Red State, Blue State, Rich State, Poor State

So there’s a couple of statistics/political science professors. They wrote a book. And it explains everything.

I’m too drunk to transform the bullet points I made into sentences, so you get bullet points. And you’ll like it.

Shorter version of the book: why Sarah Palin is a culture warrior - she’s filthy rich.

The basic image here, also seen in Tom Frank’s What’s the Matter With Kansas, is of low-income people hoodwinked into backing the GOP by culture war rhetoric. But Andrew Gelman and his coauthors in the excellent Red State, Blue State, Rich State, Poor State mount a huge pile of data to suggest that this isn’t the case. Overall, low income back strongly and consistently back Democratic candidates. Where you see culture war voting is among rich people. They explain this, plausibly, in terms of the fact that privileged people are able to do more to express their cultural preferences — both in terms of lifestyle and in terms of who they vote for. Poor people need to spend their money on stuff they need and cast their votes for practical reasons. But the well-off can afford to indulge their preferences about where to live, how to vacation, and what recreational pursuits to follow and divergent tastes in these matters continues into the voting booth.

Our current crop of candidates offers up some pretty good examples of this. The McCain family is really stinkin’ rich (inheriting multi-million dollar fortunes and owning a dozen houses) but the other three couples on national tickets are well-off on a much more banal scale. The Palin family, the Obama family, and the Biden family all have incomes running into the six figures which is much more than your average American family has. But the Palins choose to spend their money in very different ways. They’re raising five kids, getting into competitive snowmobiling, going on moose hunting expeditions, etc. This isn’t stuff that your typical coastal elites care to do with their time and money, but none of it’s cheap, either. Rather, these are the leisure pursuits of Red America’s economic elite while prosperous people in Blue America are instead raising fewer children in smaller houses that are much more expensive per square foot and spending money on cheese plates rather than moooseburgers.
The Myth
[li]Example 1: Tucker Carlson, 2007 - Ok, but here’s the fact that nobody ever, ever mentions - Democrats win rich people. Over $100,000 in income, you are like more than not to vote for Democrats. People never point that out. Rich people vote liberal. I don’t know what that’s all about.[/li][li]Example 2:[/li][li]Example 3::[/li][li]Interesting, because it sounds like it’s right if you glance at what state voted for who[/li][/ul]

[li]Strong correlation between income and voting in red states (strong in Texas for both counties and voters)[/li][li]Weak correlation between income and voting in blue states (virtually nothing in Maryland)[/li][li]Rich states vote for Democrats, Rich voters vote for Republican[/li][li]Poor states vote for Republicans, poor voters vote for Democrats (Bush 62% 200k+, 36% 15k-)[/li][/ul]

[li]State trend has appeared in last thirty years; didn’t exist in 1976[/li][li]Rich/poor voter divide has increased in last thirty years[/li][li]David Brooks vs. Thomas Frank[/li][LIST]
[li]Both socially liberal[/li][li]Disagree on economics[/li][li]Both live in Washington DC area[/li][li]Very easy to find a county that votes for the GOP and is poorer than Democratic-voting county in Maryland[/li][li]Journalists live in rich states, availability bias[/li][/ul]


Demographic vote shifts
[li]Last 30 years have seen white-collar and professionals go Democratic[/li][li]Last 30 years have seen owners and both skilled and non-skilled workers go Republican[/li][/ul]
Minority misidentification
[li]People who aren’t black think blacks are 24% of the population, not the correct 12% of the population[/li][li]Majority of whites think blacks are doing as well or better than everyone else economically; actually have 60% less average earnings[/li][li]Similar conclusions about immigrants and minorities seen in Europe[/li][li][/li][li][/li][/ul]
Rich voter culture war
[li]Evangelical protestants have moved to the GOP[/li][li]Catholics and Mainline protestants have moved to the center[/li][li]Changes mostly concentrated in voting behavior of the rich in poor states[/li][li]Religion correlates with income in poor states, reverse in rich states[/li][li]Top 10% of income = 15% of votes[/li][/ul]
Partisan gap
[li]Voters issue opinions still mixed (85% of GOP is not conservative or does have official opinion on an issue; roughly same for Democrats)[/li][li]Southern conservative democrats go to the GOP[/li][li]Killer Bartels graph - the rich make out like bandits under the GOP while everyone else is screwed, under Democrats everyone does pretty well instead[/li][LIST]

[li]Geographical industry difference on donations[/li][li]Regressions that Democrats can’t do much better by moving left economically, which I don’t buy[/li][li]Public vs. elite[/li][li]Enormous public margins for taxes on the superrich[/li][/LIST]
[li]Explains an awful lot, doesn’t it?[/li][/ul]

Andrew Gelman Links

Some people post while high and out comes gibberish.

McCullough posts while high and writes…an outline.

I wrote up the outline earlier while sober, I’m just too lazy to pretend to turn it into a book review.

