Once is weird. But is twice a trend?
In two consecutive elections in Portland in less than six months, ranked-choice voting results were problematic in the city of about 68,000 people.
It made history this past Election Day, in what one expert said was probably the first-ever tie in an RCV election.
The apparent tie in the at-large City Council race was later decided by drawing a name from a bowl, with Brandon Mazer declared the winner. That result was overturned by a recount, and Roberto Rodriguez was ultimately awarded the seat.
But the recount produced another bizarre twist: an additional 36 votes that were not counted on election night, which ultimately helped sway the election. This vote discrepancy is being reviewed by the city as well as the Nebraska company whose software runs RCV for the city.
Coupled with what happened during last June’s Charter Commission election – when a threshold requirement essentially vaulted candidates with considerably fewer first-place voters over a candidate who had the second-largest number of first-place votes in the at-large election – RCV is now in a harsh light.
Is it working? And even if it has technically worked the way it is supposed to, should it be reexamined?
Charter Commission agenda
While not necessarily one of the primary issues facing the nine elected members of the 12-person Charter Commission, some commissioners now say this may be a good opportunity to improve the system.
At-large Commissioner Marpheen Chann, who chairs the commission’s elections committee, said in his opinion RCV works and has increased voter participation and engagement in elections that otherwise might have attracted just a few thousand voters.
Turnout on Nov. 2 was about 21,000 voters. The Charter Commission election in June, which historically draws tepid interest from voters, produced 8,884 ballots.
But Chann also said there is room for improvement, which is why his committee is considering recommending the city move toward a proportional RCV system, which he said some experts have recommended.
“In terms of the tie issue, I’m more than happy to look at solutions that inspire more public confidence, as much as (City Clerk Katherine Jones’) salad bowl was a hit,” he said. “Drawing lots here was more of a formality that then allowed for a recount to be requested, which gave us a winner.”
Commissioner Robert O’Brien, who also served on the 2010 commission, said he supports this examining RCV and fixing problems. He noted the last commission decided to only extend RCV to mayoral elections, not the City Council or other elected offices.
He also noted citywide RCV came about via referendum.
Patricia Washburn, an at-large commissioner who trailed considerably after the first round of voting in June but won her seat thanks to the runoff results, admitted the formula could use adjustment.
But Washburn, who is a member of the elections committee, said she isn’t sure it is the commission’s job to “get down on that level of detail” and said she continues to support RCV because it gives more of a voice to each voter.
“In general I think we are looking at recommendations that support and uphold greater access to voting and elected office, both to improve diversity and opportunity, and to give voters more choices,” Washburn said.
Catherine Buxton, the third member of the elections committee, and Nasreen Sheikh-Yousef are the other two at-large commissioners. Neither of them responded to requests for comment.
Commissioner Michael Kebede, who chairs the full commission, declined to comment until the elections committee has decided whether it will vote on this issue.
‘A more perfect system’
Portland adopted RCV for its mayoral elections in 2011, a result of a recommendation by the 2010 Charter Commission. The city later expanded RCV to City Council and School Board elections in 2020 via referendum.
Under the City Charter, a winning candidate must receive more than 50 percent of the vote, with voters ranking the candidates by preference. If no candidate passes that threshold, there are runoffs, where the candidate with the fewest first-place votes is eliminated, and that voter’s next preference gets support. That process continues until someone breaks 50 percent.
This model caused some confusion in the Charter Commission race in June, where 10 candidates were on the ballot for four at-large seats. Experts said a lower victory threshold than 50 percent would have been preferable.
Westbrook this month became the second Maine municipality to adopt RCV for local elections. The state also uses RCV for primary elections for governor, and all legislative and congressional contests, and in the 2020 presidential election.
Westbrook City Councilor David Morse, who proposed RCV to the City Council, said Westbrook is fortunate to be able to observe Portland as its clerk is developing regulations to implement an RCV charter amendment. He said it will anticipate challenges Portland has experienced, including ties and recounts.
“I’ll note that our charter amendment anticipates the possibility of using RCV in a future Charter Commission election and lays out the procedure for how those rare and unusual elections will work,” he said. “Our clerk wanted to ensure that there was no ad hoc rulemaking needed in the unexpected event of a Charter Commission being called.”
Morse said the two recent events in Portland were in line with what people expected when they endorsed RCV: the people who hold public office need the support of the majority of voters. He noted recounts happened before RCV and resulted in “challenging circumstances.”
“Westbrook’s ability to observe Portland’s experiences helps Westbrook and everywhere else as we work towards the goal of a more perfect system,” Morse said.
