Turn off your TV set, recycle that New York Times, and close your CNN browser tab, folks, ’cause none of them are reporting on the real election news of 2013: the undemocratic—and probably unconstitutional—Minneapolis municipal election!
The people of the world’s former flour-milling capital were duped into adopting a “ranked-choice” voting (RCV) system, a variant of instant-runoff voting (IRV). In RCV, voters rank their top three candidates (rather than all of them in IRV) from the entire pool. If no candidate receives an outright majority (as St. Paul’s incumbent mayor Chris Coleman did this year), then the candidate receiving the fewest votes is eliminated, and his or her votes are redistributed to their voters’ next choice(s). The process is repeated until a candidate wins a majority of votes.
Like many progressive responses to obvious problems, it’s the next worst solution—if not an even worse one—to plurality voting. (For the record, range voting, e.g. giving each candidate on the ballot a score from 0 to 4 with an affirmative “no opinion” option is the best voting system for single-member seats. The candidate with the highest average score wins. Special rules apply, such as minimum vote-share requirements for write-in candidates to prevent them from stealing elections.)
RCV and IRV suffer from numerous problems, some of which are novel to people used to plurality voting. For one, vote counts must be centralized and aren’t possible at the precinct or ward level—not impossible to accomplish but slower and more inefficient. The other problem is that IRV/RCV raises the number of spoiled ballots. If people rank the same candidate twice, for instance, their votes might be thrown out. Such errors don’t necessarily invalidate all overvotes, but it happens. Thus, IRV-type elections regularly result in more spoiled and erroneous ballots than plurality vote elections. Take that lizard people voters!

From the 2008 Minnesota U.S. Senate election, which resulted in a recount. Supposedly this person intended to vote for Franken, but his vote was thrown out as an overvote. Personally, I disagree because he didn’t circle the oval next to “Write-in, if any”.
The proponent of the new method is an outfit called FairVote Minnesota, which gives ten reasons to favor RCV over plurality voting (I think):
1). Upholds the principle of majority rule
Not this year! Here are Minneapolis’ 2013 results, sorted by party and then total votes:
NAME | Party | 1ST | 2ND | 3RD | Pct. Total (1st) | Pct. Total (2nd) | Pct. Total (3rd) |
---|---|---|---|---|---|---|---|
Captain Jack Sparrow | Count All Rankings | 264 | 583 | 1,440 | 0.3% | 0.8% | 2.4% |
Bob “Again” Carney | Demand Transit Revolution | 56 | 98 | 172 | 0.1% | 0.1% | 0.3% |
Betsy Hodges | DFL | 28,935 | 14,399 | 6,742 | 36.6% | 20.8% | 11.2% |
Mark Andrew | DFL | 19,584 | 12,757 | 8,977 | 24.8% | 18.4% | 14.9% |
Don Samuels | DFL | 8,335 | 14,170 | 11,178 | 10.5% | 20.5% | 18.5% |
Jackie Cherryhomes | DFL | 3,524 | 6,470 | 8,045 | 4.5% | 9.3% | 13.3% |
Bob Fine | DFL | 2,094 | 3,751 | 4,506 | 2.6% | 5.4% | 7.5% |
Stephanie Woodruff | DFL | 1,010 | 2,128 | 2,633 | 1.3% | 3.1% | 4.4% |
Alicia K. Bennett | DFL | 351 | 628 | 854 | 0.4% | 0.9% | 1.4% |
