I know, I'll stop soon. I have made an editorial decision to cover local affairs here, ranging from somewhere inside the house so no further than the B.C. border, but I will make exceptions.

This one is for the U.S. election, so variously misinterpreted in so many ways. The strange misapprehension I'd like to address today is the idea that there's an emerging long-term majority for either major party. Barring a tremendous leadership failure in either camp, it's not going to happen.

The nature of the two-party system is that each party needs to capture just over half the voters in any election. More is great, but the further you over-reach, the more likely you are to alienate your most fringe supporters, who will then defect to Nader or Buchanan, or simply stay home on election night. So there are natural tensions that encourage a platform that appeals to just over 50% of the voters. In local elections, the candidates tend to adjust their platforms to local political conditions, thus causing southern Democrats and blue-state Republicans like Mike Bloomberg or Arnold Schwarzenegger (cool site, eh?).

This presidential election saw the usual tight split, about 51-48, give or take a point. That basically means that both candidates came very close to winning positions, but Bush came a little closer (this passes over the various tactical issues like why the turnout was so high, and the effects of get out the vote (GOTV) efforts by both parties, but that's implicitly part of the game of getting 50% of the voters).

What are Democrats to do? Well, they could rend their garments, declare the middle of the country idiotic, decide they need to do all kinds of neat things. They probably needn't bother. It would be like the Oakland A's firing their GM because the team missed the playoffs by 1 game, after several years of consistent winning seasons on cut-rate budgets.

Maybe if Kerry hadn't come off as a demented stiff, they would have won this time. Maybe Americans really did want to back the incumbent in wartime, and not even Zombie Reagan could have beaten him.

But when either party loses, it usually reflects either a poor candidate, a poor strategy, or a failure to find policies that will attract those marginal median voters that qualify as election-deciding moderate independents.

This middle point moves around as voter attitudes change: Nixon declared "we're all Keynsians now" as he enacted policies favouring the spend-in-lean-times, save-in-fat-times doctrines of that economist. Nowadays, no politician would publically favour Keynesian policy, as the median point has moved to a place that favours avoiding deficits (this wasn't a factor in this election, as Bush has been the one running Keynesian deficits, and Kerry was touting himself as a fiscal conservative; I think budgetary deficits were a minor issue for most voters). Similarly, Bill Clinton found and rode a winning issue with welfare reform, despite the general opposition in his own party to changing the welfare system. He followed the median votes, and was rewarded. Public health care? the median voter wasn't ready for that one.

What this means in the long term is that unless the Democrats go stupid, they will respond to this election with what are likely to be minor tactical tweaks, and notable policy tweaks aimed at bringing the party closer to the interests of those median voters. Similarly, the Republicans will be chasing down those median voters just as hard, but neither side should have a long-term structural advantage unless internal party dynamics drive them away from the interests of median voters.

It could happen: the dedicated fringe of either party is unlikely to be satisfied with the kinds of compromises that attract median voters, and if sufficiently displeased, are liable to stay home or vote for another candidate.

But the median voter is the centre of gravity here, the point about which the Republicans and Democrats must necessarily orbit. Each party can try to either move the minds of the medians closer to their own platform (either by changing the minds of the middle voter, or by convincing more of their natural voters to vote, thus moving the median in a favorable direction), or move their own platform closer to the centre, but the other party can always respond in the same way.

There are gross simplifications in this model: political attitudes are multidimensional, and it is quite possible for a voter to hold naturally Democratic positions in one dimension, and naturally Republican positions in another dimension. Libertarians are one classic example, tending towards socially liberal attitudes and fiscally conservative ones. As a group, they tend to split interestingly between voting for Republicans, voting for Democrats, voting for libertarians, or making a principled decision to stay home because the return-rate on the task of voting is too low (since, after all, a single vote rarely makes a difference; alas, a sort of tragedy of the commons occurs when all libertarians think this way, in that libertarian votes are seen as unattainable, and they are thus ignored by party platforms). But one way or the other, each party is always striving for that median voter more than it is striving for any one policy, and unless one party or the other lapses into extraordinary incompetence (say, nominating Walter Mondale for president again), the margins should never sway much beyond 50/50.

Whether any particular Democrat or Republican will be happy with the machinations and policies necessary to get that median voter is quite another matter.