Charles Murray [perhaps] does a Tucker Carlson, provoking me to unleash the usual torrent of graphs

Charles Murray wrote a much-discussed new book, “Coming Apart: The State of White America, 1960-2010.”

David Frum quotes Murray as writing, in an echo of now-forgotten TV personality Tucker Carlson, that the top 5% of incomes “tends to be liberal—right? There’s no getting around it. Every way of answering this question produces a yes.”

[I’ve interjected a “perhaps” into the title of this blog post to indicate that I don’t have the exact Murray quote here so I’m relying on David Frum’s interpretation.]

Frum does me the favor of citing Red State Blue State as evidence, and I’d like to back this up with some graphs.

Frum writes:

Say “top 5%” to Murray, and his imagination conjures up everything he dislikes: coastal liberals listening to NPR in their Lexus hybrid SUVs. He sees that image so intensely that no mere number can force him to remember that the top 5% also includes the evangelical Christian assistant coach of a state university football team. . . .

To put it in graphical terms:



In blue America (where Charles Murray, David Brooks, David Frum, and I live), rich people are a bit more economically conservative then poor people but this is balanced by rich people being more socially liberal. As a result, rich and poor have similar voting patterns in the blue states.

In contrast, in red America (where Charles Murray and David Brooks locate the forgotten majority), rich people are both economically and socially conservative (at least they were in 2000, which is when the data for this graph came from).

America’s top 5% of income includes rich households in red, purple, and blue America—and, as the graph above shows, this represents a large variation in political views.

Another way to put it is that, as we say in our book, the culture war is not a battle between rich liberals and poor (or middle-class) conservatives or even a battle between rich conservatives and lower-income liberals. Rather, the culture war is between rich liberals and rich conservatives.

It’s not the Prius vs. the pickup truck, it’s the Prius vs. the Hummer.

There are more rich conservatives than rich liberals (just as there are more rich Republican voters than rich Democratic voters) but the minority of upper-income Americans who are rich do play an important role in our society. It’s worth spending some time thinking about rich liberals, but it’s also worth remembering that most rich Americans are not liberal.

I think Charles Murray is interested in religion too, so let me throw in this set of graphs that subsets the population according to religious attendance:

Just to hack at this a little more: data from 2000, 2004, and 2008 showing the income distribution of voters self-classified by ideology (liberal, moderate, or conservative) and party identification (Democrat, Independent, or Republican):

There are more rich conservatives than rich liberals.

And here are some maps of how different ethnic groups voted in 2008 (click for the large version):

The second row of maps gives the answer to your questions about white America, or at least those white Americans who vote. (The evidence is that, compared to voters, nonvoters have lower income and are more likely to favor income redistribution. So I don’t think that moving from “white voters” to “white America” would change our story.)

43 thoughts on “Charles Murray [perhaps] does a Tucker Carlson, provoking me to unleash the usual torrent of graphs

  1. In fairness to Murray, perhaps you ought to check the context of Murray’s quote. Frum may be distorting Murray’s opinion. After all, the rich are more liberal on social issues, when compared to middle-income folks in the same state.

  2. I want to agree with you, but most of what you show here is the distribution of income conditional on political ideology. This is different than the distribution of political ideology conditional on income, which is what Murray seems to be talking about. What is the fraction of the top 5% that is liberal, etc.?

      • I came to comment/ask about a similar issue (though with respect to states, not ideology). In, e.g., the first graph, are the rich people in solid republican vs. battleground vs. solid democrat states equally rich? Or is it the case that a larger proportion of the top 5% are in, say, solid democrat states than in battleground or solid republican states (or vice versa)?

        • A related problem: in the graphs in this post, are the upper 5% even represented? Or are the rich the top 33%? With data from surveys, sometimes the surveyer just drops the top 5% because there are too few people surveyed in that category and it would be misleading.

          My impression (which may be wrong) was that in the last election, the very rich (say, top 3%) voted Democrat and the rich (the next 15%) voted Republican.

