Peter Hutchinson on meager plurality winners: Ranked Choice Voting

Over at MPR’s commentary link, Peter Hutchinson (remember him? Mike Hatch does) laments the current state of elections: “Our election system doesn’t work. People know that. Exhorting good people to run for office in a system that punishes practical experience and rewards ideological extremism won’t do it. Similarly, exhorting people to “show up” won’t get them to turn out if they feel that the system offers them nothing but bad choices or that their vote won’t count. It’s going to take something a lot more ambitious if we are going to get our democracy back.     A great place to start would be to require that a candidate get a majority of the vote to win an election. … For state-level elections, some propose using RCV in partisan primaries and then again in the general election. But given that voters don’t bother to show up for primaries (just 9 percent in Tuesday’s election), why not eliminate the primary altogether, as is done in local RCV elections? The whole election can be handled in November. Just put all of the candidates on the ballot, let them make their pitch to the voters, and in a high-turnout election let the voters rank them from top to bottom. Then use those rankings to assure that the winner is the one who gets the support of a majority.”

We’ll take it. The AP says: “Minnesota employers added 6,800 jobs in July, the second month in a row that the state gained jobs. The Department of Employment and Economic Development released the July job numbers on Thursday. While jobs were added, the state’s unemployment rate climbed two-tenths of a percent to 5.8 percent, due to seasonal adjustments.”

Rachel Stassen-Berger of the Strib adds to the context of the Nolan-Cravaack race in the 8th District. She says: “Already, outside groups have poured more than half a million dollars into the district — and that was before the DFL race was settled in a messy three-way primary. … Cravaack is in an enviable position. He has collected more than $1 million in contributions and hardly spent anything while the DFLers were slugging it out. Fresh from the primary, his fundraising dwarfs Nolan’s. The DFLer, who has been out of elected politics for 30 years, had less than $90,000 in the bank, according to his last report. Cravaack, a devotee of Republican vice presidential candidate Paul Ryan’s stringent budget proposals, has proven politically savvy in picking up a union endorsement, which shocked some outside GOP observers, but which could prove a crucial boost in a district known for its strong union ties. That, too, has Democrats nervous that this fight will be grinding, expensive and uncertain.”

Experts are thinking the coming flu season could be a rough(er) one. Rupa Chenoy at MPR says: “After a couple years of the same old flu, there are two new types in town. ‘If you want to know where they came from … The ultimate origin was Victoria, Australia, and Wisconsin,’ said Kris Ehresmann, the infectious disease director at the Minnesota Department of Health, where she and other top officials gave a preview of the coming flu season.  Ehresmann said fewer people than normal were sick with the flu over the past two years. That could be because the strain of flu was the same, allowing more people to become vaccinated. With new flu strains, Ehresmann says Minnesota may experience a regular flu season.”

Also at MPR, Sasha Aslainian reports: “The ad war over the marriage amendment begins at noon today. Freedom to Marry, a national group working to defeat the amendment, is launching its first television advertisement in Minnesota in what will likely be the first of many volleys to win votes on the marriage issue this election season. … Minnesota for Marriage, the main group working to pass the amendment, has not yet run ads, and a spokesman says they will come later in the campaign.”

What are you going to do? Paul Walsh at the Strib reports: “Shawn G. Swanson, 21, was on top of an SUV and then fell to the pavement while it was being driven about 11:30 p.m. Saturday by another 21-year-old St. Paul man on a dark road in Oakland Township, according to the Burnett County Sheriff’s Office. Swanson died the next day. The driver was arrested on suspicion of homicide by intoxicated use of a vehicle. He was released from jail after posting bond Monday and awaits charges. The Sheriff’s Office said Swanson, the driver and three others were among a group from the Twin Cities involved in car surfing on Glendenning Road in Oakland Township.” There are explanations for why the late teens and early twenties are the most dangerous ages for young guys.

