“The Party of the American Working Man and Woman”

by Larry Bartels on June 18, 2012 · 7 comments

in Campaigns and elections,Political Parties

Jonathan Haidt writes that “the Republicans are increasingly becoming the party of people who currently hold jobs.” He really means white people without college degrees who currently hold jobs. Or maybe southern white people without college degrees who currently hold jobs.


How did a nice psychologist like Haidt wander into this minefield? A couple weeks ago, Haidt suggested that “When working-class people vote conservative, as most do in the US, they are not voting against their self-interest; they are voting for their moral interest. They are voting for the party that serves to them a more satisfying moral cuisine.” Andrew, among others, vigorously challenged the premise that “working-class people vote conservative,” which seems to have inspired Haidt to brush up on some political science literature on trends in partisanship and voting behavior.

Haidt settles on the definition of the white working class offered by Ruy Teixeira and Alan Abramowitz, which includes white people without college degrees who are currently employed. (Excluding retirees and homemakers makes some difference, but not a lot. Focusing on education to the exclusion of income matters more; the resulting “white working class” is more affluent than the rest of the public. And of course, the big switch here is from “working-class people” to the white working class.) Using this definition—and focusing on party identification rather than voting behavior—Haidt concludes that, yes, “the white working class is leaving the Democratic Party.”

Ah, the white working class, yet again. I wish some psychologist would study why this particular topic generates so much interest and emotion. Perhaps it has something to do with the resonance of the term “working class”—and with the flexibility of that term in successive iterations of the debate.

For what it’s worth, the trends among Haidt’s new-and-improved white working class seem to me to be entirely consistent with what I reported in  “What’s the Matter with What’s the Matter with Kansas?” and in Chapter 3 of Unequal Democracy—namely, that there has been a decline in Democratic attachment among whites without college degrees, heavily concentrated among southerners and middle- and upper-income people, and more for party identification (a lagging indicator of responses to earlier political changes) than for presidential voting behavior. More specifically, using Haidt’s definition of the white working class since 1975 (avoiding McGovern and Watergate) and the ANES data:

(1) The “working white working class” has become more Republican by about 3 points per decade on a 100-point party identification scale.

(2) This shift is more than twice as large for southerners as for non-southerners, and about 40% larger for middle- and upper-income people than for low-income people.

(3) Overall, there has been no discernible trend in presidential voting behavior among the “working white working class.”

(4) While southern working whites without college degrees have become more Republican in their presidential voting behavior (by 4.5% per decade), non-southern working whites without college degrees have become more Democratic (by 1.6% per decade).


Haidt concludes that the political behavior of the white working class “depends [on] how you define class, and it depends on several moderator variables, including employment status.” Well, yes; but it depends much more on region—and much, much more on ignoring people who don’t happen to be white.

“If Democrats want to be the party of the American working man and woman,” Haidt writes, “they should first figure out whether they are in fact losing the working white working class, and if so, why.”

I would have said: If Democrats want to be the party of the American working man and woman they’re currently doing pretty well, averaging 54% of the major-party vote among all employed people in the past five presidential elections. (The Democratic vote share among people not currently working is 57%; hardly a momentous social divide.) But that way of putting it requires an acknowledgement that, nowadays, “the American working man and woman” is more often than not a college graduate, an African-American, or a Latino.

P.S. While Haidt’s new post defends the premise of his claim about working-class voters, he tones down the substance of his earlier argument about the crucial political role of moral values, writing that his critics “are probably all correct when they say that economic concerns played a stronger role in recent electoral shifts than the sorts of moral/cultural issues that I and Thomas Frank were talking about.” I would just add that that seems to be the case regardless of whether the (white) working class is defined by income (as in the unpublished paper of mine cited by Haidt) or education (as in the published version of the same piece).

UPDATED: Revised calculations using corrected ANES sample weights.

{ 7 comments }

Comments on this entry are closed.

Previous post:

Next post: