Maskin’s argument is that voters should list candidates in order of preference, so we wouldn’t have to guess whether Clinton would have beaten Obama in a two-person contest. If a majority of voters for Edwards, Richardson and the other also-rans put Clinton higher on their lists than Obama, she would win the contest under Maskin’s system. But if Obama ranked higher than Clinton on a majority of voters’ lists, then he would win. After all, most people would have preferred him.
Now, Maskin’s being pretty vague here. There are a variety of preference-based voting systems. The most common is instant runoff voting (called single transferable vote in multi-seat elections), where the candidate with the lowest number of first choice ballots has them redistributed to the candidates ranked second, and so forth until one candidate has a majority. This system isn’t optimal; it isn’t a Condorcet method, meaning it doesn’t always select the candidate who would beat each of the other candidates in a runoff election. But it, along with any Condorcet method, would eliminate the spoiler effect.
Matt, however, thinks the solution’s too small:
While this system seems to have problems that I can’t quite identify (political scientists and economists, this is where you tell me what they are) it seems vastly preferable to a system that can’t seem to accommodate having more than two candidates. But Mallaby wasn’t just writing a column about how our voting system doesn’t encourage people to express their preferences, but why the system as a whole is screwy. It the next paragraph, he writes “On the basis of a three-point margin over Obama that tells us little about which of the two candidates voters actually preferred, Clinton has transformed her prospects.” While he’s certainly correct that a three point win in the current system doesn’t mean much, the real reason it’s all messed up is because small, unrepresentative states like Iowa and New Hampshire have all too much influence on selecting nominees. Iowa’s caucuses, which are derided for not being accessible or private, actually allow people to express preferences — for example, someone could support Biden in the first round, and after he gets less than 15% of the vote, move on to Clinton in the second round — but the fact that a candidate can launch him or herself to the nomination from Iowa, or at least get a 15% bounce in the national polls, is infinitely more ridiculous than the actual method used to select said candidate.
And, oh yeah, we have the electoral college. As long as we allow Iowa-New Hampshire to have hegemony, or at least outsize influence, over the primaries and then use the electoral college for the general election, any better method for actually voting is just rearranging deck chairs on the Titanic.
Well, this is true, clearly. While I certainly prefer small-state primaries where underfunded candidates have a chance to a national one that would be almost entirely decided by fundraising prowess, I admit that Iowa and, yes, New Hampshire’s outsized roles in the process are nothing short of ridiculous and unrepresentative. Even worse is the way the media chooses to cover them; there’s not even a workable solution for that problem, for as long as the public keeps paying for breathless horse-race coverage, there will be breathless horse-race coverage. And Lord knows the electoral college has to go.
But a preferential voting system wouldn’t have as minimal an impact as Matt thinks. As previously noted, it does away with the spoiler effect; Nader’s voters would have been redistributed in 2000, with most undoubtedly going to Gore, with the end result, even if this process was done on the state level, a handy Gore victory. In New Hampshire last week, Edwards supporters, who would probably place Obama higher in preference than Clinton, would have had their votes distributed overwhelmingly to him, meaning that even as the left wing of the party was slightly split between the two candidates, the final vote would show a united tally (and an Obama victory). While I’m sure these examples appeal to me and Matt because we’re both Gore/Obama supporters, it’s better on principle for the dominant ideological impulse in the electorate to be represented in the final election result than for that segment of the population to be split and a minority faction to win.
More importantly, however, negative campaigning would substantially decline under a preferential voting system. Candidates would be competing not just for first-place votes, but for the second and third-place votes of their opponents’ supporters. The Clinton campaign would be less likely to level racially charged attacks against Obama if it knew that this would lead to them getting placed third behind Edwards by Obama supporters in a later primary, and thus suffering in the final count. This effect is less helpful in a general election; it wouldn’t have deterred the SwiftVets in 2004, given as Kerry voters were bound to place Bush third, behind Nader, anyway. But it would help prevent candidates from killing each other in a primary, and thus result in a more united party for the general election.
So Matt’s right that preferential voting isn’t the be-all and end-all of electoral reform. But I think it would do more good than he gives it credit for.