  3. Good points, though in (partial) defense of Charles Murray (I just finished reading his book), he concentrated heavily on the “blocks” of “Superzips” (which included New York, San Francisco, and D.C.) The minority of wealthy people who are liberals are more heavily concentrated in these areas than (say) Dallas. It so happens these areas are the cultural, political, and economic capitals of the country, and it is possible for highly motivated, well-organized politcal minorities (in this case wealthy liberals) to “punch above their weight” in influencing the political direction of the society. Nonetheless, it’s always worth putting this matter in proper perspective.

    The culture war in a nutshell: one group of upper middle-class white people who claim to read books they don’t versus another group of other upper middle-class white people denying reading books they have :)

    Incidentally, did any one read about Goldman Sachs CEO Lloyd Blankfein running PSAs in support of gay marriage? Freakin’ hilarious…

  4. I am starting to believe that I am becoming old-fashioned or plainly obsolete. I thought the received behaviour was that, before one could comment on a book, let alone blog on it, he would actually read it. Apparently, now it’s ok to quote second-hand sources and focus on 1-line excerpts. Murray’s books somehow seem to elicit a visceral response especially in those who have *not* read them. A few days ago, Krugman. Now (sigh) Gelman. I am not pissed off: I am not a fan of the book, but my judgement is more nuanced than most. But I am disappointed, because I remember the time when you (prof. Gelman) used to do your homework more often.

    Regarding “Coming Apart”: the author makes clear that the new elite is not specifically liberal. This is an overall theme of the book, but is specifically covered on page 95ff. And Murray makes clear throughout that the Republican elite has abdicated his role of moral leadership. You may agree or not. Personally, I think it’s an idea dangerously close to Carlyle’s thought and I find it closer to proto-fascism that to libertarianism. But at least I am trying to respect the author.

    And I apologize for always leaving acerbic comments, when I do so. I have bee a big fan of the blog since its start, and feel more like a lover betrayed by sloppiness.

    • Gappy:

      I don’t see what was sloppy here. I very clearly wrote that I was commenting on what Frum wrote. If I were being paid to review Murray’s book, of course I would read it. As it happens, I don’t have a copy of Murray’s book and I’m not writing a book review, I’m explaining something about voting patterns.

      What I was doing in the blog entry was expanding on Frum’s point. In particular, the differences between red and blue states can help explain how Murray, like Carlson before him, could casually make an error about political attitudes of upper-income Americans. In short, things look different in red America than in blue America, and that could be useful to know.

      I have no visceral response to Murray; I just think there are some subtleties in public opinion that Murray might not be aware of. I’ve spend a lot of time thinking about income, political attitudes, and voting, and I think it can be a useful contribution to share these graphs.

      I’m hoping that Murray and his readers will see these graphs and get a more nuanced understanding of the differences in attitudes comparing Americans of different income levels.

      P.S. No need to apologize for the comments. I appreciate the chance to explain further.

      • I would agree with you if you had titled your blog entry “Frum saying Charles Murray does a Tucker Carlson”; or “Hey, have you people forgotten what a kick-ass book is “Red State Blue State””. Not the case here. You quote Murray in the title, yet your charts neither support nor refute any of Murray’s theses; moreover your view is cross-sectional, whereas his is temporal.

        “If I were being paid to review Murray’s book, of course I would read it” a valid excuse for critiquing a second-hand statement from a book one didn’t read? I think not. If someone misrepresented a statement in your book without reading it, would you consider it acceptable?

        • Gappy:

          You write that my charts “neither support nor refute any of Murray’s theses.” Huh? They definitely refute Murray’s claim that the top 5% of incomes “tends to be liberal—right? There’s no getting around it. Every way of answering this question produces a yes.” I never purported to evaluate any part of Murray’s book except for that sentence. Similarly, I never purported to evaluate any part of Tucker Carlson’s broadcast except for the one sentence of his. And I never actually watched Carlson’s show either.

          If I did misrepresent Murray (presumably because Frum misquoted him in some way), then I’m happy to be corrected. Until then, I’ll assume Frum quoted Murray correctly.

