26

In America, most elections are based on first-past-the-post voting, where whoever gets a plurality of the votes wins. For instance if candidate A gets 40% of the vote and candidates B and C get 30% of the vote, then candidate A wins, even if 60% of the electorate are steadfastly opposed to candidate A.

This problem can be solved by a system called ranked-choice voting, aka instant-runoff voting, where voters rank candidates from best to worst. Suppose there are six candidates. Then each voter ranks the candidates from 1 to 6. Then the voters's first-choice votes are all tallied up, and whichever candidate gets in 6th place is eliminated, and his voters are reallocated to whoever was their second choice. Once this reallocation is done, the 5th place candidate is eliminated, and this process is repeated until only one candidate is left. (And there are some rules on handling ties, just as there are in the first-past-the-post system.) For more information see here.

Ranked-choice voting has many advantages. It eliminates the concern that third party candidates act as spoilers. And if the Republican primaries had used ranked-choice voting, Trump wouldn't have won the nomination, since a majority of the electorate was against him, it's just that the anti-Trump vote was split.

But my question is, what arguments have been made against ranked-choice voting. It seems like an obviously superior system to me. The only downsides I can think of is that it may be too hard to understand for some voters, and there may be implementation costs. But it's been implemented in San Francisco and several other jurisdictions, and it doesn't seem to have caused any catastrophic problems.

EDIT: My question is not about the difficulties of switching to ranked-choice voting, but about what disadvantages (if any) there are of the system itself.

46

This problem can be solved by a system called ranked-choice voting, aka instant-runoff voting

First off, there are multiple voting systems based on ranking your choices. The system you're describing is just one example, and it's a pretty bad one, so it's frustrating that people refer to it as "ranked-choice voting", as if it's the only ranked system.

This system is more specifically referred to as "Instant-Runoff Voting". (Though taken literally, there are other systems that use instant runoff rounds, but "IRV" always means this particular system.)

Second, it has a number of problems:

IRV has a spoiler effect

You said:

It eliminates the concern that third party candidates act as spoilers.

It doesn't, though. It actually leads to two-party domination in every country it's adopted in.

It's true that voting honestly for a fringe candidate (Yellow) is safe, since they'll be eliminated immediately and your second choice will go to a mainstream candidate (Green). So IRV is better than plurality in this case (where Red would win even though a majority opposes Red):

Fringe Yellow candidate does not spoil the election for Green

However, when the third party candidate becomes more competitive (which is the whole point of adopting an alternative voting system), voting honestly for them takes away votes from your second choice, who will get eliminated first, and then your most-disliked candidate will win in the following round. If the third party hadn't run, your more-preferred candidate would have won, so the third party is acting as a spoiler, and in fact the effect is worse than in plurality (red bar extends further to the right):

Yellow moves closer to Green and splits the vote, allowing Red to win

Video illustration of this effect:

Video thumbnail
Favorite Betrayal in Plurality and Instant Runoff Voting

These effects have been modeled graphically on 1-dimensional and 2-dimensional political spaces, and produce bizarre win regions where the population moving toward a candidate causes them to lose, and vice versa. In this example, if the population's opinions shifted to the right, reducing support for Red and increasing support for Yellow, it would cause Red to win instead of Green!:

Yellow is slightly farther from Green, causing the win regions to become non-monotonic

In 2 dimensions, we can see IRV giving the election to Green even when the population is centered exactly on the Yellow candidate. There's also a bizarre Yellow island that's nowhere near the Yellow candidate, and not present in any of the other voting systems pictured.

Hare (IRV) creates non-monotonic win regions in 2D political space

Approval voting or Condorcet ranking both have results that make sense, with the candidate nearest to the population center winning:

Condorcet or Approval voting produces nearest-neighbor win regions

Animated 2D comparison of different systems:

Video thumbnail
Yee Animations 0.8

IRV excludes moderates

Similar to above, if there are two more-extreme partisan candidates and a moderate/compromise candidate in the middle, the compromise candidate is eliminated early (for not being anyone's first choice), even though they are the best representative of the population as a whole, and a more partisan candidate is elected instead. This is called the "center-squeeze effect".

This is essentially what happened in Burlington Vermont's 2009 election, which led to IRV being repealed. Montroll was "centrist" relative to the other major candidates, but was eliminated in the 4th round, and a more extreme candidate won, even though Montroll had a higher approval rating than the winner.

