Donate

Ex-Insider Scott Shafer on Why Politicos Hate Ranked-Choice Voting

| October 24, 2011
  • Share:
  • Facebook
  • Pinterest
  • Reddit
  • Email

People wait in line to vote on November 4, 2008 in San Francisco, California. (Photo by David Paul Morris/Getty Images)

The November 8th mayoral election in San Francisco is ranked choice voting’s biggest moment. Other cities (including Oakland and even San Francisco in previous less-contested elections) have used this “instant runoff system” before. But this is by far the most consequential: A major U.S. city that’s politically progressive and has a diverse population with many different languages. Plus there are  –gulp! — 16 candidates with none likely to win a majority in the first round.

Exciting? Perhaps. But I can tell you from first-hand experience that political insiders — campaign strategists, pollsters, and most candidates — would like nothing more for ranked choice voting to be, well, recalled by the voters.

A little background. Before becoming a journalist at KQED I worked in politics and government. I was press secretary to Mayor Art Agnos and Chief of Staff to then-State Controller Gray Davis. I was also a partner in a political consulting firm that, among other things, managed campaigns for local and statewide candidates.

Political strategists are about one thing: Winning. How do you win? Raise a lot of money, buy ads on TV and radio, dig up dirt on your opponents, rack up enough endorsements, mobilize your ground operation and try to win free media coverage.

Ranked-choice voting makes almost all of those things harder…

Money? If every voter gets three votes instead of one, donors can be pressed to make contributions to multiple candidates, possibly diminishing each slice of that pie.

Advertising with hard hitting ads? What do you say? If you attack your opponents (voters say they hate negative ads, but guess what? They work.) you risk alienating their supporters, and possibly forfeiting those all-important second and third choices.

Endorsements? Instead of one whole endorsement, you get one-third of an endorsement. Newspapers, political groups et al now often recommend multiple candidates for each race.

You could even argue that voters are less motivated to volunteer (really grassroots, not to be confused with “astro-turf,” machine-driven “volunteers”).

Another reason Politics, Inc. dislikes instant runoffs — they make a lot of money on that actual second-election runoff. In a window of 30 days or so, the push to raise money for the runoff is fast and furious. Fundraisers go after all the donors to the losing candidates — or hit their original donors up for a second contribution. A lot of that cash lines the pockets of pollsters, strategists and campaign operatives. With ranked choice, the holiday shopping season is a little leaner.

Perhaps the main reason political insiders dislike ranked choice is that they lose control of the election. They know the runoff system and have thrived under it. Ranked choice has so many variables it seems like almost anything can happen, as last-place candidates are eliminated and their votes are redistributed.

In last year’s 21-candidate District 10 Supervisorial election, Malia Cohen was in fourth place after the initial round of vote counting. But after 19 rounds of eliminating last place candidates, Cohen won. No one could have predicted that outcome. Political consultants tend to hate that kind of uncertainty.

When Jean Quan edged out Don Perata 51-49% after starting out behind 35 to 24%, Quan called it a “really a victory for grassroots effort in this city.” But Perata’s team cried foul. Perata’s consultant John Whitehurst (now running Dennis Herrera’s campaign for SF Mayor) called it “a travesty” that his candidate won 78 percent of the city’s precincts but ended up losing. He called ranked-choice voting “an injustice,” adding “Oakland will pay the price.”

Which raises a question. How do you measure whether ranked-choice voting is a “success?” Whitehurst implies that it depends on how popular or successful the winner is. If Quan is deemed a failure by voters, does that mean ranked choice failed? Or should assessments of election systems be independent of how popular the winner is in office?

Related

Explore: ,

Category: Politics

  • Share:
  • Facebook
  • Pinterest
  • Reddit
  • Email
  • http://fairvote.org Rob Richie

    Interesting piece, Scott. Two key facts, however:

    * When matched one-on-one against Jean Quan in Oakland, Don Perata in fact lost a majority of precincts. His “wins” in 78% of precincts were by plurality,, just as he “won” Oakland with less than 34% of 1st choices. But the whole point of ranked choice voting is to compare the top two candidates — and more voters ranked Quan ahead of Perata citywide and in most precincts. That’s why she won.

    * Malia Cohen was in third place in District 10, and was less than 0.3% behind the first round leader. It’s hardly remarkable that she was able to defeat rivals less than 1% behind her in first choices.

    Keep in mind that runoffs in San Francisco had major issues, which is why RCV was passed. Those issues included:

    * In the last 14 runoffs in 2003, ten of them had participation rates that were less than 65% of the first round. And in nine of those races, the runoff winner had __fewer_ votes than the first round leader. In contrast, of 15 RCV races that have gone to multiple round of counting in San Francisco, the participation rate of the final count has only once fallen below 74% of the first round — and never has a winner had fewer votes than a first-round leader.