The GleanBest Buy founder and former CEO Richard Schulze is having a hard time getting the company to disclose its financial details. Mike Hughlett in the Strib says: “Schulze, the company’s founder, has a group of private equity companies lined up to help bankroll his plan to buy the struggling company. ‘A number of leading private equity firms have informed me that they are prepared to make significant commitments, subject to due diligence,’ Schulze wrote in a letter Thursday to Best Buy’s board. Schulze says the company has so far refused to let him examine its financials. … Schulze, who controls 21 percent of Best Buy’s stock, said in Thursday’s letter that he is ‘prepared to roll over into this transaction at least $1 billion of my own equity — and potentially all of my existing stake.’ “

The Stillwater Lift Bridge is closing Sept. 10. Andy Rathbun at the PiPress says: “Starting that day, the bridge will close to all vehicle, pedestrian and bicycle traffic through late December. River traffic will also be affected. The bridge will be raised only on three Sundays — Oct. 21, Oct. 28 and Nov. 4 — said Kevin Walker, spokesman for the Minnesota Department of Transportation. Rehabilitation work on the aging bridge is the cause for the closure. A $3.4 million rehab project began July 9 and is meant to maintain the structure until it can be converted to a bicycle and pedestrian bridge in 2017.”

Remember the lawyer who represented that Wisconsin religious group (aka “cult”)? David Hanners at the PiPress writes today: “The state board that monitors lawyers wants a Hastings attorney suspended for a long list of offenses, among them filing suits in bad faith and harassing judges, other lawyers and people she was suing. The lawyer, Rebekah Nett — who has represented a Wisconsin group that former members say is a religious cult — has 20 days to respond to the allegations. … In a 26-page petition for disciplinary action, the two top people of the Office of Lawyers Professional Responsibility wrote that Nett, 36, had engaged in a long standing campaign of unprofessional conduct in Minnesota and Wisconsin. ‘(R)espondent has engaged in an extensive pattern of bad faith litigation and the filing of pleadings intended to harass, embarrass, delay or burden third persons, in various courts,’ board director Martin Cole and first assistant director Patrick Burns wrote in their petition filed with the Supreme Court.” Put another way, she’s giving litigious attorneys a bad name.

The invariably entertaining John Hugh Gilmore has been blogging more regularly at his Minnesota Conservatives site. His latest post on Kurt Bills’ victory in the GOP Senate primary Tuesday, where TeamBills “soundly trounced” two competitors, asks: “Which stoned intern wrote this fiction? Does the campaign think the reader doesn’t notice a lack of numbers in the release? Soundly trounced? A 51 to 49 win is not trouncing, even for the marginal liberty denizens. Bills has run a lousy campaign, going to safe areas (parades!) and never reaching out to moderate democrats and independents. Indeed, Bills has made no overture to the base of his own party which he and his unserious, moronic Paul supporters pretend he represents. … Party leaders, such as they are, were mute in their praise of this smashing win. The requisite MN GOP press releases were issued so one is at least comforted that the mechanics of a functioning state-wide party were in order. But Shortridge, Fenton, Johnson & Anderson lead a fool’s brigade to Tampa. They made a conscious decision to align themselves with what they thought was the ascendant movement in the party. Yes, the very definition of leadership, that. Someone should break it to Jeff he’ll never be governor. One assumes the other three know how to dog paddle.” I give points to anyone who can apply humor to the cartoonish state of modern politics.

You can also learn about all our free newsletter options.

Comments (13)

  1. Submitted by Arito Moerair on 08/16/2012 - 02:16 pm.

    GOP says no to RCV

    The GOP hates RCV, which is ironic, because Tom Horner probably allowed Mark Dayton to win.

    On the other hand, George W Bush would have never been president, and we never would have gotten eight years of Timmy.

    • Submitted by Eric Ferguson on 08/16/2012 - 02:59 pm.

      I don’t get GOP opposition

      We’ll never know if Horner helped Dayton, but given the result, if you’re Republican, why not try RCV? I’m not convinced RCV would have spared us Pawlenty, but then again, there’s no way to ever know. At least with RCV, we would know who voters really wanted if they could vote without having to be strategic, and who a majority really want between the top two. I have a feeling RCV would work even better for partisan races than non-partisan since at least the party label gives a clue about who an unknown candidate is.

      • Submitted by Dale Sheldon-Hess on 08/16/2012 - 07:47 pm.