Here are the ideal win regions for 14 candidates, with each winning if the population center is nearest to them:

Voronoi diagram of 14-candidate election

Here's what happens in IRV. All the candidates near the center are eliminated, and only the more extreme fringe candidates can win:

Diagram of win regions of IRV 14-candidate election

These two effects lead to political polarization and two-party domination. You can see how IRV skews Australia's House in favor of the two main parties even though 1/4 to 1/3 of the population would prefer third parties (while STV leads to a more proportional Senate):

Bar graph of primary votes vs seats in Australia's House (IRV) and Senate (STV)

IRV is not a Condorcet system

In the Burlington election, Montroll was also the Condorcet winner, meaning he would have won against every other candidate in head-to-head elections. The overall preferences of the population were unambiguous: Montroll > Kiss > Wright > Smith > Simpson. But IRV is not a Condorcet system; it eliminated Montroll and elected the population's second favorite Kiss instead.

If your system doesn't elect the most-liked candidate (the "Utilitarian Winner"), it should at least elect the most-preferred candidate (the "Condorcet Winner"). IRV guarantees neither.

IRV enables tyranny of the majority

Here's a (contrived) example:

  • Candidate A is loved by 55% of the population, and hated by 45% of the population (55% overall approval rating)
  • Candidate B is liked by everyone (85% overall approval rating)

Under IRV, the polarizing Candidate A would win, because they are preferred by a majority, even though the population as a whole would be much happier with Candidate B winning the election.

"Utilitarian" voting systems like Score/Approval choose the candidate with the higher approval rating, which is considered a better outcome by advocates of these systems.

Another way of viewing this is that Candidate A is a great representative of half of the population, while Candidate B is a good representative of the entire population.

Majoritarian voting systems are not as inclusive, leading to adversarial politics, inefficiency, and even civil wars.

2016 election

And if the Republican primaries had used ranked-choice voting, Trump wouldn't have won the nomination, since a majority of the electorate was against him, it's just that the anti-Trump vote was split.

There's not much good data about this, but it's likely that IRV would still have elected Trump:

Score or Condorcet voting would likely have elected Sanders or Kasich, as they had the highest approval ratings:

  • 1
    Well, I think the point of elections is majority rule, and then you should use other mechanisms like checks and balances and the like to deal with the fact that there's a minority who doesn't agree with the majority. In any case, do you know of any voting system that improves on IRV in other ways but still makes A win over B in the example you gave at the end? – Keshav Srinivasan Jan 27 '17 at 6:46
  • 5
    @KeshavSrinivasan I'm just curious how you came to that conclusion. Isn't the purpose of elections to choose the candidate who best embodies the will of the people? Anyway, Condorcet systems are also majoritarian, so you might like those. Schulze is the only one I've seen talked about, so I assume it's superior to the others in some way, though I don't know the details. – endolith Jan 27 '17 at 15:26
  • 2
    @KeshavSrinivasan Score Runoff Voting uses Score for the first round, to choose the two candidates with highest approval ratings (replacing party primaries), then faces them off in an instant runoff where one will get a majority of the votes. So it's majoritarian in a sense, but only after eliminating the extremely polarizing candidates first. Simulations say it resists strategic voting better than pure Score, as well. – endolith Feb 6 '17 at 15:12
  • 2
    @blip There are many ranked-choice systems, but the phrase "ranked choice voting" is typically used to mean this system in particular, at least in the US. I guess taken literally, there are other voting systems that use "instant runoff" rounds, too, but likewise the term "IRV" almost only refers to this system. I'll edit to be more specific. Center squeeze does not happen in all ranked systems. Condorcet ranked systems like Schulze would have elected Montroll, for instance. – endolith Feb 7 '17 at 1:40
  • 3
    Where did you get (or how did you make) these figures? I love them. – indigochild Feb 7 '17 at 3:33
12

Trump

And if the Republican primaries had used ranked-choice voting, Trump wouldn't have won the nomination, since a majority of the electorate was against him, it's just that the anti-Trump vote was split.

Maybe. However, by the end of the primaries, Donald Trump was winning an absolute majority of the vote. Instant Runoff Voting (IRV) or plurality, an absolute majority is a winner. It's possible that ranked choice would have helped in the early primaries (and Trump would have never achieved the inevitability at the end), but it's by no means guaranteed.

If your goal is getting rid of Trump, then I would suggest getting rid of partisan primaries instead. That could have led to a candidate like John Kasich or Jim Webb winning.

