MENU

Reason.com

Free Minds & Free Markets

Maine’s Primary a Pioneering Experiment in Ranked-Choice Voting

Voters participate in first use of a candidate rating system for state races in the U.S.

Ranked votePeter Mautsch / Maranso GmbhKeep an eye on Maine's gubernatorial primary today not to see who wins (because we might not actually have a winner declared this evening) but to see how the winners are selected.

Today's primary marks the launch of Maine's use of ranked-choice voting for determining the winners of several state-level elected offices. It will be used to decide which Democrat and which Republican will face off in November (along with several independent and third-party candidates).

In ranked-choice voting, rather than just picking a single winner, voters can rank their preferences. In order to win a ranked-choice vote, the top candidate must get a majority of the vote among all the candidates. Getting a plurality is not enough—it has to be more than 50 percent. If no candidate gets a majority, the candidate with the least votes is ejected. Then the votes are tallied again, but the votes from those who went to the ejected candidate now go to their second-ranked choice. And so it goes until one candidate claims more than 50 percent of the vote. That candidate is the winner.

There are four Republicans and seven Democrats duking it out for the governor's race (incumbent Republican Gov. Paul LePage is getting term-limited out), so determining the outcome may get pretty complex.

A handful of cities use ranked-choice voting for local elections (San Francisco just used it to determine its mayor). Maine is the first place to implement it for state races. In addition to the governor's race, one congressional primary on the Democratic side and one state lawmaker race on the Republican side have enough candidates to call for ranked-choice voting.

Voters approved ranked-choice voting in a ballot initiative in 2016, but there's been a fight with state Republicans over implementing it (LePage didn't get a majority vote in either of his elections). The state Constitution specifically requires only a plurality of the vote to determine the winner. The state's Supreme Judicial Court warned in an advisory opinion that a constitutional amendment would be necessary or the outcomes of ranked-choice voting could potentially lose a court challenge. There have been both legal and legislative fights. A judge ruled in March to allow ranked-choice voting for this primary, but its future in Maine is still muddled. To try to resolve the conflict, also on the ballot today in Maine is an initiative that would overrule a lawmaker's attempt to delay implementation of ranked-choice voting and potentially repeal it if the state's constitution is not amended. So essentially Maine voters today have to also approve ranked-choice voting again or potentially lose it.

The case for ranked-choice voting is that it's an alternative to our more common winner-takes-all system that sometimes pushes voters to feel like they have no choice but to support the two major parties or discourages them from voting entirely. It encourages third-party and independent votes because you're no longer "throwing your vote away" by voting Libertarian, or Green, or any other party. You can rank a Libertarian Party candidate first, then pick a Democrat or a Republican that comes closest to your belief system as your second choice. If the Libertarian Party candidate performs poorly, your vote wasn't a wasted effort. It also means that if lots of people rank a Libertarian candidate as their second choice behind a Democrat or a Republican but the major party candidate doesn't get a majority of the vote, the Libertarian has a better chance of coming out on top than he or she would ever have had in a traditional election.

So ranked-choice voting is popular among those who want to encourage alternatives in races and to break the stranglehold by the two major parties and by those with the most party influence. The Washington Post interviewed a taxi driver in Maine who thinks LePage would never have been elected governor in the first place if the state had ranked-choice voting.

The Washington Post notes that the Republican candidates are still resisting ranked-choice while the Democrats are embracing it (two Democratic candidates are collaborating and encouraging voters to select them both, similar to what happened in San Francisco's mayoral race). The Republican Party in Maine says it thinks ranked-choice voting is "confusing" and will lead to lower voter turnout.

The data doesn't back them up there. FairVote, an organization devoted to pushing ranked-choice voting, observes:

Voter turnout in cities that have adopted RCV is comparable to, and often higher than, turnout in other cities. In elections using RCV in the Bay Area in 2014, voter turnout decline was less than in other parts of the state and voter turnout was generally higher than past non-RCV elections.

