More on the Politics of the Super-Rich

by Larry Bartels on May 27, 2012 · 5 comments

in Data,Political Economy,Political Parties

Andrew argues, based on “extrapolation from preferences of the top 5%, data on campaign contributions, and data on political attitudes of the top third of income,” that “there are lots more rich and powerful Republicans” than Democrats. While extrapolation from the top third, or even the top 5%, to the “super-rich” seems perilous, some new data on campaign contributions handsomely support his claim.

Stanford political scientist Adam Bonica has done terrific work mining public campaign donation records for insights into the behavior of campaign contributors. Using a scaling algorithm similar in flavor to those often applied to congressional roll  call votes, he has mapped more than 50,000 candidates for federal and state offices and more than 11 million distinct campaign contributors on a “liberal-conservative” dimension.

Among Bonica’s millions of contributors, he has recorded $460 million in contributions from 377 of the Forbes 400 richest Americans. In a presentation at Stanford earlier this month, he compared the ideological distribution of his Forbes 377 to the distributions of Democratic and Republican candidates.


The (dark grey) distribution of super-rich contributors has three clear modes. The most numerous sub-group overlaps the distribution of Republican candidates (though mostly not the most extreme conservative Republican candidates). The least numerous—but still appreciable—sub-group overlaps the distribution of Democratic candidates (though mostly not the most extreme liberal Democratic candidates). The middle sub-group consists of people who made significant contributions to both Democrats and Republicans.

While these data seem compelling with regard to the partisan alignment of the super-rich, they do not speak to Andrew’s additional claim that “rich Democrats tend to be moderate on economic policy, whereas rich Republicans tend to be highly economically conservative.” (Bonica maps candidates based on which donors they receive contributions from, and donors based on who they contribute to; there is no direct measure here of anyone’s policy preferences or voting behavior.)

For what it’s worth, I suspect that Andrew is mostly right on this score as well—but also that there is a great deal of politically significant variation even within the domain of “economic policy.” For example, the finance industry super-rich in Bonica’s data look, on average, much like the rest of the super-rich. However,  even those who contribute mostly or entirely to Democrats are probably not “moderate” on the issue of financial regulation—a fact that may be relevant to understanding why the regulatory response to the Wall Street meltdown was not more vigorous.

As Thomas Edsall notes in today’s New York Times, heavy reliance on contributions from Wall Street has different implications for Democrats and Republicans, given the two parties’ very different ideological centers of gravity. (According to Edsall, the finance industry is the largest single source of cash for both parties’ campaign committees, and accounts for more than 40% of the contributions to Romney’s super-PAC Restore Our Future.) ”Romney has made it very clear how he will help Wall Street if he wins,” Edsall writes. “It’s President Obama who is in a conflicted position. Obama could conceivably take a more aggressive regulatory posture. The more likely scenario, given recent history, is that he would be apprehensive about the consequences of regulating hedge-fund operators and investment bankers.”

{ 5 comments }

Andrew Gelman May 27, 2012 at 4:11 pm

Larry:

Thanks for the info. Bonica’s work seems consistent with earlier work by Thomas Ferguson on the political contributions of the richest Americans in earlier years.

Here’s what we had in Red State Blue State:

Probably the best evidence [about the political views of the richest Americans] comes from studies of political contributions. Political scientist Thomas Ferguson has tracked political donations of top corporate executives and the Forbes 400 richest Americans (or their equivalents, in earlier periods). The data presented in his 1995 book, Golden Rule, indicate that America’s superrich have generally learned Republican, but with some notable exceptions that have changed over time. Certain industries have persistently higher rates of contributions to the Democrats. In the New Deal, these included industries with a strong interest in free trade. Since the Reagan years, finance, and high technology firms have been much friendlier to Democratic presidential candidates than most of the rest of American business.