Jeffrey Alan Wagner | DFL | 164 | 292 | 375 | 0.2% | 0.4% | 0.6% |
Gregg A. Iverson | DFL | 144 | 215 | 437 | 0.2% | 0.3% | 0.7% |
Doug Mann | Green Party | 779 | 1,052 | 1,105 | 1.0% | 1.5% | 1.8% |
Rahn V. Workcuff | Independence | 65 | 118 | 162 | 0.1% | 0.2% | 0.3% |
Neal Baxter | Independent | 145 | 337 | 493 | 0.2% | 0.5% | 0.8% |
John Leslie Hartwig | Independent | 97 | 185 | 248 | 0.1% | 0.3% | 0.4% |
Cam Winton | Independent Responsible Inclusive | 7,511 | 3,751 | 3,974 | 9.5% | 5.4% | 6.6% |
Merrill Anderson | Jobs & Justice | 108 | 190 | 216 | 0.1% | 0.3% | 0.4% |
Dan Cohen | Jobs, Downtown Casino | 1,798 | 2,283 | 2,072 | 2.3% | 3.3% | 3.4% |
Bill Kahn | Last Minneapolis Mayor | 97 | 164 | 194 | 0.1% | 0.2% | 0.3% |
John Charles Wilson | Lauraist Communist | 37 | 66 | 112 | 0.0% | 0.1% | 0.2% |
Edmund Bernard Bruyere | Legacy-Next Generation | 70 | 59 | 83 | 0.1% | 0.1% | 0.1% |
Christopher Clark | Libertarian | 188 | 449 | 531 | 0.2% | 0.6% | 0.9% |
Christopher Robin Zimmerman | Libertarian | 170 | 435 | 422 | 0.2% | 0.6% | 0.7% |
Troy Benjegerdes | Local Energy/Food | 148 | 224 | 323 | 0.2% | 0.3% | 0.5% |
Kurtis W. Hanna | Pirate Party | 200 | 250 | 418 | 0.3% | 0.4% | 0.7% |
Ole Savior | Republican | 693 | 795 | 657 | 0.9% | 1.1% | 1.1% |
Mark V Anderson | Simplify Government | 975 | 1,035 | 905 | 1.2% | 1.5% | 1.5% |
Tony Lane | Socialist Workers Party | 219 | 265 | 397 | 0.3% | 0.4% | 0.7% |
Jaymie Kelly | Stop Foreclosures Now | 196 | 419 | 412 | 0.2% | 0.6% | 0.7% |
Abdul M Rahaman | We the People … | 338 | 363 | 367 | 0.4% | 0.5% | 0.6% |
James Everett | 347 | 550 | 806 | 0.4% | 0.8% | 1.3% | |
Mike Gould | 204 | 368 | 548 | 0.3% | 0.5% | 0.9% | |
Joshua Rea | 108 | 190 | 267 | 0.1% | 0.3% | 0.4% | |
James L. Stroud | 64 | 98 | 154 | 0.1% | 0.1% | 0.3% | |
Cyd Gorman | 39 | 96 | 136 | 0.0% | 0.1% | 0.2% | |
TOTAL | 79,057 | 69,233 | 60,361 | 100.0% | 100.0% | 100.0% |
(Source: StarTribune.com)
Although the purported winner is DFLer (that’s Democratic-Farmer-Labor for you Minnesota history dunces) Betsy Hodges, it’s savagely obvious that a majority of voters wanted someone other than her to be mayor. In fact, while DFL candidates received a de Blasionian 81.1 percent of the total vote, 55 percent of them still were against Hodges.
To say that many of these voters may’ve chosen Hodges as their second or third choices so their votes “really counted in the end” is condescending, semantic nonsense because it tells those voters that their first or second choices weren’t serious. She might have been the “Condorcet winner,” i.e. the candidate who would’ve beaten all other candidates in head-to-head matchups, but it doesn’t matter because RCV leaves that question in dispute. The 2013 Minneapolis mayoral election results blatantly discredit FairVote Minnesota’s claims that RCV upholds majority rule.
2). Eliminates “wasted” votes
Tell that to anyone whose three candidates were eliminated, a phenomenon that makes RCV even worse than IRV and plurality voting. Indeed, because some votes (probably) weren’t included in the runoffs, eliminated voters had no say in who was ultimately elected mayor. This is probably a denial of due process.