IRV might have helped in the general election. Presumably Jill Stein voters preferred Gary Johnson and Hillary Clinton to Trump. And Clinton voters probably preferred Johnson and Stein to Trump. It's not clear how Johnson voters felt though. Libertarians have traditionally been split on second choice candidates. In any case, IRV is heavily dominated by the first choice. Since Trump and Clinton were the overwhelming first choices, it would have still been mainly between them.

Ranked choice in general

  1. Ranked choice ballots are more complicated to fill out. Instead of filling out a single choice, voters have to fill out a list.

  2. Ranked choice voting is harder to understand. It's not as obvious why one candidate wins over another.

  3. Ranked choice is harder to process. Instead of just counting votes, the system has to count ordered lists.

  4. Ranked choice doesn't fix problems with partisanship or oppression of minority groups in favor of majority opinion. But those are common complaints about the current system.

Note that these aren't insurmountable details. In particular, computers make the implementation difficulties much less important. But they are problems with every ranked choice method.

Also note that this doesn't include any of the arguments for ranked choice voting methods. This question only asks for arguments against. Balancing the competing arguments would be more opinion-based.

IRV in particular

Plurality is better than IRV by some criteria. In particular, monotonicity, consistency, and participation. Quick summary, there are tactical reasons to list different orders than one's actual preference.

IRV does not meet the Condorcet criterion. So in a high partisan environment, it tends to devolve into the same choices as plurality voting.

Of course, IRV is also better than plurality voting on some criteria. In fact, every voting system has some criterion on which another voting system is superior. In particular, the Condorcet criterion and Later No Harm are incompatible, so no voting system fulfills both.

  • 4
    Regarding using the 2016 Republican primaries as an example: 1. One of the biggest advantages of ranked voting is that tactical voting is far less necessary, so you can not just take results of a majority vote and assume people would have had the same 1st preferences if it had been a ranked vote. And 2., with ranked voting, primaries are no longer necessary, because a party has no disadvantage anymore by running multiple candidates. But the rest of this answer is a really good synopsis. – Philipp Jan 22 '17 at 12:25
  • @Philipp IRV/"RCV" does not have those benefits, though. It still has vote splitting and spoiler effects which require tactical voting. – endolith Feb 6 '17 at 16:32
  • While I like the 'Trump' part of the first, it's likely much fairer to say "fewer fringe candidates winning". IRV ultimately (in theory, at least) leads to a more centrist result as the 'less extreme option' tends to be shared by more people that the extreme options. – user1530 Feb 7 '17 at 0:33
  • Hmm...I need an analogy here. Without IRV, you could end up with pepperoni and licorice flavored ice cream for dessert. With IRV, you might end up with French Vanilla. Not everyone would choose French Vanilla if they only had one vote, but most people would agree French Vanilla is an OK compromise. :) – user1530 Feb 7 '17 at 0:35
  • @Brythan I think you make a lot of good realistic informative points, but the speculation of results of the last election if they had been held under IRV are not so useful - e.g., Bernie and others would probably have been candidates under IRV so your starting hypothesis about the candidates present is already wrong. – Craig Hicks Feb 8 '17 at 18:20
2

A. Big part of an election IS to show a mandate for the winner. To show a mandate, you need to have popularity. Or "absolute preference" for a candidate - " I like candidate XYZ too win".

Ranked voting shows only "relative preference" - "I like candidate XYZ more than ABC" but it could very well be that I dislike both.

Thus a winner of a ranked voting cannot assure the voters that he or she has the mandate to govern.

Edit: another way to look at this, the winner of a rank voting system is the least disliked candidate, while a regular voting produces the most popular winner.

  • suppose 3 cands a, b, c with ranked voting: // a > c > b = 50% // b > c > a = 50% // c > * > * = 0% // Cand "c" the extreme case of the least disliked but never most popular candidate. But c will be eliminated in the first round and it will be a toss up between a and b. // On each round IRV eliminates the candidate with least #1 votes in that round - that gives an intuitive explanation of why c, the compromise candidate, could not win. – Craig Hicks Feb 8 '17 at 15:41
  • 1
    i can make the example simpler for you. Say you have two candidates and two voting systems, one is popular vote and another is rank vote. Candidate A is declared a winder in a popular vote. you can infer from that, correctly, that Candidate A has the most support thus the mandate; – dannyf Feb 8 '17 at 15:49
  • 3
    I can give an example where "mandate" becomes questionable in your scenario. Two candidates, A and B, are the only ones running in the election. A has 25% support, B has 25% support, but 50% are completely dissatisfied with both candidates and do not vote. Either A or B will get elected, but 75% of the electorate are completely dissatisfied with either. Is that a "mandate"? – Craig Hicks Feb 8 '17 at 17:32
2

One of the problems with ranked choice is that it not show how much do you like or rate someone.