        • The whole point of my first post was that that a) Frum did not quote Murray correctly; and that b) you would have not incurred in the same mistake had you read the book. I am not stating that Murray is right either. He may be wrong, but for different reasons. Murray identifies the new upper class not only by income but also by educational level, occupation and domicile (“superzips”, mostly concentrated in blue states). He also associates (weakly) these upper class to socially liberal mores (more than economically liberal); not in contradiction with you first chart. In any case the descriptions of the new upper and lower classes in the book are far more nuanced than what Frum would make it to be.

          I think it’s fine to show charts from Red State Blue State once in a while. It’s also fine to discuss Frum’s statement as his own. I believe it a minor sin of sloppiness [like omission, only less grandiose] to rely on second-hand information without checking the source; especially considering that the sourcea $14 kindle download away and is a quick read. “I am not paid to write a book review” and “I’ll stand by that one-liner, I have no time to read a book” are excuses that would fit well in a David Lodge book. But hey, it’s a free country. I am sure Murray would at least agree with the last statement.

        • The quote is on page 44 of the book. I don’t know how to copy from Kindle. But to put the quote in the context *of the sentence* (let alone of the book): Murray asks the rethorical question: The new upper class “tends to be liberal—right? There’s no getting around it. Every way of answering this question produces a yes.”. And from the book, the top 5% and the new upper class are not one and the same. To your credit, you quote Frum, but Frum has it wrong.

  5. That first graph, if you ignore the poor and look only at the social axis, not the economic, it looks like richer people are always more liberal than the middle-class people in the same place. This (not seeing the poor, only considering the social axis) is probably how Murray experiences his circle of acquaintances.

  6. My review of “Coming Apart” is in the February issue of The American Conservative.

    As Gappy says, Murray’s book does not emphasize that upper reaches of the social stratum are particularly liberal or Democratic. He presents data to show they are pretty mixed in politics, except in the most influential metropolitan areas like DC, NYC, LA, and SF, where they are more liberal. (That’s where the people with the highest SES nationally tend to be found disproportionately.) I don’t think you would terribly object to his pages on this subject. Overall, partisanship is not a big theme in the book.

    What is a big theme in the book is that the upper 20% tend to live conservatively, but talk “nonjudgmentally.” Murray would like them to talk the talk as well as walk the walk about how to live, so that their example gets through better to the lower classes.

  7. Right, looking up Murray’s graph and text, he first looks at members of the House of Representatives on the ADA rating of liberalism and finds little difference between the country as a whole and the top 5% of zip codes when leaving out the Big Four metro areas (NYC, DC, LA, SF), where they are quite liberal. I have the prepublication version of the book, so my page numbers are different from Gappy’s, but you can just look at the last 4 pages of Chapter 3.

    • Some questions here because I do not have a copy of the book.

      What years are we talking about? Are we concerned with the present, or does Murray make reference to the past at all. Related to this how many zip codes are there in the top 5%, and where are they geographically located. This seems like cherry picking to me. Are the other 5% (not including the big four) more conservative than the rest of the country? Again I don’t know the literature, so if someone can inform me that would be appreciated.

      Does Murray or any other study actually prove that the Big Four metro areas are liberal?

      • In fact a quick wiki search generated that (currently) the big four are now NY, LA, Chicago and Fort-Worth. So the fact Murray is removing the “big four” is more worrisome to me than in the previous comment. I assume that if we take the two Texas metropolitan areas in the top 10 they might be more conservative than the mean. So, again it is unclear to me why the big four are being taken out. I doubt they can be considered outliers…

        • Regarding what are the Big Four metropolitan areas in terms of national influence, which is what Murray is concerned with, Murray writes in an endnote:

          “Including San Francisco with New York, Washington, and Los Angeles is a judgment call based on the enormous influence that the information technology sector has acquired in the last three decades, not just technologically and economically but culturally. To the CEOs of multi-billion dollar businesses who do not live in the cities I listed and are incensed at being omitted from the narrow elite, I can only observe that lots of large corporations could go bust without making a ripple on the national scene.”

