Nonprofit, nonpartisan journalism. Supported by readers.


Community Voices features opinion pieces from a wide variety of authors and perspectives. (Submission Guidelines)

Ranked choice voting would bring our system into the 21st century

Ranked choice voting is already used in many U.S. jurisdictions, including San Francisco, Minneapolis and St. Paul

Americans are proud – and rightfully so – of our democratic institutions and of choosing our leaders through free and open elections. While far from perfect, our history has led us to extend the vote to include all of our citizens and to promote the act of voting as a vitally important civic responsibility.

Jeanne Massey

What, then, are we to make of the following results from last week’s statewide primary election in Minnesota?

  • In the hotly contested 8th Congressional District, the Republican and Democratic candidates each moved on to the general election with the support of just 5 percent of the eligible voters in the district;
  • In St. Paul’s Senate District 67, the candidates to replace retiring Sen. John Harrington will face off in November after gathering the combined support of just 7 percent of the district’s registered electorate; and
  • In Minneapolis’ House District 59B, the voters’ choices for the fall election were set based on the preferences of 876 DFL voters and 98 Republicans, a mere 4.8 percent of the registered voters in the district.

Plurality winners are now commonplace

Unfortunately, these results are not unusual. Overall statewide voter turnout last week was just 9 percent, narrowly beating 2004’s record-low numbers. In race after race, our November electoral choices were constrained by the preferences of a very small number of voters who, more often than not, were themselves a minority of those casting votes. Plurality winners – those who garner less than 50 percent of the total votes cast – have become commonplace in both Minnesota’s primary and general elections. The last Minnesota governor elected with a majority of support from the citizenry, for example, was Arne Carlson in 1994.

The only way to square the circle of these results with the traditions and aspirations of our democracy is to conclude that our electoral system is broken.

Many of us believe that ranked choice voting – RCV for short – is one fix that is achievable and effective. Under RCV, voters rank candidates in order of preference (1st, 2nd, 3rd, etc.) and by doing so allow for an “instant runoff” process. In a single-seat election, if no candidate receives a majority (50 percent plus one) of first choices, the least popular candidates are eliminated and their ballots get reallocated to remaining candidates based on their voters’ next choices until one candidate earns a majority of support. It can also be used in multi-seat elections to expand representation to even more voters.

Functions like a traditional runoff

RCV functions like a traditional runoff but does its work in a single election. RCV eliminates the need for municipal nonpartisan primaries and rolls the winnowing process of the primary into the higher turnout general election.

In partisan elections at the state level, where primaries are conducted to decide which candidate from each party will advance to the general election, RCV is used in both stages – the primary and again in the November general election – to ensure winning candidates receive at least half the votes, without the expense and effort of a second or “runoff” election. This gives voters more choice in choosing candidates and eliminates the “wasted vote” or “spoiler” worries that stifle viable third-party candidates.

RCV is used in democracies around the world, such as Ireland and Australia, and in many U.S. communities – including Minneapolis and St. Paul – from coast to coast. RCV is popular with voters and has been shown to make elections more inclusive, participatory and representative

Candidates must make broad appeal

RCV also helps mitigate excessive partisanship; to win a single-seat RCV election, a candidate must win a majority – not just most – of the votes. As a result, candidates have to appeal beyond their traditional base to the broadest possible reach of voters. Knowing that the path to electoral success requires this sort of outreach encourages more civility and more focus on common ground, and less of the vitriol that’s been so corrosive to our political process these past several years.

RCV can’t be adopted in time for this fall’s election, but we can work to hasten the adoption of statewide RCV for future elections and to bring our voting system into the 21st century. The idea has increasing support from the public, from academics and other analysts and – most tellingly – from the candidates and elected officials who are on the front lines of our political debates and who know firsthand what the current broken system is costing us and what we need to do to make it better.

Jeanne Massey is executive director of FairVote Minnesota.


Write your reaction to this piece in Comments below. Or consider submitting your own Community Voices commentary; for information, email Susan Albright

Comments (7)

  1. Submitted by Hiram Foster on 08/21/2012 - 06:16 am.

    Mayor of Minneapolis

    I just go back to the last totally dispiritied mayoral race in Minneapolis, where RT Rybak was able to use ranked voting to coast to an easy unchallenged victory. Voting gimmicks don’t solve problems they merely hide them.

  2. Submitted by Ross Williams on 08/21/2012 - 08:09 am.

    Reality Check

    ” Under RCV, voters rank candidates in order of preference (1st, 2nd, 3rd, etc.) and by doing so allow for an “instant runoff” process.”

    The problem with an “instant” runoff is that it is based on what voters know at that instant. It means voters not only have to find the candidate they support, they need to know enough about all the other choices to determine which they prefer.

    Using the primary races as examples, if serious, shows the problem. If this had been a ranked vote, these elections would now be over. If it had been used in Missouri, its likely we would be having a discussion of Senator-elect Aiken’s views on rape, rather then Senate-candidate Aiken’s views.