3). Solves the “spoiler” problem and gives voters more choice
Given that a majority of DFLers, to say nothing of the other 19 percent of voters, preferred someone other than Hodges, some of the candidates must have been spoilers ipso facto.
4). Increases voter participation
The StarTribune says turnout was higher in 2001.
5). Opens the political process to new voices
…Who are promptly eliminated.
Okay, I concede. Nine DFLers taking 81.1 percent of the vote is better than one.
6). Promotes more diverse representation
Sure, if you think minority, plurality candidates count as “diverse.”
7). Reduces negative campaigning and promotes civil, issue oriented campaigns
Or it promotes rancorous intraparty fights as candidates accuse one another of spoiling the election in favor of a plurality leader like Hodges.
8). Mitigates political polarization
What? Just a second ago we were getting diversity and now we’re getting moderation? What could be more polarizing than minority candidates winning and votes being eliminated before the final runoff?
9). Combines two elections in one so that voters only have to make one trip to the polls and taxpayers have to pay for only one election
Tell that to the people who picked for their first or second choices any of the candidates who lost to Hodges but weren’t eliminated. If, for example, DFLer Mark Andrew came in second, all 19,584 of his first-choice voters only got to vote once, i.e. their second and third choices were wasted votes, which we’re told don’t exist. Some people’s votes counted thrice, some once, and some not at all.
10). Reduces the cost of campaigning
Maybe, unless an internecine fight erupts over who’s spoiling the election and should drop out. Technically, RCV encourages “candidate cloning” by opposing parties. (Hey Mayor-elect Hodges, how much will you pay me to move to Minneapolis, throw down the $20 registration fee to run as another DFLer in the next election and split up the vote even further so you’ll win next time? Republicans? I’m totally a mercenary candidate.)
…
To be clear, none of this is to say that Hodges will make a poor mayor, or that she was an unpopular candidate, but she is not the person most Minneapolitans voted for via RCV. This contradiction is so self-evident that it lacks the subtlety to qualify as Orwellian.
Minneapolis: Where “one person, one vote” and majority rule crawl into a plastic bag to die.
**********
And while the LSTB is in Minneapolis, my favorite question from the Star Tribune‘s meet-the-30-candidates article is “Ideas to lower property taxes.” Because, why make Beacon Investment Properties LLC, owner of the 57-story IDS Center and Starwood Hotels and Resorts Worldwide, Inc., owner of the 32-story Foshay Tower pay for the never-won-a-Superbowl Viking’s new stadium when the city can just confiscate poor people’s wages with another sales tax hike? How else are wealthy, foreign, feudal land speculators supposed to make a living? Their own labor? Heavens no.
A couple of questions:
1. Why RCV is “probably unconstitutional”?
2. What is your definition of majority rule?
1. “One person one vote” means voters should be able to vote in the dispositive runoff. RCV results in “exhausted” ballots, which prevents this.
2. GFE.
1. As you know, the Dudum case explicitly rejected that argument. Try again or explain why your think the Dudum case reasoning reached a faulty conclusion about constitutionality.
I’m sure you’re aware that a young lawyer who could articulate a solid, credible argument for the unconstitutionality of RCV could make quite a name for himself/herself. Give it your best shot, citations and all.
2A. The definition for majority rule that you linked to was: “A doctrine by which a numerical majority of an organized group holds the power to make decisions binding on all in the group.” That’s rather vague as it applies to elections. Can we start by agreeing that in the context of public governmental elections “majority” or “numerical majority” means “more than half of the number people”?
2B. Since the decision for who is mayor of Minneapolis is binding on all residents of Minneapolis, wouldn’t your definition of majority rule require that more than 50% of Minneapolis residents be required to choose the mayor?
1. As you know, I am not obligated to agree with the Ninth Circuit on the constitutionality of RCV any more than I am obligated to agree with the Supreme Court that voter ID laws aren’t unconstitutional poll taxes. You must undoubtedly also know that Minnesota is in the Eighth Circuit, not the Ninth, so the Dudum opinion isn’t mandatory authority there.
