Red/Blue/Rich/Poor

[shkrobius]

I've finished "Red State, Blue State, Rich State, Poor State" by Gelman, Shor & Co. The subject is voting patterns of Americans. The thesis is that Blue states vs. Red states has nothing to do with haves vs. have-nots; the confrontation is limited to the affluent. The interests of the poor do not seem to be the real concern, despite the lip service

Comments

[stas]

rich states are growing more socially liberal (also not true).

It would be nice to see how about economically liberal (i.e. socialist). I'd imagine if state becomes more rich the pressure to spend part of it on socialist policies would grow considerably.
As for social policies, I don't see a reason why that should have any correlation with income...

[shkrobius]

I returned the book, so I can't really speak for its authors, but perhaps they would retort that "rich" is a relative term whereas you treat it as absolute. Their whole point is that median income is not a very good metric and you do not get much out of comparing this metric for different states.

What they do is plotting the voting preference vs. income percentile. They get a straight line (amazingly straight line) for each state but the line is different. I think you would enjoy reading this book; I know too little of its subject to make it justice. Boris Shor is now a professor at UoC, I was told he is tops. It is low on economic analysis, it is maily statistical inferences.

[stas]

Thanks, I'll check it out - looks like my local library has it. It also has recommendation from Nassim Nicholas Taleb on their site, which for me has big weight.
In general, I too think income level would not be a good metric, but I probably couldn't base it on any cold hard facts, so it would be interesting to see what they have.

[shkrobius]

It would be interesting to hear what do you think of the arguments. Let's return to it once you are finished.

[poltorazhyda]

I'd guess that rich people DO in fact vote for the democrats-it just depends on where you set the bar for "rich."

[shkrobius]

Ok, here what they do: within each state you can grade people from the poorest to the rischest and than put them into different batches (percentiles). For each batch you examine their voting preference (say % of those voting R) and plot it against the median income for each batch. The assertion is that for each state in the Union the plot is linear. There are no Red and Blue states in this picture. In each state the rich tend to vote R, the poor for D. You would not find a single state where the affluent vote for D more than the poor, or where more there are more affluent D voters than affluent R voters. This is not what makes states Red or Blue. It does not matter where you set the bar for "rich" because it is the universal trend. The more affluent you are, the more likely you vote R, everywhere. The difference is only in how more likely. The slope does vary from state to state. Furthermore, it seems to be more or less constant over time for each state. Given that, you can try to predict elections from income distributions. This is

[irrelative]

Wow... Those myths, I wonder where do they come from. Right now it looks they (at least some of them) are invented by the author himself just to prove them false.

"Rich people vote for the Democrats" - not only counterintuitive but also contradicts to everything I have ever heard on this matter :-)

[shkrobius]

I've seen most of these stated many times, except for Kansas, which is indeed new to me. I certainly did not know that church-going is such a weak indicator for poor. Rich people do vote D, but more of them vote R. If you live in an enclave of rich D voters you may hear the stories that those rich that vote R are not rich enough, not educated enough, have become rich only recently, or "not us" in some other way. I'd read this book first and criticize later, would I be you. Some people took it seriously, for example, David Axelrod. It is not written in partisan fashion, and all these "myths" are richly documented before being discussed.

[irrelative]

Never heard of these stories, honestly.
The most prevailing myth is that R electorate are either "rich bastards" or "redneks". But again, definition of "rich" may be different.
I am not really critisizing the book, just stating my amazement with "myths" you listed here are never heard of in my enviroment. Perhaps it does depend on environment more then we think it is.