Voting reform seems overrated
Key points
A shift from first-past-the-post to PR shifts party competition from being within parties to being between parties. It is not clear that this is good.
Under PR you get more parties. This includes far right parties. There is a significant advantage in future electoral competition once you’ve become an established party.
Coalitions have the problem of it being hard for voters to hold specific parties or leaders to account for policies.
Coalitions create a free rider problem because the reputation of the coalition government is a public good.
The median voter theorem no longer holds. If voters have extremist tendencies there is no competitive pressure not to serve them
Italian and French politics look not good. Having non FPTP seems plausible to have played a part in this in that it allowed Front National to gain a significant foothold in France while no equivalent far right party has sprung up in the UK. The Northern League is not fascist but is hard right and is currently the largest party in the Italian Parliament. The Brothers of Italy are literally neo-fascists and have been part of Italian coalition government and plausibly could be again.
Big tent Social democrats competing with Christian Democrats seemed to work fine.
My claim isn’t that first past the post is clearly better than other electoral systems, but it’s not clearly much worse and so I wouldn’t expect it to pass the very high bar of being an EA cause area.
An important point note is that these arguments apply much less strongly to the US because of the much weaker parties.
I may or may not write a more detailed account of this, but given my record of trying and failing to write good forum posts, this may the best I’ll do.
Shouldn’t the title be “Proportional Representation seems overrated”?
PR is often what people mean by voting reform, in the UK, but there are options without these problems, e.g. approval voting.
My view is the complete opposite: voting reform is the biggest bang-for-the-buck human welfare increaser, by a huge margin. But don’t equate electoral reform with proportional representation.
Two different math PhD’s have run computer simulations to estimate the benefit of alternative voting methods:
A Princeton math PhD named Warren Smith, who advocates score voting (formerly known as range voting) produced these Bayesian regret calculations. https://www.rangevoting.org/BayRegsFig
A Harvard stats PhD named Jameson Quinn used some slightly different modeling assumptions and gave the results in an inverted normalized form called voter satisfaction efficiency or VSE, where 100% VSE is the same as a Bayesian regret of zero.
https://electionscience.github.io/vse-sim/vse.html
Warren Smith found that an increase from plurality voting to score voting increased human welfare about as much as democracy vs. non-democratic random selection. He summarizes thusly:
There is a great deal of empirical evidence that bolsters such claims. In a 2014 exit poll for the Maine gubernatorial race, switching from plurality voting to approval voting led to a complete reversal in finish order. A bombastic climate change denier won the real election due to vote splitting between the Democrat and a Democrat-ish independent.
Warren Smith is actually fairly skeptical of proportional representation, mind you.
My view is that if the impact of alternative voting methods was properly understood, it would comprise the vast majority of the EA community’s efforts. You just can’t get this kind of impact (per dollar) from cash transfers, malaria nets, deworming, etc.
I think using Bayesian regret misses a number of important things.
It’s somewhat unclear if it means utility in the sense of a function that maps preference relations to real numbers, or utility in axiological sense. If it’s in the former sense then I think it misses a number of very important things. The first is that preferences are changed by the political process. The second is that people have stable preferences for terrible things like capital punishment.
If it means it in the axiological sense then I don’t think we have strong reason to believe that how people vote will be closely related and I think we have reason to believe it will be different systematically. This also makes it vulnerable to some people having terrible outcomes.
Lots of what I’m worried about with elected leaders are negative externalities. For instance, quite plausibly the main reasons Trump was bad was his opposition to climate change and rejecting democratic norms. The former harms mostly people in other countries and future generations, and the latter mostly future generations (and probably people in other countries too more than Americans although it’s not obviously true.)
It also doesn’t account for dynamic affects of parties changing their platforms. My claim is that the overton window is real and important.
I think that having strong political parties which the electoral system protects is good for stopping these things in rich democracies because I think the gatekeepers will systematically support the system that put them in power. I also think the set of polices the elite support is better in the axiological sense than those supported by the voting population. The catch here is that the US has weak political parties that are supported by electoral system.
> It’s somewhat unclear if it means utility in the sense of a function that maps preference relations to real numbers, or utility in axiological sense.
there’s only one notion of utility. if your utilities for x,y, and z are 0,3,5 respectively, then you’d find a 60% chance of the 5 option equally preferable to the guarantee of y, and so on.
> preferences are changed by the political process.
well, no. your estimate of how much a given policy will benefit you, that is changed by the political process. the actual utilities aren’t.
> The second is that people have stable preferences for terrible things like capital punishment.
no. people have utilities that relate to things like “being murdered walking down a dark alley”. the preferences they form for policies like capital punishment are estimates of how well off they’ll be under a given legal policy regime. in reality, most people would prefer a world where capital punishment is illegal. but they erroneously think capital punishment is good becaus ethey don’t understand how ineffective it is, and how they themselves could end up being unjustly killed via capital punishment.
you need to update your mental model with the disparity between actual utility from policy, versus the assume utilities that form your espoused political preferences.
that disparity between actual and assumed preferences was already accounted for by “ignorance factors” in the bayesian regret calculations, fyi.
There isn’t only one notion of utility—utility in decision theory is different to utility in ethics. Utility in decision theory can indeed be derived from choices over lotteries and is incomparable between individuals (without further assumptions) and is equivalent under positive affine transformation because it’s just representing choices.