Professor David Kimball, at the University of Missouri-St. Louis, has studied voter turnout under RCV. His study finds that RCV in American local elections has a limited impact on turnout, with more important influences on turnout including a competitive mayoral election, other races on the ballot (including initiatives) and the use of even year elections. However, Professor Kimball's study shows that, when compared to the primary and runoff elections they replace, RCV general elections are associated with a 10 point increase in voter turnout.

California's overall voter turnout in last week's primary election was miserable. It's not finalized as yet, but the current participation rate based on county reporting is a mere 27 percent of all registered voters. But in San Francisco County, the participation rate jumps all the way up to 48 percent. Only a couple of other counties had a higher participation rate, and some of them are lightly populated areas like Sierra County that have a tiny pool of voters.

Ultimately, once Maine votes today, the question will be whether voters turn up and whether they're satisfied with the outcome, even if their preferred initial candidate lost.

An update on the ranked-choice mayoral race in San Francisco: Last week it appeared that ranked-choice voting may lead to the candidate who came in second in the first round of votes, Mark Leno, overtaking frontrunner London Breed as the lower-ranking candidates were eliminated. Now that more votes have been tallied, however, Breed has retaken the lead from Leno. But her lead is only 1,600 votes and there are still more than 17,000 ballots to tally.

Photo Credit: Peter Mautsch / Maranso Gmbh

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  • Rhywun||

    there are still more than 17,000 ballots to tally

    I honestly don't understand how this is possible, unless they're using stone-age technology or something. It seems to be giving people the impression that the voting method is "complicated" (it's not - a computer program can give you the winner instantly).

  • Scott S.||

    Most of them are provisional ballots. It's not the tabulating tech that's the problem. It's pretty much everything else about casting the vote in the first place that's the problem.

  • gormadoc||

    "Tallying mail votes is a days-long process in San Francisco because each ballot is split into four separate cards, each of which has to be removed from a sealed envelope and hand-fed into one of the city's four vote-counting machines."

    Also, it looks like 14000 are provisional ballots and weren't to be counted yet anyway.

  • Ron||

    provisional ballots ned to be verified as being by a legal voter, that could take some time.

  • DarrenM||

    I notice the picture indicates you can have four votes. Would it be worth reducing this to two or three? What is the optimum?

  • June Genis||

    Actually the ideal is to be able to rank all the candidates if you want to. There is never a requirement to rank any specific number. The fewer the number that people rank the greater the chance become of running out of votes to transfer. So that potentially, especially in a crowded and diverse field, the winner in the final round might end up winning by a plurality rather than a majority if more than two candidates remain in that round.

  • Eidde||

    "(incumbent Republican Gov. Paul LePage is getting term-limited out)"

    Don't worry, he'll still be able to have a great career as a Stephen King villain.

  • Just Say'n||

    Illinois had a better system that they abandoned sometime in the early 80's. Cumulative voting where two representatives are chosen per district. Seems a lot less complicated and allows vocal minorities to have a say in their representatives

  • Eric L||

    It was three members per district, IIRC.

  • JFree||

    Not much point since they had 59 'districts' with 3 critters each -- where the duopoly both colluded to only run two candidates per candidates - and they then moved to a system of 118 districts with one critter each.

    That entire thing shows the parties themselves were gaming the system so the individual voter is only given the appearance of being important.

    Much better to have 500 districts with no election at all and one randomly chosen critter from each. None of those critters would owe a damn thing to parties even if they 'tend to agree' with one or the other. And with 500 critters, there's a lot more diversity of opinion - and inability to effectively corral it.

  • BestUsedCarSales||

    Have you ever read The Napoleon of Notting Hill by GK Chesterton?

  • JFree||

    Don't think I've ever read anything by him. Is it worth it?

  • Scarecrow Repair & Chippering||

    I'd make it three elected members just to break up the two-party bias, and each would proxy however many votes they received in that election, so tallying votes in the legislature would take longer, but so what?

    I would also add a checkbox to every ballot for a volunteer legislator. Come election night, you choose one volunteer at random to be a fourth legislator, who would proxy all remaining votes.

    This would encourage a lot more voters because their votes really would count in some fashion, and the volunteer legislator would really shake things up in the legislature.

  • BigChiefWahoo||

    The old system in Illinois gave each voter 3 votes to cast for two legislators. All three could be cast for one candidate or split between two candidates. I don't remember anyone crediting that system with increasing voter turnout. The state constitution was changed to the current system in 1978 taking effect in 1980. Pat Quinn, who later became governor, advocated the change and a lot of people went for it, believing that reducing the size of the legislature would save the state money. Ha!

  • Diane Reynolds (Paul.)||

    I'm highly suspicious of these experimental voting systems. Not because the systems we have now are perfect, or even good, but because they seem to me to just add new problems to the pile we already have.

    "Democracy vouchers" here in Seattle was an excellent example of that. It immediately became a corrupt process with powerful incumbents gaming the system leading to predictable results. But hey, the paper dug through every nook and cranny and found someone on a street corner who said they were able to support a candidate they normally might not have, so great success I guess.

  • JFree||

    Totally agree. People are basically stupid and lazy when it comes to voting - and there's nothing wrong with that at all. These systems just seem to force more work on voters with the notion that that will produce a better choice. When the only real impact imo is to enable a tiny minority of political junkies who have too much time on their hands and are spending way too much time parsing thisthatother.

    I still say the only 'voting system' that can work is ending 'elections' altogether. Instead create the notion of 'critter duty' just like we do 'jury duty'. And randomly select citizens to be critters.

  • Cy||

    I like that idea. Make them pass a fill in the blank test on the constitution first and then cut them loose.

  • SRoach||

    I do not like this idea at all. The opportunity for a group in power to influence the outcome of the legislation by ensuring only people who agree with their, special, interpretation of the constitution would be too high.
    It could make gerrymandering look tame.

    [*]I recently read a piece on how the AMA's new test would have questions, not on medicine, but on hot-button political topics. ..I can no longer find the article via google.
    [*]It's been conventional wisdom for decades that our schools are leaning further and further to the left. Wildly liberal tenured professors are generally implicated in this trend, (and in at least one anecdotal case that strikes close to home.)
    [*]In the 19th century, entrenched interests in the reconstruction south tried to create "poll tests" to prevent blacks from voting. A common restriction was literacy, which makes sense, since an informed voter would generally be a literate one, ..until you remember that it had been illegal to teach slaves to read...and until you discover that the law exempted anyone whose grandfather could vote, even if he, himself, was illiterate, (as I recall in my reading, the original "grandfather" exemption.)

    Do you REALLY think it makes sense to give ANYONE the power to decide which ideas are legitimate to sit on a legislature?

  • Azathoth!!||

    The opportunity for a group in power to influence the outcome of the legislation by ensuring only people who agree with their, special, interpretation of the constitution would be too high.

    Not if one were intelligent enough to make the test about the text and not anyone's interpretation.

  • SRoach||

    Did you read my examples of people deliberately subverting a test, or coursework, for political gain?
    Many, if not most, of these people were quite intelligent, and bent their intelligence toward subverting a system toward their perceived advantage. Liberalizing future doctors, liberalizing the educated, (and presumed elite,) and disenfranchising a recently enfranchised group whom they wanted to keep down.

    Now, if you're intelligent enough to prevent that, please say so, because, as has become glaringly obvious over the last two hundred plus years, even our esteemed founding fathers couldn't think of every dodge to their carefully crafted form of government, or we wouldn't see the Federal government stepping on states over guns and drugs (that aren't for sale across state lines).

  • JFree||

    Here's the actual Louisiana literacy test for voters in 1964

    Just pure nonsense and timed trickery intended to ensure everyone who takes it fails. And all completely textual.

  • vek||

    Seriously, don't care. IMO voters should have to pass a literacy, history, and civics test. The founders were smart enough to try to come up with some limits on how dumb a person could be to vote. They wanted to ensure total morons and dead beats couldn't, which is a wise choice.

    They chose white, male, landowners. That wouldn't fly politically today for obvious reasons, and it probably isn't a great way to weed out idiots either... But some sort of basic testing that is based on OBJECTIVE facts, not subjective BS, would be fine. If you don't know what year we declared independence from the UK you shouldn't be able to vote. If you don't know what amendment guarantees free speech, you shouldn't be able to vote. If you can't name the 3 branches of government, you shouldn't be able to vote. And so on.

    People WOULD wrangle over things, but ANY standards would be better than having NO standards like we do now. The fact that the unwashed masses are allowed to vote is half the problem with things in this country today. I'm not saying it to be a dick or whatever, but it's true. Idiots have no place deciding how to run the richest nation of earth. The founders knew better, and do gooders after them screwed it up.

  • SRoach||

    While the old law requiring you to be a landowner was well-reasoned, it does tend toward consolidation of power. How do you prevent dilution of your power? Declare all other lands state lands, and quit giving out deeds. Since the landed are the ones with the power, they have a perverse incentive to regulate land ownership to favor themselves.

    That said, we've not done it that way for a long time.

    My problem with a poll test, with any test to exercise basic rights and political obligations, is that it WOULD become a vehicle for shaping future voters to look more like those currently in power. In this climate, it would be used to make sure future voters were sufficiently liberal rather than sufficiently civic minded or sufficiently intelligent.
    The date of the Declaration of independence might not make the cut in such a test, but the date women finally got the vote would. George Washington might not make the cut, but Rosa Parks would.
    And those would be the EASY questions. (Another question would ask you if True/False, is Marriage a union between a man and a woman. Yet another if decent medical care is a Right to be granted or a Privilege to be bought.)

    Perhaps you should be happy with the system we have, then. A test of apathy is a pretty decent poll test. If you're too apathetic to drag yourself to a polling place that Tuesday, or at least take the separate steps to acquire an absentee ballot, you've proven you're too apathetic and don't get to vote.

  • vek||

    The founders system wasn't perfect, but it was an attempt at SOMETHING. It does consolidate power, but that was the point. To not let morons have power. It's not a "fair" or egalitarian goal, but it is an intelligent one.

    I get what you're saying. 100% do. And it's a real danger. But IF one could create a bi/multi-partisan commission to come up with the test it might not be too bad. And frankly being knowledgeable about lefty crap is still going to weed out dumb people. I know more about leftist history than most of the dumb lefties voting. So even if slanted having a challenging test might still produce better results, because brains is brains.

    I'm quite sure that if you issued a simple IQ test, and set the requirement at 115 (1 standard deviation) to vote, you would get faaar better opinions from that group. There would be left and right leaners with high IQs. But the leftists would probably be saner, or at least more pragmatic. The rightists would likely be the same. Being smart in and of itself tends to lead to better decisions on average when one is looking at large groups of people. Even setting it at 100 to only exclude the lower 50% of people would be an improvement.

  • vek||

    Really it's a question of balancing so called fairness with the best real outcomes. In reality women getting the vote has done more to push the country left than any other single thing. Women are more empathetic and emotional. That's just a fact. They vote for bleeding heart things at vastly higher numbers than men. If only men could vote still the whole political spectrum would be sooo far right/libertarian of where it is today it would be ridiculous.

    Now would it be fair to not give women a say? Maybe not. But it would produce a stronger, more successful, wealthier, freer society. If we ever have a total collapse and taking away womens right to vote is on the table I'm all for it. Not because I hate women, or think there aren't smart women out there (there are!),but simply because it would produce better real world results. Fairness is a luxury item, making a functional society often includes not being nice or fair about some things.