I’m sure you’re aware that an anonymous blog commenter who has basic reading comprehension skills could tell the post’s topic was not the constitutionality of RCV, as evidenced by the fact that the constitutionality statement appears only once in an em-dashed aside. Thus, as the author I’m not rhetorically required to defend that opinion any more than I already have. Give it your best shot: reread the post; understand its arguments; and comment on the substance and not the editorials.
2A. There’s nothing vague about the definition I linked to because there is only one definition of majority rule, whether for elections or legislation.
2B. Again, I don’t have an independent definition of majority rule. Rather, it is FairVote Minnesota that states on its Web site that RCV elections that result in plurality outcomes constitute majority rule.
Note also that plurality voting does not always enact majority rule, but it doesn’t claim to, hence the “plurality.” Now, I suppose you could make the argument that Minneapolitans accept by majority rule that the plurality winner of an RCV election shall be sworn in as mayor, but FairVote Minnesota isn’t making that argument. Instead, it’s claiming that RCV results in the election of the candidate that a majority of voters wanted, but this didn’t happen in Minneapolis in 2013, ergo FairVote Minnesota is misrepresenting the merits of RCV.
1A. I’m surprised that as a blogging lawyer you’ve come up empty so fast. I’m not so interested in who you agree or disagree with, so much as your lawyerly legal reasoning behind your opinions on legal matters. By the way, you mention this issue in your opening paragraph, your closing paragraph, and your analysis of FVM item 2.
2A. I’ll accept that response as an agreement with me.
2B. RCV ensures that the winner has a majority of the votes in the final round. That’s what happened in the Minneapolis mayor’s race. Hodges won with about 61% of the vote in the final round with the top two candidates. I know that may not be your interpretation of majority rule, but it is all a delayed runoff between just two candidates can guarantee. My understanding is that is the basis for the FairVote Minnesota claim. Even reasonable dogmatists may sometimes disagree.
I’m surprised that as a hostile, condescending, anonymous Internet troll you think you can shame me into enthusiastically writing legal memoranda for you on my own time. (Nice try, though.) But because I’m an honest person, particularly for the sake of others reading this discussion, I will grant you that the due process point appears elsewhere in this article. However, they are still editorials and immaterial to the article’s points.
This dialogue is over. In the future, if you want to know blogging lawyers’ “lawyerly legal reasoning behind legal matters” (wow, you’re a bad writer!), simply state your opinions and curb the arrogance.
Your subsequent comments will not be posted.
The phrase “one person, one vote” has nothing to do with the way votes are cast and counted. See Reynolds v. Sims (377 U.S. 533)
http://www.electology.org/approval-voting#Does_Approval_Voting_violate_one_person_one_vote
Most of your criticisms of Instant Runoff Voting are correct, but you left out a lot of much more powerful criticisms.
http://scorevoting.net/CompleteIdioticIRV.html
Clay Shentrup
Co-founder, The Center for Election Science
Thank you. scorevoting.net is an awesome Web site.
Ron G,
> RCV ensures that the winner has a majority of the votes in the final round.
So does the following system:
1) Voters vote for one candidate.
2) All candidates are eliminated, except for the one with the most votes.
3) That one remaining un-eliminated candidate is the winner.
That system insures that the winner has not just a majority of the votes in the final round, but *100%* of the votes.
It’s also identical to Plurality Voting. Oops.
It’s even possible for IRV to elect X instead of Y, even though Y was preferred to X by a majority of voters *and* got more first-place votes.
http://www.electology.org/core-support
Mathematically speaking, IRV only guarantees that the winner is preferred by a majority to *at least one other candidate*—and that’s assuming voters rank all the candidates. If voters truncate their rankings, it’s possible for IRV to elect the Condorcet loser.
It would be nice if IRV proponents had any education in election systems.