes doesn’t even get the right two candidates to the final round, and can thus ignore the votes of people who support a candidate who should have won, and who in fact would win if EITHER of the last-round candidates were instead eliminated first. It thus fails what scholars call the Condorcet criterion.
An alternative voting method, also supported by the LWV, which doesn’t have this problem, is Approval Voting. See the Center for Election Science (https://electionscience.org/) for more on Approval voting.
For more on the issue of false majorities, see The Majority Illusion: What Voting Methods Can and Cannot Do
https://electionscience.org/commentary-analysis/the-majority-illusion-what-voting-methods-can-and-cannot-do/
The Maine Heritage Policy Center has a report on “The Failed Experiment of Ranked-Choice Voting” which documents a wide variety of IRV/RCV contests in which the winner did not get a majority.
https://mainepolicy.org/wp-content/uploads/RCV-Final-Booklet-.pdf
True
What’s more Boulder “Elevated” and “Progressive”‘s platforms conflict, so a slate promoting a candidate on one side or the other only serves to reveal the confused status of the community at large. Increased homelessness equals less safety, as does increased density without population checks, since it increases homelessness too, so there really was no choice. It was the wrong question that got answered. One’s own integrity is superior to that of collective confusion and if more folks used integrity and logic, the social experiment would be in a better place.