    * The San Francisco Ethics Commission found that “independent” expenditures quadrupled in runoffs. That was a key reason it backed going to RCV. You can imagine in the Citizens United regime that this would be even more pronounced.

  • David Cary

    In an RCV contest, a voter has one vote, but can mark three choices for how that one vote can be counted. It is not true that a voter gets three votes. I’d expect KQED journalists to be using correct terminology. There is too much disinformation about RCV being thrown around as it is.

    Shafer’s analysis is a little too simplistic and not sufficiently backed by facts.

    For all the alleged dislike, nearly all the mayoral candidates say they support RCV in its current form, and think it’s a good idea. Some would even like to improve it. If Schafer knows otherwise, he should go public with the specifics.

    In competitive situations, if you are in the lead, you like stability and predictability. If you are behind, you want to shake things up. It’s true in politics, with or without RCV, just as it’s true in sports. A leading team rarely throws a Hail Mary pass.

    Let us know when you see a candidate touting their “one-third of an endorsement”. For many candidates, an endorsement is an endorsement and with RCV more endorsements are being made. If endorsements are less influential, it is partly because voters have been freed from the tyranny of vote-for-just-one.

    If contests under RCV are less predictable, that’s because it allows contests to be more competitive. That’s a good thing for the rest of us. Candidates will vary, depending on their situation, on whether they like more competitive contests.

    RCV doesn’t just shift control and power away from politicians, consultants, powerbrokers, and political financiers, it shifts power to voters. That’s why California voters across the state overwhelmingly would like to also be using RCV.

  • http://www.cfer.org Steve Chessin, President, Californians for Electoral Reform

    I hate to be the bearer of bad news to the campaign strategists, pollsters, and some candidates, but elections aren’t for them — they’re for the voters. And every exit poll done by a reputable disinterested organization (this of course does not include the San Francisco Chamber of Commerce) shows that San Francisco voters overwhelmingly like ranked choice voting. As you say, it takes control of the election out of the hands of the political insiders and puts it into the hands of the voters, where, I hope you’ll agree, it truly belongs.

  • http://danclark.me Dan Clark

    It is dissapointing that I have learned more from the comments on this article than from the article itself. After listening to this morning’s story on the California Report, I came to this blog post for more information–but only received it from the commenters.

    In the original story, the reporting is on how some people find RCV confusing. This is because reporters like you do not do your job! Next time, try edifying your audience instead of reporting on how dumb we are.

  • http://www.first5sf.org Top_Shelf

    Ditto Mr. Clark. Well-explained, fellow commenteers.

  • http://d10watch.blogspot.com Chris Waddling

    Actually, I kinda predicted the outcome of the D10 race after the initial RCV numbers came in. http://d10watch.blogspot.com/2010/11/feeling-wonky.html Also, Cohen ended up in third in the initial vote counting, not fourth, and her first place votes were only 54 behind Sweet. http://d10watch.blogspot.com/2010/11/on-perils-of-instant-runoff-voting.html

  • John E. Palmer

    “Politics, Inc.” doesn’t like it. This should be Shafer’s first point, head and shoulders above the others. Yes, true, but should that be the reason we do away with it–because political consultants get paid less? Baseball scouts didn’t like Moneyball for a similar reason (a more mathematical method of picking players diminished their worth)–that’s just too bad if RCV is a better way for the rest of us. Most authors seem to feature the “confusion” argument, demeaning their own and voters’ basic intelligence (ranking three choices just isn’t that hard)–I give Mr. Shafer credit here, if not his employer’s radio piece, which had “confusion” front and center.

  • Stephen Cassidy

    I disagree that all candidates dislike ranked choice voting. I supported my city’s adoption of the system as a citizen and support continuing elections under the system now that I am Mayor of San Leandro.

    What candidates need to accomplish in an election under ranked choice voting is straightforward – and constitutes an important step forward in creating a better political system. You must obtain support from all voters, whether as their first, second or third choice. Voters are no longer sliced into various demographic and political groups with some entirely ignored and the favor of others exclusively sought. Interest groups that insist upon loyalty from candidates to their narrow agenda have much less influence. You need to develop a strong volunteer organization, walk the streets, knock on doors, and seek to personally interact with and engage as many voters as possible. You can certainly still (and should) continue to create contrast between yourself and other candidates, but not in the form of negative attacks based on matters other than their records. You need to seek out and develop relationships with key supporters of your opponents – persons that will support you and influence others to support you should their candidate fall short. You end your campaign, not with a last minute hit piece, but with an optimistic message of a new future for the city, and one that includes groups that have been historically disadvantaged or marginalized. This is what ranked choice voting offers.