(Update, August 24, 2011: The office of the city attyorney in San Francisco reports that the deadline has passed for an appeal of the Ninth Circuit's ruling. The case is closed.)
On May 20, the Ninth Circuit Court of Appeals unanimously upheld a lower federal court ruling rejecting a legal challenge to the City of San Francisco's use of ranked choice voting (RCV, also known as instant runoff voting, or IRV). The three-judge panel emphatically dismissed the plaintiffs' arguments, including a particularly clear rejection of the claim that RCV violates the principles of one-person, one-vote or equal protection under the law. The court's well-reasoned and well-researched opinion comes as a growing number of cities and states are implementing RCV, and should eliminate any concerns about the legality of RCV under federal law. In addition to San Francisco, RCV is used in the California cities of Oakland and San Leandro, and about a dozen other locations, including St. Paul and Minneapolis, Minnesota, and Portland, Maine.
The lead plaintiff in this challenge to RCV was Ron Dudum, a perennial candidate who lost in his bid for the Board of Supervisors in 2006. Dudum sought to represent a majority Asian American district that year, and an analysis of the RCV rankings in the race indicate that he would have been defeated one-on-one when matched against any of the top three Asian American candidates. Just as he lost in a traditional runoff election in 2002, he was defeated in 2006 in the final RCV tally by the most popular of the Asian American candidates.
Despite the logic of his defeat, Dudum brought a suit seeking to halt San Francisco's current form of RCV. Due to voting machine limitations, San Francisco restricts voters to three rankings, which Dudum claimed violated the equal protection rights of voters who don't include one of the two strongest candidates among their rankings. Dudum's complaint indicated his goal was to move to unlimited rankings, but a statement from one of the lawsuit's leading financial backers, the California Apartment Association, demonstrates its actual goal was to halt the use of any form of RCV, based on its misperception that RCV is detrimental to conservative candidates.
Writing for the unanimous panel, Judge Berzon methodically explained and rejected each of the plaintiffs' arguments one-by-one in strong language. Calling any burdens imposed by restricting voters to three rankings "minimal at best," the panel concluded that the burdens Dudum alleged are "largely ephemeral, disappearing upon examination" and that "Dudum's contention mischaracterizes the actual operation of San Francisco's restricted IRV system and so cannot prevail."
Under an RCV system, voters get one vote and one ballot, but rank their "backup" choices in order of preference. If no candidate wins with an initial majority of first choices, the candidate with the fewest votes is eliminated, and their supporters' second choices are distributed to the remaining candidates in the first of what can be a series of "instant runoffs." The process of elimination and redistribution continues until one candidate has a majority of the votes among the remaining candidates. Because of the technical limitations of the ballot-counting machines - also found in several other cities, but soon to be removed in the next generation of optical scan voting equipment - San Francisco limits the ranking to three candidates, a restriction the court found to be constitutional.
Here's a summary of how the 9th Circuit struck down each of Dudum's arguments:
(1) The system does not allow any voter to cast more ballots or votes than other voters, definitively addressing the claim that RCV gives some voters more than one vote.
(2) Any ballots that "exhaust" during the count (meaning they don't rank any of the remaining candidates) have been counted in the election. As will happen to even more voters in a vote-for-one plurality system, they happened to count only for losing candidates, but that does not mean they didn't count.
(3) The RCV system does not result in vote dilution because the ability to rank preference does not affect the mathematical weight of any vote;
(4) Any of the "extremely limited" burdens on voters resulting from the system are justified by legitimate government interests, including the ability of the voter to express more nuanced preferences in an RCV system. The judges wrote, "in sum, we have no difficulty in holding that these important government interests are more than sufficient to outweigh the extremely limited burdens - if any - that the restricted IRV features Dudum challenges impose upon San Francisco's voters."
Although the opinion did not endorse a particular election system, it noted a number of benefits to RCV in contrast to both plurality voting and traditional runoff elections. RCV provides the voter more influence over his elected representation - if the winner is not to be his or her first choice, that voter has a chance to help elect a second or third choice. Doing so allows for more nuanced voting preferences than traditional vote-for-one systems, and, as Judge Berzon pointed out, "means that candidates with greater plurality support tend to be elected, as compared to a traditional plurality system." RCV also allows for voters to openly support third party candidates, without casting a "spoiler" vote.