Westbrook City Council President Gary Rairdon, an opponent of RCV when it went to the council, did not respond to a request for comment.
‘Black eye’ for RCV
Former Portland City Councilor Cheryl Leeman, who served as mayor when the position was more comparable to a council president, said she was at the polls on Election Day and heard “a lot of grousing” about RCV.
“It’s confusing, and we probably didn’t do a good enough job of educating before it got implemented,” she said.
Leeman was not the only one to say RCV continues to confuse voters. On Election Day, wardens at Reiche Elementary School and Deering High School said there were constant complaints about RCV.
Leeman, who ran unsuccessfully for the Charter Commission this year, said there’s been an acknowledgment the ordinance is flawed when it comes to multiple candidates running for multiple seats, and she hopes that will be corrected.
Leeman said the last two instances in Portland have shown that RCV is not working as intended because “people have figured out how to game the system” by forming slates.
“It was intended as a fair and more equitable system so that everybody, regardless of what your political coalition might be, would have an opportunity to run,” she said. “When someone with 4 percent of the (first-place) vote gets elected, that’s a problem.”
She said a similar thing happened in the at-large council race, where voters were encouraged to vote for and rank only two candidates.
“It didn’t have to be that way,” Leeman said. “I’ve never seen anything quite like it. You run in an election, you knock on doors, get the message out there, that should be adequate.”
She said the last two elections have hurt the process.
“They’ve given ranked-choice voting a black eye,” Leeman said.
‘Unusual outcomes can be helpful’
Anna Kellar, executive director of the League of Women Voters and Maine Citizens for Clean Elections, is an RCV advocate who believes the process is working the way it’s supposed to.
“Voter participation has been high, and we’ve been able to elect candidates who have the support of a majority of voters,” Kellar said. “It’s important to remember that a close election can happen under any voting system, and that recounts almost always lead to small changes in the results as ballots are reexamined.”
Kellar said RCV is still relatively new, and each election provides lessons on how to tweak the system and make it run more smoothly. They said their organizations want to work with the city to see what changes need to be made to improve the system.
“Unusual outcomes can also be helpful as we think about how to balance different goals for our elections, for example, ensuring majority rule and minority representation in an elected body,” Kellar said.
Kellar said the Charter Commission election showed that in a multi-winner election, the majority can end up winning all the seats, and minority viewpoints end up with no representation. But, they noted, that can also happen in non-RCV elections.
“The Charter Commission race has led to an interesting conversation (on the commission) now about the value of proportional representation, and ways of using RCV in multi-winner races that would lead to more voters having a representative that reflects their views, while still ensuring majority rule,” Kellar said.
While on the cutting edge, Portland isn’t alone in using ranked-choice voting.
Burlington, Vermont, adopted RCV, then rejected it, and then last winter voted to restore it. Cambridge, Massachusetts, uses it to elect members of the City Council and School Board. Berkley, California, uses it to elect its mayor, City Council, and city auditor. Minneapolis uses it to elect 22 city offices, including the mayor and councilors. New York City adopted it in 2019 for municipal elections. Oakland, California, uses it for 18 single-winner city offices. San Francisco uses it for mayor, city attorney, its Board of Supervisors, and five other citywide executive offices.
Before RCV, winning an election in Portland was pretty simple: the person with the most votes wins. Had that been the practice this time around, Rodriguez would have still been declared the winner in the four-way race with just under 25 percent of the vote; according to official election results, he had 275 votes more than Mazer.
Chris Hughes, policy director of the Ranked Choice Voting Resource Center, which provides guidance and information on RCV, previously said he didn’t see Portland’s recent experience as an indication of a trend. Rather, he said they were just two highly unusual outcomes.
Hughes said ties just don’t happen, and said to his knowledge Portland’s at-large council race was the first time in history an election using RCV had that result. He also cautioned against connecting the two elections as anything other than a bizarre coincidence.
“It’s sort of just two very unlikely events happening in one year,” Hughes said.
Hughes also previously advised the city that the model it uses was not the right one for the Charter Commission election.
In the meantime, Elections Systems and Software – the Nebraska company that helps the city conduct RCV – continues to work with Portland to try and resolve the 36-vote discrepancy between the number of votes counted electronically on election night and the number tabulated during the recount.
City Hall spokesperson Jessica Grondin said city officials wouldn’t have anything more to say about what happened until after Associate Corporation Counsel Jennifer Thompson and the city clerk complete a report for the City Council.