I highly recommend the book How Congress Evolves by the late Nelson Polsby to any student or practitioner of politics. Polsby documents the interplay between Congressional rules, demographic shift, and partisan realignment. The essential fact was that Southern congressional districts fell to the Republicans as a result of southern migration from the traditionally Republican Northeast and Midwest. (Polsby points to the rise of residential air-conditioning and military-industrial complex) The first seat to go reliably Republican was in St. Petersburg, FL. Gradually, the loss of conservative Southern Democrats shifted the Congressional Democratic Party to the left. Gradually, a more liberal Democratic Caucus changed House and Caucus rules to force conservatives to concede, retire (usually handing seats to Republicans), or switch parties.

Winners Losers
Texas +4 Ohio -2
Florida +2 New York -2
Arizona +2 California -1
North Carolina +1 Massachusetts -1
South Carolina +1 New Jersey -1
Georgia +1 Pennsylvania -1
Utah +1 Michigan -1
Nevada +1 Illinois -1
Oregon +1 Minnesota -1

It is with that context that I look at a post by Brian Faughn over at the Weekly Standard. His thesis is simple:

The New York Times reports that while the subprime mortgage crisis has slowed the population shift away from states such as California, New York, New Jersey, and Massachusetts, the trend for the decade is clear: the red states are gaining people and electoral votes while the blue states are losing them

With the conclusion:

This would represent a shift of eight seats from Kerry states to Bush states. A Democratic candidate who held all of Kerry’s states would also need to win Florida, or a similar combination of smaller states, to gain the presidency.

I quibble with his language here. He says "a shift of eight seats" when it is really a shift of eight electoral votes. And that’s the rub. I have two points that I significantly differ with him. The first is that I wonder who actually fills those seats. The second is captured by his caveat, "states change character and become more or less competitive for parties over time."

First, who fills the seats. Let’s take the states one at a time. Based on Polidata projections, Brian lists the states that win and lose. While I only have specific knowledge about some of these states, we can do some rough estimates. It seems that a large source of migration into Texas is Hispanic which is still significantly Democratic. It seems that the Texas seats would go GOP/Dem 1/3 or 2/2. The growth in Florida is in South Florida, which went substantially blue in the last election. Perhaps a split, although gerrymandering could result in a 0/2 GOP pickup. Arizona is unclear to me. Most of the Southern states are likely to be GOP pickups.

Out West, it is probably a different story. My gut is that Nevada’s seat goes blue or we lose Rep. Jon Porter (R)’s suburban Clark County seat, resulting in a wash. It is hard to imagine that Oregon’s growth is somewhere other than the Portland area, which we lose. Utah is, of course, the exception.

Looking to the states that lose seats, it is actually kind of grim for Republicans. Clearly we lose the California seat because of redistricting. There is no seat to lose in Mass. We will lose the Illinois seat because the partisan gerrymander will combine two GOP suburban districts. Michigan is also a highly gerrymandered state that over-performs GOP at the congressional level, not to mention the strong possibility that we lose MI-07 this cycle. New York lossage is almost certainly from upstate, and there is a strong chance that we lose the state senate by 2010, which could create the circumstances to lose more than 2 seats. Pennsylvania is another state with a strong partisan gerrymander that will likely be broken by a Dem governor and state House.

The upshot is that who fills the seats is, at least, mixed.

Second, the issue that "states change character." The swing states are different than they were several years ago. The argument for Colorado and Virginia being purple is now transparent, something that might not have been true in 2004. Selling a Southern evangelical in 2000 and 2004 to West Virginia and Arkansas seemed easy, but a zillionaire Massachusetts Mormon? Bush suprisingly pulled off New Mexico, but it seems unlikely to be a repeat performance for the GOP. The upshot is that the sentence, "[a] Democratic candidate who held all of Kerry’s states would also need to win Florida, or a similar combination of smaller states, to gain the presidency," seems remarkably un-farfetched.

Indeed, the challenge for the GOP is going to be fighting back against the intra-state trends. In Virginia, southeastern Pennsylvania, and New Jersey, the GOP needs an agenda that is more relevant to the suburbs. In the West and Florida — and nationwide — we need better Hispanic numbers. And in the rust-belt, we need a response to irresponsible Democratic anti-globalization demagoguing.

Interestingly, the 2008 GOP presidential field features three different kinds of heterodox candidates who try to address these failings. Rudy Giuliani might well offer an answer in the inner-suburbs. John McCain provides a path to greater penetration in the Hispanic vote and a personality that appeals in the upper-Midwest. And Mike Huckabee offers a populism that could help consolidate the weak Southern states and the rust-belt. The fourth option is, of course, the status quo. National Review, in an article about how broken the GOP coalition is, characterized it like this:

Romney and Thompson, meanwhile, are fighting over who is the most conventional, paint-by-numbers conservative circa 1987.

In conclusion, I think that Brian is right in some sort of static analysis. But the world isn’t static. The Reagan and even Bush coalitions are basically gone. It is very, very dangerous for the GOP to look at 2012 with anything but great apprehension. That’s why we need a candidate at the top of the ticket in 2008 who has something different to offer. And that’s why we need a Congressional party that is willing to substantially address some of our flaws. And I am not seeing it.