    The purpose of primaries is to narrow the choices so that people can focus on evaluating a few candidates. It is simply unrealistic to expect voters to fully educate themselves about every candidate who puts their name up for an office and rank all of them in any informed way. Some of us remember Joe Donovan winning the DFL primary for Secretary of State a couple decades ago. The incumbent Joe Donovan had retired, but Joe Donovan the truck driver defeated the DFL endorsed candidate to replace him. Only on election night did the voters discover their mistake. With ranked choice voting, that would have been too late.

    The Missouri example of a Senator Aiken makes a nightmare of the dream that ranked choice voting requires appeal to moderates. Aiken would have finished second in a three way primary ahead of his Republican rival, but polls show he would have been the second choice for most of his rival’s supporters. Ranked choice voting would allow the most ideological extreme candidate to win without ever being subjected to the glaring contrast with an opposition candidate who appealed to moderates in their own party.

    Of course, ranked choice voting would change the election process so we don’t really know what the outcomes would be if these primaries had been general elections. Its likely more people would vote and the candidates would adapt their appeal to different voters. We don’t know what the unintended consequences would be. But the idealized version presented here is almost certainly not an accurate description.

  3. Submitted by Dan Hintz on 08/21/2012 - 09:53 am.

    Dishonest article

    The premise of this article is false: RCV elections do not eliminate plurality winners.

    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 re-allocated, Lee was up to 84,457 votes (43.4 percent). The second place candidate, John Avalos, 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. An exhausted ballot is one in which the the first chosen candidate has been eliminated, and the remaining choices (if any) have also been eliminated. Any ballot that did not include Lee or Avalos as one of the choices ended up being an exhausted ballot. If you look at just the votes to Lee and Avalos, Lee got more than 50 percent of the votes. But to call that a majority, you have to discard the nearly 30 percent of voters who did not name Lee or Avalos anywhere on their ballots. Its no different than looking at just the votes for Dayton and Emmer (and ignoring Horner’s votes) and saying that Dayton had a majority.

    The same thing happened in Oakland in 2010. After the votes were re-allocated, approximately 12 percent of the ballots were exhaused. The elected mayor, Jean Quan had 53,897 votes from 119,607 voters after reallocation, which is only 45 percent. The second place candidate, Don Perata, who actually had far more first place votes than Quan (35 percent to 24 percent) – got 51,720 votes after reallocation, or 43 percent. If you only count the votes that went to Quan or Perata, then Quan had a majority of those votes, but to make the claim that this was a majority winner election, you have to discard the ballots of the 12 percent of voters that did not choose Quan or Perata with any of their choices.

    The only way to fix the exhaustion problem is to force voters to rank every single candidate on the ballot. That would mean, for example, that if I lived in the 6th Congressional District, I would have to rank Michelle Bachmann (conservatives – substitute your own horror candidate) and that my vote could end up registering as a vote for her. If I don’t rank her, or any of the other choices, my ballot would be invalid. Obviously, you can’t force someone to vote for a candidate, even if just ranking them as a lower choice, so there is no way to fix the exhaustion problem.

    I realize that fixing the problem of plurality elections is one of the big selling points of RCV advocates, but RCV doesn’t fix that problem. I don’t know if Ms. Massey does not understand RCV very well or if she is not being truthful, but if you are asking people to change the way by which we vote, you should not be making false claims.

    • Submitted by Rachel Kahler on 08/21/2012 - 10:53 am.

      What an excellent response

      Your response definitely shows the flaws of RCV. I can still see the appeal of RCV, given the fact that the most radical (politically-speaking, not necessarily ideologically speaking) voters are the most likely to vote in the primaries. However, the answer to that is to get more voters voting or find some other way to narrow the field. Honestly, I think the answer is to bring voting to the voters rather than the other way around. I don’t have a solution that does that without problems, but I’m certain it can be done by someone with more experience in such matters.

  4. Submitted by Alec Timmerman on 08/21/2012 - 01:19 pm.

    Not really telling the whole story

    You mentioned 67 winner only getting 7% of the vote, making the implication that RCV would somehow fix that. With RCV Foung probably would have gotten 7.2% of the registered voters. It was a primary in August, RCV is not going re-adjust the entire solar calendar in order to generate excitement for a primary in August.

    It’s a good debate to have, but you diminish your case when you include superfluous information and imply your proposal would have an impact.

    Foung would have won even without Dimond

    So, assume all 806 of the Dimond votes were in play. Assume Humphrey would get a majority of the Dimond votes. To win, Humphry would have to get 526/806, or 65%.

    Humphry received 39% of the vote. To think he would have gotten 65% of Dimond’s votes is a pretty big stretch. It would be a 26% point gain over his performance in the general population.

  5. Submitted by Connie Sullivan on 08/21/2012 - 01:46 pm.

    RCV proponents seem still to be stating that the system will always bring winners with a majority of the votes cast. Not true (most winners do so with pluralities). They seem still to be claiming that the system will increase voter turnout. Not true. They seem incapable of dealing with discussions of why certain towns or cities have decided to go back to the old voting system after trying out, and finding huge holes in, the RCV system. They refuse to deal with the problem of too many candidates in the pool without the winnowing process a primary provides.

    There are some advantages to the ideal RCV situation. But advocates would win more friends and influence more people if the took reality, real experience, into account, and re-thought their absolutism and their disdain for any other system of voting.

Leave a Reply