3 ms·
RCV doesn't solve what is claimed in the article. It is interesting since countries that they give as examples are clear evidence of this. (Yes, I'm going to ad
by godelski 5y ago
RCV doesn't solve what is claimed in the article. It is interesting since countries that they give as examples are clear evidence of this. (Yes, I'm going to advocate cardinal systems like Approval, Range/Score, and STAR)
> More moderate candidates.
RCV doesn't encourage this at all. Australia is a clear example of this. In the house Liberal and Labor (the two major parties) control 85% of seats. The next major party (National) is part of a coalition with the Liberals and this sums to 95%. Things are only slightly better in the senate. Look at any country that uses IRC/RCV and you'll see a similar pattern. The system didn't encourage more parties and more moderates, Australia still has divided politics like America.
As to the more parties, I have an alternative explanation. They use a parliament, which uses proportionate representation. Don't buy it? Go look at countries with parliaments and don't have IRV and you'll find similar distributions. We also need to recognize that in America our parties are closer to coalitions in other countries out of the necessity. I'm not sure how anyone can look at someone like AOC or Sanders and think they are the same "party" (by other country norms) as Biden or Pelosi (similarly on the Republican side, though Trump caused a consolidation).
> ranked-choice voting can drastically reduce the possibility of spoilers.
Yes, but no. It reduces spoilers when a candidate is not going to win anyways (e.g. Jo Jorgensen spoiling Trump). But this does not prevent spoiling when candidates are similar (e.g. Bernie spoiling Biden), which is the specific type of spoiling that we are concerned about![0] This is known as the Favorite Betrayer[1] and is of grave concern if we want more parties (IRV/RCV fails this)
> More cost-effective __than other runoff elections__
The reason for the qualifier is because cardinal systems are extremely cost-effective. They have higher VSE than RCV (maximal VSE is from Condorcet methods, but not significantly higher than any cardinal system[2][3]), they scale better (you give candidates independent values instead of comparing, although you can rank if that's easier for you and you don't mess anything up), and you don't have to do any run-offs/mini-elections/rounds (STAR requires 2 rounds of voting max), which it isn't uncommon for IRV to cause many rounds of voting. This greatly increases complexity and reduces transparency (cardinal systems are trivial to calculate).
Why is transparency important? Just look at Arizona. Now imagine if the system was more complicated (no matter what side of the argument you're on there should be an argument for a clear and easy to calculate system for the winner. Multiple rounds of vote counting greatly increases complexity and chance for mistakes, which compound).
> Less negative campaigning.
There's no evidence for this. Australia's ads don't look that different from American ones. So I'm not sure what they are getting on about. There's plenty of attack ads.
I also want to plug Election Science[4] (I'm not affiliated), another non-partisan voting group (they previously advocated for STAR but have shifted to Approval with the explanation being that Approval is "good enough" and lower complexity/higher transparency)
[0] https://electionscience.org/library/the-spoiler-effect/ https://electionscience.org/library/the-spoiler-effect/ (see video for example: https://www.youtube.com/watch?v=JtKAScORevQ https://www.youtube.com/watch?v=JtKAScORevQ)
[1] https://electowiki.org/wiki/Favorite_betrayal_criterion https://electowiki.org/wiki/Favorite_betrayal_criterion (A voter can never get a worse result by expressing the maximum support for their favorite candidate)
[2] Voting methods animated (author is a HN user): https://www.youtube.com/watch?v=-4FXLQoLDBA https://www.youtube.com/watch?v=-4FXLQoLDBA
[3] VSE (which isn't the only metric btw. Especially compare bounds due to strategy): https://electionscience.github.io/vse-sim/VSE/ https://electionscience.github.io/vse-sim/VSE/
[4] https://electionscience.org/ https://electionscience.org/
- specialist 5y agore less negative campaigning. Thanks for the contrary evidence. It's an oft-cited benefit of public financing of campaigns. I now can't remember where I got the notion that RCV helps too.