Andrew Gelman has a post up about an exchange we had over the weekend. Here’s a historical image of the two-party share of the House vote:
My view is that the right way to understand this is that historically Democrats averaged about 55 percent of the vote because of the weirdness around the one party politics of the Jim Crow South. Then from 1994 on you have a 50/50 average and you should understand the 2008 Democratic landslide as an unsustainable outlier. Gelman points out that this is pretty speculative on my part and that as of January 2009 it was at least possible to argue that the 2008 result was a return to historical norms. Obviously I can’t prove I’ll be right about the future, but I’d like to go on record with my prediction—I think over the next decade or two the parties will be about equal on average. With the current distribution of voters and districts, a 50:50 split of the national vote leads to a GOP majority, so I expect we’ll see small Republican majorities as the most common outcome.