        Unfortunately, RCV doesn’t remove the need to be strategic. Consider what happens when a third-party (C) starts to actually do well and can get more than a third of the (first-rank) votes:

        45%: A > B > C
        25%: B > A > C
        30%: C > B > A

        Here, party B is preferred over party A by 55% of voters, but the lose anyway because they are eliminated first; they A faces C and clobers them 70%:30%. But, if the C-favoring voters were *strategic*, they would vote for “the lesser of two evils”, B, so that B wins. This is the exact same dynamic that happens with the current voting system, and which will pull it towards delivering the exact same results.

        There are other voting system reforms which don’t have this problem; with approval voting, it is never to your advantage to give anything less than your full support to your honest favorite.

    • Submitted by Dale Sheldon-Hess on 08/16/2012 - 07:52 pm.

      It’s not always the GOP; it’s whichever party is smaller in the state. When RCV was proposed in Alaska, it was the Democrats that were opposed.

      This is because the smaller party’s best-hope is that two nearly-identical candidates split the vote for the larger party, letting them sneak in. This becomes harder under RCV (although the system still manages to punish cross-over and truly moderate candidates; which is why the larger-party’s base will have no trouble supporting it.)

  2. Submitted by James Hamilton on 08/16/2012 - 02:37 pm.

    A better idea

    If a primary fails to draw more than 25% of registered voters, we simply cancel the election and leave the seat open.

    • Submitted by Thomas Swift on 08/16/2012 - 06:29 pm.

      I’ll second that!

      And raise you one. If the seat remains open two consecutive terms, it gets eliminated…I think we’re on to something!

  3. Submitted by Eric Ferguson on 08/16/2012 - 03:03 pm.

    reconsider primaries

    Whether we go to RCV or not, maybe we should reconsider primaries. The whole reason for holding primaries instead of letting the caucuses decide is so many more people turn out and the most ideological voters can’t make the decision alone. However, in actual experience, few people vote in the primaries, and the party endorsements are almost always ratified. So what’s the point?

  4. Submitted by Dan Hintz on 08/16/2012 - 03:56 pm.

    Majority vote

    If the goal is to require a candidate to have a majority of the vote to win an election, then RCV is not the answer.

    San Francisco had an RCV election in 2011. Mayor Ed Lee received 59,775 first place votes out of 194,418 votes cast (30.7 percent). After all the second and third place votes were transferred Lee was up to 84,457 votes (43.4 percent). The second place candidate had 57,160 (29.4 percent) after the second and third place votes were reallocated. The remaining 52,524 votes were on “exhausted” ballots and did not figure in the outcome. Had the pre-RCV run-off system been in place, those 52,524 voters could have chosen between the first and second place candidates and you would have had a majority winner. Under RCV, nearly 30 percent of voters were disenfranchised and the mayor was elected without a majority of support, even taking into account second and third choices.

    Oakland had a similar election outcome in 2010. Mayor Jean Quan was elected with 53,897 votes from 119,607 voters after reallocation, which is only 45.06 percent. The second place candidate – who actually had far more first place votes than Quan (35 percent to 24 percent) – got 51,720 votes after reallocation 43.2, with just under 12 percent of ballots being exhausted.

    • Submitted by Ed Day on 08/16/2012 - 11:36 pm.

      Then find a way to address exhausted ballots

      The main problem with a normal runoff election about a month later is that turnout appears to go down significantly, disenfranchising voters in a different way.

      That said, I think the other goals of RCV of voter engagement and determining the candidate with the broadest appeal (ideally a majority) are still valid. In what was essentially a three-way race in 2006, reallocating the Hutchinson votes would have made Hatch a clear winner. (And most likely, a less resounding victory for Tom Emmer in 2010.)

      The 2002 race for governor is trickier, as from what I could tell from working a nonpartisan GOTV drive was that Tim Penny was the candidate a large majority would have been comfortable electing, but not their ideal first choice. In both a normal and RCV, he would not have been in the top two.

      However, in a system like Cy Young voting, which weights the rankings with point values (first place =10, second =7, etc) there’s a fair chance Penny wins (assuming people are honest and not writing in Donald Duck for everyone after their first choice.) I’m assuming some mathemeticians can figure out a fair way to do this.

      Currently, RCV would work best in clear three-way races.

      • Submitted by Dale Sheldon-Hess on 08/17/2012 - 02:00 pm.

