Paradox of the Median Voter

The pundits are lousy with ideas for how the Democrats can reform their party to come back in 2004. But while this is the story right now, it's really not that big of a defeat. Democrats only lost two seats in the Senate and five in the House. Compare that to 1994, when Democrats lost 53 House seats (including the Speaker's) and 10 Senate seats.

Both elections had low turnout, which tends to favor Republicans. But in 1994, the B variable (which can be thought of as the difference between the two parties' platforms) was spectacularly high, thanks to the Contract with America. (Despite the high B, there was low turnout. Why? Probably because only traditionally-Republican voters were fired up by the high B, while voters who usually vote Democratic took a nap. If Democrats had had a similar Contract-with-America-style platform, perhaps they would have been able to energize their supporters.)

The important thing to consider is that the Contract with America was decidedly not centered around the median voter. And this leads to the median voter theorem's downfall: it is only useful in predicting elections when nearly everyone votes. When the election is still mostly about getting out the vote, the game is getting your supporters fired up, not centering in the middle. Democrats would be wise to keep that in mind, and reject the contrary view.

As I explained two entries ago, political parties target the median voter - that voter who has an equal number of voters on either side of her in the political spectrum. And yet because voter turnout is so important, the parties must also focus on turnout, not just positioning.

Getting out the vote is about more than just phone banks. This election featured recorded calls from Bill Clinton and Martin Sheen urging voters to mark their ballots for Democrats. But when voters are faced with both unattractive candidates (Gray Davis, etc.) and only a small perceptible difference between the two parties' platforms (a small B value), even a phone call from President Bartlet is not enough to get Democrats to the polls.

Paradoxically, the rational voter theory places a premium on the B value for most voters, the benefit that the voter will receive if his candidate wins. If both parties are centered around the median voter, the B variable is necessarily zero. So in order to win elections, parties need to do more than just target the median voter. Because the median voter doesn't vote.

Turns out, the Democrats may have been paying a little too much attention to this rational voter mumbo jumbo. Instead of trying to get the Democratic party to move toward the center, why not get the center to move toward the Democratic Party? What happened to the power of persuasian? What happened to leadership? More and more I think it went out the door when Bill Clinton left office, and Democrats simply forgot what it the Clinton Administration was about. Was bailing out Mexico a center-position? Was NAFTA? Was a tax increase? None of these positions were based around the political center. Instead they were based on their merit. If only the current crop of Democrats could learn the lessons of the Clinton years, they would most certainly have won back the House and kept the Senate.

comments :: post a comment
comment on this article


(not published)
(optional)

No HTML allowed; press Shift-Enter to insert line breaks
enter code