What's the Matter with Kansas? Not all that much, it turns out.

So, a number of years ago, political scientist Thomas Frank did something that very few political scientists ever do: he wrote a book that normal people actually read.

His What's the Matter with Kansas? made the argument that low-income people in red states like Kansas had drifted away from the traditional New Deal partisan alignment where they had supported Democrats for economic reasons and had instead begun supporting Republicans for cultural reasons.

And it is true that red states tend to have lower median incomes, and blue states tend to have higher median incomes. And it is true that, in contrast to earlier parts of this century, red states almost never turn blue, and blue states almost never turn red. So, is Frank right? Is this pattern driven by low-income, culturally conservative voters defecting away from their own economic interests because of cultural issues?

Since Frank's book came out, a variety of scholars, most notably Andrew Gellman, have put out empirical work that shows that Frank's argument is fairly off-base. While it is true that low-income people tend to be culturally conservative, it isn't true that they have moved en masse to support Republicans. In fact, the partisan distribution of folks in lower income brackets hasn't really changed that much. A majority are still Democrats, and they still care about economic issues, insofar as they really have any clear policy preferences at all (the coherency of ideological and policy preferences tends to increase with education and income).

The really interesting thing that comes out of Gellman's work is that the state-level party polarization that characterizes modern American politics (i.e., the fact that there is almost no variation from election to election on which party wins in a given state) is driven by the rich, not the poor. The real difference between the rich, blue states and the poor, red states is that, in the rich IN the red states are both culturally and economically conservative, and the rich in the rich states are socially and economically liberal.

So, in blue states, the rich defect from their own economic interests (lower taxes, etc) to support Democrats, largely for cultural reasons. In red states, the rich are more culturally conservative (despite the fact that wealth tends to be correlated with social liberalism), and tend to support Republicans.

So, what's the matter with Kansas? Nothing, really. It just has a lot of upper-middle class, very conservative people, and there aren't enough liberal poor people to win elections for liberals.

Interestingly, one upshot of these findings is that there is more policy consensus across income groups in blue states. In red states, it resembles more of the class-conflict model that has so dominated thinking about politics for the last 200 years.