Bankrolled by the California Apartment Association (CAA), the plaintiffs hired the well-known political advocacy and law firm Nielsen, Merksamer, Parrinello, Gross & Leoni, LLP. The CAA's online statement discussing the case suggests that its motivations were blatantly political and clearly designed to eliminate RCV rather than the lawsuit's professed goal of allowing voters more than three rankings. Here is the nearly incoherent statement provided by CAA Legal Committee Chairman Dave Wasserman of San Francisco explaining why they were involved in the case:
"So the issue across the state, and indeed the nation, is how ranked choice voting (RCV) can be utilized to essentially crowd the field so as to ensure that the most qualified, and even most popular, candidate is defeated. The dilution of votes works against the strongest candidate. Just ask Don Perata who lost the Oakland Mayoral election. In big elections, like a mayoral one, the progressive left will field a slew of people who will garner second and third choice votes. The statistical odds favor their ultimate victory as compared to the main vote getter (candidate who has the most first choice votes) based upon how RCV tabulates voting points. In San Francisco, we have two supervisors who won with a minority of first choice votes but who garnered more second and third choice picks. This means, with RCV, that we oftentimes cannot get behind a strong candidate, given the likelihood that a lesser candidate will take the election. As such, our political sway is greatly diminished under the RCV scheme. Just ask the folks in Oakland."
It's hard to know where to begin in correcting Wasserman's claims. Briefly, however:
- 2010 mayoral candidate Don Perata wasn't the most popular candidate in the Oakland election. More Oakland voters preferred Jean Quan to Perata when comparing them one-on-one - and Quan won the most votes of any Oakland mayoral candidate in decades.
- Any "dilution of votes" due to "crowding the field" theoretically would hurt the voters who spread their vote among several candidates, not backers of a single candidate. This claim directly challenges the alleged basis of the lawsuit, which was that limiting rankings to three hurt voters.
- "The progressive left" is no more likely to have more candidates running than conservatives, depending on the context -- just look at the 2012 presidential election. Indeed, this year in a special election in California for Assembly District 4 there was one Democratic candidate matched against seven Republicans in the first round. The Democrat led in that first round, securing 30% of the vote as compared to 21% for the top Republican But the Republican won a runoff election, just as a Republican would have almost certainly won with ranked choice voting. Both runoffs and ranked choice voting uphold majority rule for whichever side has the most support.
It is the last part of Wasserman's statement that may suggest what's really going on: "Our political sway is greatly diminished under the RCV scheme." What RCV elections demonstrate is that big money interest groups like CAA have less power to control outcomes than in a one-on-one runoff election, when they can pour money against the candidate that interest group opposes. In 2003, the San Francisco Ethics Commission formally endorsed moving forward with RCV as quickly as possible because it had determined that runoff elections created more power for special interest groups. In the 2002 runoff elections, independent expenditures had quadrupled over the first round - and largely were spent in negative ads attacking candidates.
Rather than help any one side in the partisan game, RCV helps voters - and creates incentives for candidates who can get to know more voters directly in their effort to earn majority support against their strongest opponents. Rather than disenfranchises voters, RCV promotes candidates who work to connect with voters in their community, earning second or third choice support in a way that hardworking conservative candidates can do as well as progressive candidates. Experience from RCV elections suggests that earning second and third choices is best accomplished by showing that you are ready to listen to voters; they may not see you as their first choice, but can believe that at least you will not ignore them. As with several RCV elections, Jean Quan's victory in Oakland in 2010 seems grounded in her attending far more community meetings and having far more direct contact with voters than her much better-funded rival Perata. But well-funded special interest group like the CAA may lose out if their preferred candidate tries to win by relying on spending more money.
The Ninth Circuit's decision is an important legal victory, both for RCV and election reform as a whole. The judges' discussion of the challenges inherent to all election systems, including the most commonly used "winner take all" or "first past the post" system, highlights the very real need for election reform on all levels of government in the United States. Their ruling leaves RCV in place for this November's highly competitive, San Francisco mayoral election. The race currently has several strong candidates, but ultimately city voters will elect as their leader the individual with strong enough support to advance out of the early rounds, but broad enough support to be able to defeat his or her strongest opponent one-on-one. You can find out more about the election and RCV in the Bay Area at Demochoice's new San Francisco page and FairVote's Bay Area RCV resource page.