    Maybe I'm an asshole, but that's kind of my whole thinking on voting. Some people simply consistently have wrong opinions, so why let them vote? The founders didn't. That may have made them assholes too, but they were assholes who created a very functional society.

  • SRoach||

    You're describing Sortition.
    I don't know if I want a handful of random people, who have previously had no motive to learn the law, deciding laws.
    I think we'd get a bunch of people tweeting their latest wild hair idea, and changing policy on a dime.
    I also think such a group would be more vulnerable to lobbying then would a group of pre-informed representatives, because the ONLY information they had on the issue would be that provided by the lobbyist.

    Have you ever read "The Probability Broach"? I can no longer link to the webcomic version of it, as it's been pulled down, (send Mr. Bieser rent and food money,) but the system of appointed, at large representation looks pretty good...even it it would theoretically allow for a congress of 325.7 million trying to assemble all under one roof.

  • Diane Reynolds (Paul.)||

    My preferred system is you announce an election, and arrest anyone who signs up to be a candidate.

  • SRoach||

    I laugh...but I think Russia has done something like that.

  • vek||

    LOL Great idea!

  • JFree||

    I don't know if I want a handful of random people, who have previously had no motive to learn the law, deciding laws.

    I prefer laws written so normal people can understand them - written by people who know they will soon merely be subject to them.

    more vulnerable to lobbying

    Depends on whether you think lobbyists succeed by imparting knowledge or campaign money. No election = no campaigning either. At any rate - a much larger assembly is what jacks up the cost and lowers the effectiveness of lobbying.

  • SRoach||

    Both.
    And also making friends.
    Further, you forget that many a congresscritter has left office only to be employed by the very industries they were legislating for and against.

  • JFree||

    OK. And your solution is - what?

    You're kind of reinforcing my solution. Doesn't cost much to give a couple of bogus jobs to ex-critters after they've done your bidding for many years because they won elections repeatedly. Quite different to give 500 jobs every couple years to ex-critters who were only there for a couple years.

  • DarrenM||

    Lawmakers are dependent on their staff anyway. The staff would probably stick around while the legislators changed, which would provide stability plus hopefully prevent the more lamebrain ideas from seeing the light of day.

  • SRoach||

    We already have bureaucrats crafting law.
    We have "burrowed" political appointees doing everything they can to interfere with the stated policies of their branch head and continue the policies of the former branch head.
    How is this different? How is it better?

  • Bubba Jones||

    My poli friends hate these systems because the voters do stupid things. Voters think it's a point based system so they put their second choice last, because they are trying to outsmart the vote count.

  • SRoach||

    I'm...not seeing the downside.

  • Bubba Jones||

    The end result is that the collective 3rd choice gets elected instead of the 2nd.

    Which is a positive if your first is everyone else's 3rd.

  • SRoach||

    It also implies that people who can't, or won't, read the instructions invalidate their own vote.

  • Eric L||

    This type of voting was used in many cities in Ohio for local elections up until the 1960's. A big reason it was abandoned was abandoned because it made it difficult for the Dem and Reps to control who got elected. They system made it much easier for voters to elect independents, small parties and candidates of the two major parties who were not beholden to the party bosses.

    I have belong to several organizations, including a church I used to attend, that used this type of system to elect people to their respective governing boards. I never found it difficult, and was fine with the results even if my first choice candidate did not get elected, but maybe that's just me.

  • Diane Reynolds (Paul.)||

    Some systems might work well where membership is voluntary. I think a problem arises when we're talking about politics where everyone is subject to the results whether they like them or not.

  • Ron||

    I agree with Diane with the last presidential election as an example we may have ended up with either a green party president which is just further left of democrats or the libertarian candidate and my opinion of the libertarian candidates of last election was pretty low, they were far worse than Trump. I for one would probably not give a second choice. thats not smart but considering our choices last election i almost didn't vote except to keep Hillary out. Considering I'm in California my vote didn't count and I could of had the honor of not voting for the rubbish choices we were given.

