Slashdot Mirror
Western-Style Voting 'A Loser'
sethawoolley writes "In light of the upcoming elections in the US, author William Poundstone was interviewed about voting systems by Mother Jones. In it he advocates the benefits of Range Voting as a solution to Arrow's Impossibility Theorem. Approval, Borda, Instant Runoff, and Condorcet Voting, which are often solutions advocated by the Greens and Libertarians (in the US), are discussed, as well, in light of Warren Smith's recent empirical research using Bayesian Regret. My local party (of which I'm the Parliamentarian) uses Single Transferable voting, but we're considering using Range Voting in the future. One thing is for certain: any system is better than the West's out-dated plurality voting system."
Excuse me, but a great number of what I'd call 'Western' countries use other systems than pluralist votes. For example, the German Federal Diet is elected by a hybrid of the first-past-the-post election system and party-list proportional representation. Proportional systems are also used in countries like Finland, Austria, Spain and many others. Remember: Just because the USA and the UK use it, it doesn't make it "Western" by default. (Just because -their- minds boggle when we here get along well with a four-party coalition government....)
That's completely wrong. Range Voting consists of adding up the numbers and then taking the average. As anyone knows, that's linear in the number of candidates and votes. Even if you do it by counting "pseudovotes" (this candidate got that many ones, twos, threes, etc up to nines), the granularity of the ballot is a constant, so it's still linear.
As for Condorcet, counting a ballot takes quadratic (0.5*n^2) time with respect to the number of candidates. If A, B, and C are ranked on a ballot, then you just check if A is more highly ranked than B, A more highly ranked than C, and B more highly ranked than C.
Finding out who the winner is is linear in the best case - that there's a candidate who's preferred to all the others one-on-one and that's the first candidate you checked, and quadratic in the worst case if there's still a candidate who's preferred to all the others. If there is a cycle, the methods vary, but in public elections, that would be exceedingly rare. Though for the case of completion, I'll note that most of the good Condorcet methods (like CSSD which Debian uses) are n^3 in the very worst case. In either case, determining the winner once the votes have been totaled up into the matrix takes logarithmic time in terms of the number of ballots (since all you have to do is compare numbers in the matrix or the averages list).
Another advantage with Range or Condorcet is that you can count the ballots where they're gathered and then only transmit a small amount of data (the pairwise counts for Condorcet, or the numerators and denominators for the average for Range), instead of having to count everything at the central place as in IRV.
That it is compulsory in Australia helps to moderate our politics by ensuring that the almost the whole population turns out to vote, not just ultra-motivated special interest groups (churchies, to pick a purely random example).
Too bad about the how-to-vote cards then, no? Though there's nothing about IRV that demands you have to rank absolutely all the candidates, the implementation you have is flawed.
I never understood why the US keeps mucking about with these increasingly bizarre voting systems. Pretty much every other democracy - Western democracy - I know off either has a 1) parliamentary system, or 2) uses multiple votes.
Parliamentary systems: Here, the populace elects parliaments, usually with proportional representations. The parliaments then elect the 'single seat', such as the head of government.
Multiple votes: Here, the populace elects the 'single seat' directly. If in the first n [n>0] votes no candidate achieves an absolute majority, then a final plurality vote is conducted.
As said, pretty much every "Western" democracy other than the US seems to use some variant of those two. I personally like the first better as it keeps the center of power in the parliament, which is sort of a good thing for a democracy. But either solves the problem in a clean, easily understood and verifiable manner. So... what's the deal with the US and their funky voting systems craze?
Also, I'm rather thankful for the various people pointing out the blatant mis-use of the term "West".
When the elections are over, the winning party needs to form a government, and they do this by making a coalition with one or two other parties so they represent more than 50% of the voting people in the country. This way all major opinions should be represented in a government. A new party might not be a part of a new government, but they are able to use there representation power in the parlement, for example when new laws are discussed and voted for.
Well, I'm from Belgium as well, and our electoral system is one of the worst ones around! For these reasons:
- As a resident of Flanders, I can only vote for Flemish parties, this means that, at best, I'm only allowed to vote for a bit more then half the decisions made in this country... This means that I, being Flemish, can only vote for Flemish interests, how absurd is that?
- Up until a month ago, the 3 major tendencies (Liberals, Conservatives/Catholics/Humanists and socialists) were all in the government (either regional or national), and guess what, we will now have the same 3 tendencies (except for the Flemish socialists) in our future government! How is it exactly that the people chose if everyone is still in the government anyways?
- Whatever party you chose, you _know_ that they won't be able to fulfill what they promised us, since they will have to make a coalition and find middle solution for everything anyways...
The Belgian system in all its glory has become a particracy, where the heads of the different political parties have much more to say about who rules what then the people. Our system is probably one of the most anti-democratic systems there is around, and this had grave consequences... In Flanders up until the last elections, the biggest party was an extreme rights party (well duh, they're the only opposition), in Walloon, the French socialists have had their hands on on local and regional matters for the best part of the last and the beginning of this century, leading to corruption scandal after corruption scandal, and since they have only been thought to think for themselves, their education system is so lame language wise that most of them never even get a chance at working in the Flemish part. Our country became just two cultures stuck together round a common economic interest (Brussels), without any prospect of ever forming a true nation.
Bravo, please copy our electoral system, it's great!
Not so. Single Transferrable Voting fails the monotonocity criterion. Basically, ranking someone higher can cause them to lose, and ranking someone lower can cause them to win. There's debate on how often this might come up in practice. It might be missing the larger point, though, which is that in STV, it's very hard to predict what impact your vote will actually have.
STV is the only mainstream electoral method which fails the monotonocity criterion. Even the much maligned plurality method, which everyone is familiar with, passes. Voting for someone will never cause them to lose, and not voting for someone will never cause them to win.
Arrow's Theorem says we can't have everything, but I consider the monotonocity criterion as something which is an absolute must. At the very least, if you are contemplating switching away from the plurality system to something else, be sure that it is strictly better than plurality, which STV is not.