          In general, Murray worked hard to anticipate objections, so rather than my continuing to transcribe answers, may I suggest you take a look at the book? Fans of Dr. Gelman will find much of interest in it.

        • No I think I will pass on the book. I have read a similar argument from Lawrence Mead in his most recent book. However he focuses on African Americans instead of the poor. In fact that the Mead book was published a few months ago is pretty funny.

        • This looks suspiciously like trying to draw a bullseye around where the arrow landed. I guess just looking at rich people wasn’t good enough because it didn’t give Murray the right people to sneer at.

      • If you want to read what Murray actually has to say about politics without buying the book, go to Amazon

        http://www.amazon.com/Coming-Apart-State-America-1960-2010/dp/0307453421

        click on “Look Inside” and search for the word “Laguna” as in “Laguna Beach” — that will take you to p. 95, where his discussion of Red and Blue SuperZips begins. I don’t think Dr. Gelman would find all that much to disagree with Murray’s statements like this:

        “But other places didn’t fit the stereotype [of elites being liberal]. Outside Beverly Hills, Santa Monica, and Malibu, Kerry [in 2004] didn’t get even a majority in the other wealthy Los Angele areas for which votes can be broken out. Moving farther south into the wealthy towns of Orange County, Kerry won Laguna Beach, but nowhere else, getting a meager combined 35% of the vote in Newport Beach, Aliso Viejo, Tustin, and Yorba Linda. On the East Coast, the towns of the new upper class in the SuperZips surrounding New York City were not particularly blue. …Moving away from the coasts, it becomes impossible to think of the new upper class as being predominantly liberal. “

  8. To all:

    As we discuss in Red State Blue State, you have to be careful when looking at aggregate statistics (whether at the level of states, counties, or metro areas) and making conclusions about individuals.

  9. Pingback: “The culture war is between rich liberals and rich conservatives” » mythago

  10. Andrew, would it be possible to revise those graphs (Plates 10 & 11) to include population represented by using line thickness? Since they represent voters, it would be nice to see how many are in each (sub)group.

    • It would also be useful to note how much of the population is in each of the 3 groups of states. If 90% of the population were in solidly Democratic states, that would make a difference. Also, I trust you have weighted states by population, since otherwise Wyoming is going to make Republican state rich people look a lot more conservative.

  11. Pingback: The Agenda « Verities and Vagaries

  12. Um, this set of comments got me thinking of Bayesian inference – no really. One can guess the priors (very roughly) of a number of the participants, set largely by political philosophy plus knowledge of Charles Murray’s track record (such as co-authoring the infamous The Bell Curve). But there is also a “likelihood” that relates to how much the commenters are prepared to adjust their views based on the arguments and clarifying comments of others. I’m sure this in Jaynes’ text somewhere :-)

    • Thanks D.O. I figured as much, just was lazy and did not want to do the leg work. Cherry picking data should be as penalized as plagarism. It is not taken as seriously as it needs to be.

  13. Pingback: Charles Murray on the new upper class « Statistical Modeling, Causal Inference, and Social Science

  14. Pingback: Overcoming the Merely Therapeutic | Koinonia

  15. Actually, south Orange County is a good point and upper middle class burbs in San Diego. A lot of those folks attend mega churches which is Murray’s theme that the upper middle class is more religous than the lower white class. Rick Warren is always criticized that his church has too many upper middle class whites.

  16. Persoanllity, I would not be surprise that now the upper middle class is both fiscal and social conservatives, teh 75,000 to 150,000 income bracket is where more people attend church and vote Republcian. And I’ll bet the 20,000 and under are in some ways the most liberal since they don’t go to church and don’t get married and have more kids out of wedlock. And I’ver met more people under 30,000 that believe gay marriage is ok than 10 years ago. So, maybe some of the sterotypes don’t hold anymore.

  17. Well, I dsagree that the poor among whites are that religious. I think the upper classes are more religious than most people think, because evangelicals complain that their churches are too much in affluence areas. Think Bill Hybels, Rick Warren and even Joel Osteen don’t have many in their congreation under the 20,000 unless they were laid off recently.

Comments are closed.