  • Mickey Rat||

    Of course the flaw in all this is the assumption that enough voters will have ranked candidates in such a pattern that you will get to 50% plus one. It assumes that there are candidates that voters deem to unacceptable to rank.

    In the end, it is a gimmick, one that encourages gamesmanship by political operators.

  • NoVaNick||

    This system won't work at all in any districts where a single party has a large registration and funding advantage. In other words, its a complete joke to have it in deep blue places like SF, or deep red places like Utah. Maine may be more balanced, but the dems have a funding advantage this year because of Trump, so they can load up the ballot.

  • Jgalt1975||

    How are you defining "won't work"? In a district where one party has a large registration and funding advantage, that party is likely to win regardless of the voting system you use.

  • Happy Chandler||

    It seems to be working in the SF mayoral race. There are races that don't include D vs R.

  • DarrenM||

    You could eliminate primaries altogether and have any number of candidates in the general. In someplace like San Francisco, it would still be one Democrat facing off against another Democrat in the end.

  • sarcasmic||

    Oh, shit! That reminds me, I've got to scoot down to the town office and vote in favor of repealing this abomination.

  • xsnake||

    "One man, one vote." Hmmmmm now the voter gets multiple votes. Doesn't seem legal.
    But hey.....the collective is goosey for it. If Marx is happy, what could be wrong?

  • Robert||

    Missing from the Supreme Judicial Court's advisory opinion is any discussion of whether Maine's constitution, when it refers to the election of someone to a gov't office, includes primary elections, which are only for party nominations. It seems to me it would not, although there may be case law saying the opposite. The statute adopted RCV for primaries as well as gen'l elections, so it seems the primaries have to be by RCV since the constitution doesn't specify.

  • DarrenM||

    I've long thought this made more sense. Our current system allows someone with a minority view to win with a plurality. (I believe Hitler was elected Chancellor with only about 40% of the vote at the time. What would have been the result if Germany had RCV?)

  • chemjeff radical individualist||

    More of these voting experiments, please!

  • New Cassandra||

    The system of voting you are describing is called Preferencial Voting and has been the standard in Australia for many, many years. It is not multiple votes, it is ensuring the least unpopular candidate is elected.
    I'm currently on vacation in Canada and watched the Ontario election last week. Under your simplistic First Past the Post system, the Conservatives were elected in a landslide but they only got 40% of the overall vote. How is that fair when 60% of people did not want them?
    Preferencial voting solves this.

  • vek||

    I don't think this would be some magical fix all... But I think it is a better system. It would almost surely give 3rd party candidates, both Libertarians and idiots like the Greens, a lot better odds of winning. That's a good thing.

    One of the things I have always liked about most foreign parliamentary systems is that they have a lot of 3rd party people in their parliaments. That is all literally JUST because of the mechanics of how they count votes and apportion seats. We'd probably have actual Big L Libertarians making up 10% of congress in the US if we did things differently. Same probably for Greens. But either way it is more fair than our system completely rigged in favor of Rs and Ds.

  • BILKER||

    yeah, yeah, that's the ticket. RCV esp in kommiefornia and new yawk will, with their huge advantage in democrat/socialist/progressive voters ensures that the only persons elected to any office will be democrat/socialist/progressives. The RCV would be a viable alternative as long as open primaries are absolutely banned. That might give other parties a slim chance to, at least, get a candidate past the primaries. Coxs' "victory" in commiefornia was surprising. the other candidates , Newsome and Vilar, who ran their cities into the ground before finally being termed out.

  • Walt Peterson||

    In 2012. Libertarian candidate Dan Fishman in Mass 6th district deprived the Republicans of ousting the corrupti and unpopular John Tierney. Now, in 2016, he may do same in his race to unseat the incompetent Democratic State Auditor, Suzanne Bump. The Republicans would be wise to support ranked choice voting in Massachusetts.

GET REASON MAGAZINE

Get Reason's print or digital edition before it’s posted online