Brooks (brooksmoses) wrote,

On the two-party system and voting by party

I posted this as a comment in a friend's locked entry, but decided to repost it here so I could point people at it. My friend was complaining about the two-party system in general, and in particular about people who felt it mattered a great deal to vote Democrat[1] or Republican despite the fact that their positions weren't that different -- and despite the fact that neither candidate really represents the voter's own positions very well. Since I do feel it matters that I vote Democrat regardless of whether the Democrats are leaning liberal or conservative this year, I wrote the following.

Part of the reason for my "D vs R mentality" is that, while I don't necessarily agree with everything the Democratic Party tends to endorse (though I do agree with them far more than you do, I think), I do think that the direction of change in government matters.

Why do I think it matters? Because parties move to follow the center of the electorate; in a economic analysis of optimal behavior for a two-party system, you have both parties playing "king of the hill" for the the middle of the electorate bell curve. If one of the parties starts winning a lot, that means the middle of that curve is over on their side of the voter break, and so the opposing party is going to start moving that way to try to regain voters. This is, I think, the clear and simple reason why Kerry and Edwards were practically out-conservativing Bush and Cheney this year.

And so it's important, I believe, that "my party" wins, because that moves the center back towards the liberal side of things, and both parties will start having more liberal positions as a direct result.

This also, I think, is evidence that the two-party system isn't nearly as bad as it looks, so long as voting positions can be reasonably approximated as a one-dimensional continuum. Sure, the individual candidates may not be all that different, and so on any given election it may not look like one's vote has much effect (and there won't be a candidate that one can really agree with) -- but, in the slightly longer-term view, the positions of _both_ parties get shifted to the median view of the voters. Given the 51-48 split, I'd say they came pretty close to hitting the center in this election, too.

(The failure of the one-dimensional continuum model to represent voter positions is a topic for another discussion. I don't think the multidimensionality makes any practical difference to the analysis for any winner-takes-all electoral strategy, though.)

Another interesting thing that occurs to me: the two-party system means that the winner is going to be at nearly the median position of all voters. A lots-of-parties system would mean that the winner would be who had the largest "clump" of voters to him- or herself, which in a multidimensional non-normal distribution is unlikely to be anywhere near the median. Witness, say, Gore and Nader splitting the "liberal" vote and (allegedly) giving Bush his win four years ago -- if we had serious third parties and retained the plurality-takes-all voting process, that sort of split would be happening all the time, many times over.

This also raises the point that, if you don't like the two-party system and want it changed, the place to start is to work for a better voting system. Multiple party systems need runoff elections, at the very least.

[1] Yes, lilairen, I am intentionally using the noun form of the party name there, not the adjective form. I also "voted Kerry", not "voted Kerryish". (And I do know that in both cases, it's idiomatic grammar that's not strictly correct.)

  • Post a new comment


    default userpic

    Your reply will be screened

    Your IP address will be recorded 

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.