It’s a seriously amazing book for how they tease the correlations from the data.

I don’t know about American voter distribution but I’ve sure seen a lot of movies, so…

They may get their ideas about blacks from movies which routinely overrepresent both the population share and the income level of blacks, in the interest of promoting racial equality. Blacks are the principal media stereotypes for “other”/“victim”/“look how tolerant we are”, due to the history of slavery and racism and probably also due to simply looking the most different from their white target audience. To correctly reflect America’s contemporary ethnic composition, movies should feature a shitload more Asians and Hispanics and a good deal fewer Africans, and only the Asians would typically appear in high-income jobs. did a piece yesterday with three nice maps that summarizes the state vote by income group. If only low income voters were counted, Obama would have damn near wiped McCain.

I really don’t think African-Americans make up 12% of the movie population (until the mid-80s, 2% of major movie stars had been African-American), and if they do, it’s likely not as full, “normal” characters. So I think it’d rather be a reduction of whites and more ethnic diversity.

Though I’m pretty sure the miscalculation of black wealth has more to do with thinking that black people get some magical welfare checks that they can live the high life on, instead of the miserable existence they actually provide.

African-Americans tend to make up a very small part of both the total cast and of the big stars whose names sell movies, but starting around the 1980s you got a virtually mandatory black co-star or two whose characters represent an equal or higher professional position relative to the white star(s) of the movie. Naturally, the audience doesn’t count how many black extras are standing around in the background; but if there’s one black boss (as co-star with a good deal of screentime) of three white professionals then you’ve got your impression of 25% blacks with high income, contrary to the real situation where such situations are much rarer.

Though I’m pretty sure the miscalculation of black wealth has more to do with thinking that black people get some magical welfare checks that they can live the high life on, instead of the miserable existence they actually provide.

Overestimation of welfare likely also plays a part, sure.

I can see movies for population percentage, but it doesn’t explain the income thing. The income misinterpretation goes back a ways, and I’d say the only blacks in movies until about the mid-1990s were poor criminals.

Family incomes by race here. I strongly suspect that the 10% higher median income for asian-headed families is due to more workers and hours per-family, rather than a wage differential. Poking around the web gets a bit of debate; this argues that the common “wages are the same or higher” is due to methodoligcal errors, and finds asian men have 13-20% lower wages due to their race. Which is a slightly different question, but there you go.

•State trend has appeared in last thirty years; didn’t exist in 1976

Because of gathering techniques (I.E. 'we’ve just started to collect this data)?

Counterpoint: Action Jackson. Released in 1988, and that’s only one example. Now, trending may have some other data, but there were several high-profile black actors playing good guys by that time.

I think the miscalculation comes less from movies and more from TV. I call it “The Cosby Effect”.

The rich culture warrior angle is interesting. I had always considered that the environmentalist camp, for example, was able to act that way or impose their will due to having the excess income to spend on the latest green craze while poor countries and people just want to eat, but never considered that the right does the same thing. I would provide an example about a principle that they can stand on given the luxury that they are rich, but they don’t really stand for any principles any more, so that’s kind of useless.

They do mention this, the income data is available back to the 1930s in various formats. Really all you need is state income distributions and vote outcomes.

Yeah, me too. Apparently it’s the new post-materialist politics.

On the misperception of black numbers and income, apparently that shows up consistently for minorities across a wide range of cultures, so I think blaming it on media is a bit much. That first link shows Hungary wildly overestimates the number of immigrants, which I’m pretty sure is not due to immigrant overportrayal on until-recently-state-communist television. It’s interesting data though, I can’t find really in-depth discussions on why.

A couple more interesting bits:

Restricting to only white voters removes half of the rich-poor voting gap.

Jamie Galbraith makes the interesting case that all the income inequality jump is incredibly concentrated in just a few counties.

County data confirm this: The big income winners in the late 1990s were concentrated in just four counties–Santa Clara, San Francisco, and San Mateo in California (all in the environs of Silicon Valley), and King County in Washington (Microsoft) as well as in Manhattan, the home of the bankers who made it happen. Take the big tech counties out, and the rise in inequality between counties in the late 1990s disappears.

Well, yes, if you get rid of all the rich people, inequality does indeed disappear. I bet you could remove Atlanta, Philadelphia, Detroit and Oakland and halve the number of black people in this thingie.

Well, I’m more than a little surprised there’s that much of a voting gap in rich asians, blacks, and hispanics.

Why are you so sure? I don’t know the Hungarian media but it’s certainly conceivable that they run a lot of scare stories about immigrant crime etc. Overrepresentation in the negative rather than positive sense.

I’m just guessing like you are. :)

You write outlines before posting on the internet?

I seriously don’t understand how you have the time for all of this. Are you on meth or something?


Once I was done reading it took maybe a half hour to flip through and write all that down.