        Turnout depends on a lot of factors, and often a close high-ticket race like mayor will see INCREASED turnout in the runoff. It’s true though that no one turns out for the runoff for dog catcher.

        The point-value system you suggest is called a Borda count, and it’s better than RCV, but still has shortcomings. There’s an improved system that’s like the Borda count, called score voting, where instead of each rank having a defined score, you can give any number of points (up to the maximum) to any number of candidates; give all but one 10 points if that’s what you want (great if you just want to say “anyone but…”) Moving beyond rankings to this sort of evaluative system (where each candidate’s score is independent of all others) turns out to be a huge improvement for voter satisfaction; in fact, setting the max score to just 1 (so you can give each candidate either a 1 or a 0) gives a very large improvement, but with very little change in ballots or voting machines (just change the plurality “vote for one” to “vote for any”.) This minimalist form of score voting is called approval voting.

        And RCV is actually pretty poor in a close three-way-race, because it often usually eliminates the compromise candidate first. I have an example in an earlier reply upthread that illustrates the problem.

  5. Submitted by Dan Hintz on 08/17/2012 - 12:55 pm.

    Can’t be fixed

    The only way to fix the problem of exhausted ballots is to force everyone to rank every single candidate on the ballot. That would mean that no matter how vile or offensive a candidate is to you, you would have to cast a vote for that candidate or your ballot will be invalid. Obviously, that isn’t going to work – you can’t make someone vote for a candidate they don’t want to vote for, meaning that you can’t fix the problem. Maybe Hutchinson just doesn’t understand RCV, but as a public figure speaking on behalf of RCV, he really should know what he’s talking about and not be claiming that RCV will produce majority winners.

    Lets say that Tim Penny was the candidate that won under RCV in 2002. Is it really a better outcome to elect someone who was the first choice of very few people, but the second choice of many? There have been several RCV elections where the candidate with the most first place votes didn’t win, and rather than bring consensus as RCV is supposed to do, the winning candidate was instead seen as illegimate by many people. After this happened in Burlington, Vermont, voters ended up repealing RCV. In Oakland, the RCV-elected mayor who trailed badly in first place votes, has been a complete disaster and there have been efforts to recall her. Aspen, Colorado and Tacoma, Washington both repealsed RCV after trying it, and there are efforts to do the same in both Oakland and San Francisco. The promsied benefits of RCV don’t actually work out when put into practice.

    • Submitted by Ed Day on 08/17/2012 - 07:58 pm.

      Good point on exhaustion; we need real runoffs

      While I figured that completing the slate could eliminate exhausted ballots and don’t think most people would have trouble ranking three pro pols like Pawlenty, Hatch, and Hutchinson, the thought of forcing everyone to put a check mark next to both Ralph Nader and Pat Buchanan makes your point in hilarious way.

      As for Penny, he was polling almost even with Moe and Pawlenty until the Wellstone plane crash, which seemed to drive everyone back to their base. (And Penny was the best, most experienced candidate imo). So yes, I think a electing a candidate with broad appeal would be a better outcome than a highly partisan candidate (which are most) who can’t even reach out enough to get half the vote.

      I think your earlier comments make a good case for head-to-head runoff elections. Though the prospect of a runoff would change the dynamic of the general election, here are a few runoffs that could have been interesting:

      Ventura vs. Coleman, 1998. Who do the Humphrey voters choose? (Ventura had 37 percent of the vote, losing in a landslide to NOT Ventura).
      Pawlenty vs Moe or Penny, 2002.
      Rep. Bachmann vs. Tinklenburg (IP canidate Anderson got about 10 percent of the vote — how
      many of these were simply protest votes. Tinklenburg was the IPs preferred candidate, but were not allowed to cross-endorse)

  6. Submitted by Dan Hintz on 08/20/2012 - 10:51 am.


    California and some other states have what is called a nonpartisan blanket primary. Everyone goes on the ballot for the first round, and the top two candidates – regardless of party affiliation – advance to the general election. This accomplishes a lot of what RCV is supposed to accomplish – forcing candidates to appeal to a broader electorate rather than just partisan primary voters – without the problems associated with RCV.

    As a Humphrey voter, I would have had a tough time with Coleman v. Ventura.

Leave a Reply