If some brothers are voting between an spider and a dog to be their pet, if an the arachnophonic kid vote "dog is better than spider", it will have the same meaning than a kid that likes both (but prefer dog) and vote "dog is better than spider".

  • 2
    This is more of a comment than a real answer. – Alexei Nov 20 '17 at 12:54
1

With situations you outlined, there are some bad assumptions you're making. First, in the scenario where A gets 40% of the vote and B and C get 30% of the vote apiece, you assert that 60% of the population does not approve of candidate A, despite his win. You get this number by adding B and C together. However, this does not mean that B and C attract equal opposition to A. Rather, it means that 60% of the electorate opposes A, but 70% opposes B and 70% opposes C. A first past the post system does not mean that the candidate is disliked by the electorate if he doesn't receive a vote, but rather he was not the first pick. C supporters may lean more towards A, but see C as the better candidate. However, if C dropped out, they still like A more than B. These metrics can't be accounted in a first pass the post system, which does not care about the votes not received, so long as the winner receive the most.

This is true with the assertion that Donald Trump would not win. All we know is that the support of the less successful candidates meant that Donald Trump was not their first pick, as opposed to #NeverTrump. You can only assume that Trump will get a portion of the support of people who still liked him over other candidates.

If we look at the original set up of a 40%/30%/30% split, Candidate A is still likely to win in a First Pass the Post because he only needs to secure 36% of C's support (11% of the total available voters) to win. Where as B would need to secure 70% of C's support (21% of the total available voters). Keep in mind that C's support does not mean disapproval of A or B. It could be B < A < C or A< B < C. In our real world model, You assume that Kasich Supporters would support Cruise before Trump, without taking into account that Kasiche supporters might have stronger feelings of opposition to Cruise than Trump.

So that's just with some math of yours not with your system.

With your system, lets add to the mix of canadates A, B, C the new comers D, and E. If I arrange my ballot so that C < B < A and never rate D and E because I cannot stand either of them (effectively never voting for them), what happens to my vote if A, B, and C are eliminated? Here my vote is never given to a any candidate because I did not choose either. To correct would you require to rate D and E? If so, what if I use my freedom of speech to still choose not to rate D and E and my ballot is thrown out because it is not complete? What if I do cast a vote for a winning candidate in my order, and still refuse to vote D or E... does that discount my vote?

  • 1
    You have a good and overlooked point here. Taking the 2016 election, and assuming that ranked voting replaced the primary, we would have 5 Democrats (not Sanders), 3 third party (Sanders, Johnson, Stein) and 17 Republicans. The GOP will win this election just by exhausting the field, assuming that no Democrats rank any Republicans. Also, the winner will not be Trump, Bush, Cruz or any top candidate, because those were all eliminated in early rounds. It will be somebody far down the list, like Santorum or Jindal, and the more Rs the Ds rank, the farther down the list of candidates we go. – Chris Strickland Dec 20 '17 at 19:38
0

Any election system is unfair if it has single-member districts. Fair representation can only be achieved with multiple-member districts.

Of course, executive governments are usually single-member. But parliamentary / congress systems with single-member districts can be easily abolished.

  • 1
    Can you make this more of an answer to this specific question? IRV is a single-winner election method, but it's used in multi-winner constituencies like Australia's House, which is a valid criticism. – endolith Nov 20 '17 at 15:45
-2

A prime argument against ranked choice voting is - there's nothing particularly wrong with the current voting system. Why change? What could that change bring that we don't have today?

Yes, we had a dismal 2016 presidential election, with two decidedly flawed candidates. Even the third party candidate sucked that year. The 2016 US election was a perfect storm of lousy choices.

That isn't the fault of the voting system. It's the fault of the parties for letting the situation devolve, and the fault of the voters for not demanding better.

Doesn't matter what voting system you have. As long as you have an electorate who is so easily distracted by smokescreens and irrelevant platforms, and doesn't focus on what is important for the future of the nation, you'll get poor results.

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.