For 2004, Ferguson consolidated the lists of top executives and richest families into a lot of 674 firms and investors. Out of this list, 53% contributed to George W. Bush’s reelection campaign and 16% donated to Kerry, with Bush doing better among the oil and pharmaceutical industries and Kerry getting more from investment banks and hedge funds.

Larry Bartels May 30, 2012 at 9:13 pm

Adam’s analysis has gotten some well-deserved attention in the past few days, including in the Washington Post (http://www.washingtonpost.com/blogs/ezra-klein/post/the-politics-of-the-super-rich-in-one-chart/2012/05/29/gJQAlWvyyU_blog.html) and in the Los Angeles Times (http://www.latimes.com/news/politics/la-pn-study-finds-wealthy-donors-mostly-lean-toward-centrist-candidates-20120530,0,6030066.story).

The Times’ David Lauter sees daylight between Adam’s own characterization of his findings and mine:

“What one sees in the chart depends a bit on the eye of the beholder. In a comment on Bonica’s research, political scientist Larry Bartels of Vanderbilt University focused on the degree of overlap between the 400 and the Republican side of the chart. The data support the idea that the wealthy skew to the right, Bartels wrote.”

“Bonica, himself, stresses the other side of what the chart shows. ‘There’s this notion out there that billionaires are heavily skewed to the right,’ he said at the Stanford forum. ‘The data does not show that. They do lean a little bit more to the right,’ but overall ‘there seems to be a preference among the wealthiest individuals to support more centrist candidates.’”

I don’t see the phrase “wealthy skew to the right”–or, for that matter, anything else at odds with Bonica’s characterization–in my post. However, the confusion may reflect the fact that there are two distinct questions here.

First, how many of these super-rich people are Republicans (or rather, how much of their money goes to Republicans)? That isn’t obvious from the chart, but I’m sure Adam could generate an answer, and I’m sure the answer would support Andrew’s claim that “there are lots more rich and powerful Republicans” than Democrats.

Second, do “wealthy donors lean toward centrist candidates,” as Lauter’s headline has it? That’s a harder question to answer–not only because there is no direct measurement here of anyone’s views about any political issue, but also because most contributors’ “ideologies” are inferred from contributions to multiple candidates, often including a mix of (relative) centrists and extremists. However, it might be informative to compute the 25th and 75th percentiles of the distribution of contributors and provide some examples of prominent elected officials who are classified at those spots. If they are reasonably described as “centrists,” then it would seem reasonable to describe most super-rich contributors as “centrists” as well.

gogole June 2, 2012 at 5:02 am

According to the latest stats I could find, the top 5% of households had an AGI > $154k/yr. When exactly did $154k/yr become “super-rich”? In a major metropolitan area, a household consisting of a high school teacher and a police officer could easily be in the lower end of that group.

I’m not a big fan of using income measures (as opposed to net worth measures) to define the “rich” in general, but if we’re going to use them then we need to do a better job than this. I recognize that the data may not be available, but then we should either attempt to collect the data or simply say we don’t know rather than “extrapolating”. Lumping together the political preferences of households with $154k in annual income from wages/salaries with households that have annual incomes of 3-5+ times that amount from passive investments doesn’t tell us anything anything at all as the demographics have little in common.

Andrew Gelman June 2, 2012 at 7:22 pm

Gogole:

I think you’re misreading here. Neither Larry (in the post above) or me (in my earlier post or my comment above) referred to top 5% as “super-rich.” What we see is a bunch of different strands of evidence, all pointing the same direction.

Anonymous June 4, 2012 at 2:04 pm

You mean that people with higher incomes prefer the party of lower taxes? There seem to be a lot of bloggers delighted at their own intelligence in proving a truism.

Why not focus on what kind of person donates to a party against self-interest? Are they more noble or is the reason self-preservation? How many of the Forbes 400 live in Democrat controlled urban environments? How many find a quid-pro-quo in these donations?

Comments on this entry are closed.

Previous post:

Next post: