When does your income determine the way you vote?

Image for article titled When does your income determine the way you vote?

It's election day in the U.S., and Republicans appear set to make major gains in Congress. So why do Americans vote the way they do? The answer is all about income...to a point.


Historically, Republicans stand with business, while Democrats side with labor. It's a pretty coarse division - most political classifications are - but it should mean that, in general, richer people will tend to vote Republican, while poorer people will vote Democratic. And that pattern has more or less held true since the times of Franklin D. Roosevelt and the New Deal.

Except that seems to create a paradox - Democrats tend to do much better in richer states along the coasts, while Republicans clean up in the poorer states of the South and Midwest. So how do you square that with the income pattern I just described? According to research led by Columbia's Andrew Gelman, it's all a matter of geography.

In Democratic states, there's practically no difference between how the rich and poor vote, which tilts the scales back towards the left. In right-leaning states, the rich almost completely vote Republican. On a national level, the overall pattern doesn't look any different from how it did in the 1930s or 1940s, but that's just because two opposing, extreme regions happen to average together to the same basic trend.

Gelman and his coauthors explain:

"This trend has gradually developed since the early 1990s and has reached full flower in the elections of 2000 and beyond. As a result, richer states now tend to favor the Democratic candidate, yet in the nation as a whole richer people remain more likely than poorer people to vote Republican. How much more likely? In presidential elections, the share voting Republican has tended to be 5 to 20 percentage points higher among voters in the upper third of the income distribution than among voters in the lower third."

Trying to sort all these different variables and forces out is generally a quick way to get a headache. For instance, there's a fairly well-known correlations between income and level of education. If better educated people tend to be richer, and richer people tend to vote Republican, then better educated people should vote Republican, by and large. Then why is so much of the political narrative filled with talk of rich, coastal liberal elites?

Well, liberals might be elite (whatever that means), and yes, they're probably coastal, but they likely aren't rich. Barack Obama did very well in 2008 among highly-educated voters, but his success was greatest among highly-educated people making less than $75,000 a year and dropped off sharply the richer you got.


A favored conservative talking point is the idea that Democrats want to turn the United States into Europe. According to Gelman, there's actually a weird sort of truth to that, although definitely not in the way that argument is generally meant:

"We see an intriguing connection between these international comparisons and the variation between states in U.S. politics. When it comes to income and voting, the richer states of the Northeast and the West Coast-but not those in the southern and central 'heartland'-look more like Europe, with rich and poor voting similarly and Democratic and Republican voters separating themselves more on social than on economic issues."


As Gelman and his fellow researchers explain, economic policy has long been non-negotiable in Europe, because the welfare state is too popular to dismantle and too expensive to add to in any significant way. That's actually starting to change a little, although you only need to look the current situations in Britain and France to see just how difficult that process truly is.

The point is, even issues that are in some sense economic - such as how immigrants should assimilate into their new European homes - take on a much more social complexion, and that generally means you can't predict how groups will feel about an issue based on their incomes.


Democrat-leaning states obviously don't have a welfare apparatus in the same sense European countries do - although they tend to have more of one than their Republican counterparts - but even so, the same mechanics seem to be in place, making social issues a far more likely point of difference between Democrats and Republicans in left-leaning states. Again, that means income is less likely to play a role in one's political decision-making than it will in states where the main issues are economic.

There's a lot more to all of this - whenever you try to explain the voting habits of hundreds of millions of people, that's pretty much a given. For more insight into these trends - and what they might mean for Election Day - check out the original paper.



Chip Overclock®

In their book THE ART OF STRATEGY, which was a layman's introduction to game theory, Dixit and Nalebuff devote a chapter to why it is in the best interests of political parties to appeal to the fringe elements. Turns out the game theoretic math shows doing so captures both the fringe (the far right or left as the case may be), the non-fringe (left or right to moderate), and even some elements of the _opposite_ side, by taking extreme viewpoints. If you take a more moderate position, your target demographic can't easily discriminated between you and your opponent. It's kind of nuts, but it actually made sense. So you get this constant pressure to radicalize your public viewpoint, and the result is the Tea Party or Socialism.

Personally, I already voted, I'm registered as an independent, and self-describe as a pro-gun liberal.