Utility in moral philosophy refers to value and typically refers to the value of experiences (as opposed to other conceptions of the good like satisfaction of preferences), is comparable between individuals without further assumptions and isn’t equivalent under positive affine transformation.
An individual’s utility (on either of the definitions) may or may not be changed by the political process.
Consider a new far-right party entering the political sphere. They successfully changed political conversations to be more anti-immigration and have lots of focus on immigrant men committing sexual violence.
A voter exposed to these new political conversations has their choice behaviour changed because they now feel more angry towards immigrants and want to hurt them, rather than because they think that more restrictive immigration policies would make them personally safer, for instance.
This same voter also has utility—in the moral philosophy sense—changed by the new political conversation. Now they feel sadistic pleasure when they hear about immigrants being deported on the news, leading to better subjective experiences when they see immigrants being deported.
I strongly reject the claim that we should imagine voters as exclusively deciding how to vote in terms of the personal benefits they derive in expectation from policies. I think people support capital punishment mostly because it fits with their inbuilt sense of justice rather than because they think it benefits them.
We could (probably) represent this voter as being an expected utility maximiser where they have positive utility from capital punishment, in the decision theory sense. This is a different claim from the claim that a voter expects their subjective experiences to be more positively valenced when there’s capital punishment.
I’m afraid I can’t comment on what ignorance factors do or do not account for under Bayesian regret without rereading the paper, but it’s of course possible that they do account for that disparity between actual and assumed preferences.
Is this worse than under FPTP, though, when there are often effectively two choices? There will be a leader of the coalition, so you could vote for anyone else, but when there are effectively two parties, and you hate one of them, your options are increasing the risk of letting the party you hate win, or failing to hold your preferred party to account.
What does PR stand for? Perhaps Proportional Representation? Would love if the article was edited to clarify this.
I think this piece about the center-squeeze effect might address your concern about other voting systems leading to greater prominence of extreme candidates. In short, both plurality and ranked-choice voting tend to eliminate centrist candidates early, as many voters may like a centrist candidate but prefer to vote for a more extreme candidate, either because (in RCV) they like the extreme candidate more or (in FPTP) the centrist candidate is not viable. Approval voting gives an advantage to candidates that aren’t everyone’s favorite but are acceptable to voters in all parts of the spectrum.
(TBH I’m pretty undecided between voting systems, but I’ve long held that plurality voting is a bad system. Nowadays I’m sympathetic to approval and quadratic voting.)
Coming back to this thread now having thought about it more. Speaking from my personal experience as an American citizen, I think the spoiler effect lowers voters’ confidence in the electoral system, especially that of idealistic, young voters.
I was a Bernie supporter in 2016. I wasn’t excited about Hillary being the Democratic nominee, but because I understood the incentive structure created by FPTP, I chose to support Hillary in the general election because I really hated Trump. But, there was a substantial number of Bernie supporters who defected from the Democratic base after Bernie lost. Mainstream Democrats seemed to be shaming them into voting for Hillary, on the grounds that:
if you vote for Jill Stein (the Green candidate) instead of Hillary, Trump will win.
if you vote for Gary Johnson (the Libertarian candidate) instead of Hillary, Trump will win.
if you write in Bernie, Trump will win.
if you don’t vote, Trump will win.
This has been called vote-shaming, and I think it makes American political culture a lot more toxic because it pits ideologically similar people (like center-left and far-left progressives) against each other. Many people don’t vote at all, both because of voter suppression, and because they don’t feel represented by the major candidates. Eligible non-voters in 2016 were also more likely to be younger, less educated, less affluent, and non-White (source), which suggests that the system is not representing these groups as well as it could be. It is a problem that citizens of the world’s oldest continuously running democracy feel disempowered—it means that the government is not as responsive to citizens’ interests as it should be. Vote-shaming puts the blame on individuals for not voting, instead of the system for causing vote-splitting.
Just so you all don’t think that this only happens on the left: I have a friend who didn’t really like either major candidate. He leans conservative and strikes me as someone who might have preferred the Libertarian Party or Bernie Sanders. Despite not liking Trump that much, he voted for Trump in the 2016 general, because he thought Hillary was worse.
Some statistics:
In 2016, just 54.8% of the voting-age population (VAP) voted in the presidential election; 59.2% of the voting-eligible population (VEP) voted.
In 2020, this increased to 62% of the VAP and 66.7% of the VEP. (source)
I think that increasing voter turnout would make the government more responsive to citizens’ interests, and I think changing the voting system we use would help with this because it would help citizens feel more empowered to vote.
Note: I’m not saying that vote-splitting, or even problems with the voting mechanism in general, is the only issue with the U.S. electoral system. I think there could be other problems introduced by a new voting system such as approval voting—practical problems that degrade the political system similarly to the way that I think vote-splitting does (since we know that no voting system is theoretically perfect).
Yeah I mean this is a pretty testable hypothesis and I’m tempted to actually test it. My guess is that the level of vote splitting that electoral system has won’t have an effect and that that whether not voting is compulsory, number of young people, level of education and level of trust will explain most of the variation in rich democracies.
I don’t have a strong stance on any of your specific claims or your overall conclusion, but here’s one small piece of evidence that weakly pushes against the idea that coalitions are bad:
This quote is from the post Deliberation May Improve Decision-Making. (I haven’t read the cited studies, don’t know what the effect size is meant to be, and don’t know how important “deliberative quality” actually is.)