Ruy Teixeira is a senior fellow at the Center for American Progress Action Fund.
By Ruy Teixeira | Sep 03, 2019
Tom Edsall’s most recent Times column claims, “We Aren’t Seeing White Support for Trump for What It Is.” The column is based heavily on a paper by political scientists Herbert Kitschelt and Philipp Rehm entitled, “Secular Partisan Realignment in the United States: The Socioeconomic Reconfiguration of White Partisan Support since the New Deal Era.”
Edsall’s central takeaway from the paper is:
Kitschelt and Rehm find that the surge of whites into the Republican Party has been led by whites with relatively high incomes—in the top two quintiles of the income distribution—but without college degrees, a constituency that is now decisively committed to the Republican Party.
Based on the actual data in the paper, there are several problems with this characterization—though, given that the paper authors themselves encourage this sort of interpretation, Edsall can be forgiven for his take.
Start with the basic categories the paper authors use. There are four categories, both overall and for white voters: low education/low income (LE-LI); low education/high income (LE-HI); high education/low income (HE-LI) and high education/high income (HE-HI). Low education is less than a BA; high education is a BA or more; low income is the bottom two-thirds of the income distribution; high income is the top third.
The category the authors focus on the most is low education/high income, with the implication that such voters are not really “working class.” I beg to differ. One does not have to be low income to be working class. A household income of, say, $90,000 should not eject one from the working class—it merely means you’re somewhat better off than your poorer brethren. Not that $90,000 suggests such a fabulous standard of living, especially for older workers supporting a family, who are undoubtedly heavily represented in this group.
Definitional issues aside, the real problem here is that the authors’ data does not really show what they imply. First, look at the difference between the LE-LI and LE-HI groups in terms of occupational mix (Tables D1 and E3 of the paper). LE-LI is 76 percent blue and lower white collar plus mixed service functionaries (basically clerical jobs) plus low service functionaries. LE-HI is 61 percent from the same basket of jobs (and note that there is significant representation here of independent contractor-type construction workers and carpenters who aren’t counted in the 61 percent). So, different, but not that different.
Second, consider the size of these two different groups. The LE-LI is over half (!) of voters, while LE-HI is a mere 16 percent. Given the relative size of these two groups, it would only make sense to say the LE-HI group was “leading” the stampede to the GOP/Trump if there was little change in the much larger LE-LI group and much sharper change in the smaller LE-HI group.
But this is decidedly not the case. It turns out that (Figure 3) the white LE-LI and LE-HI voters had identical pro-Trump voting patterns in 2016. Moreover, since 2008 the figure shows a remarkably precipitous group in Democratic support among LE-LI voters. Hmm. It seems like, given the very large size of this group and this precipitous drop, it is the LE-LI group that is “leading” the move toward Trump, not the more affluent group.
All that said, I would be tempted to recommend that you read this paper because there really is a lot of interesting information in it. However, be forewarned that the paper is larded with enough political science jargon to sink a battleship. I speak the lingo so can slog my way through it without too much difficulty; your experience may be different.
As a final thought, I see the erroneous interpretation of this paper as feeding into the ongoing effort to depict Trump’s supporters as not having any “economic anxiety.” Hey, look they’re doing fine, it must be that they’re just racists! I note that this view is usually promulgated by people whose household incomes are far above those of the even relatively-well-off white working class. “Economic anxiety” can be considerably broader than not being able to pay your bills this month or put food on the table. To understand our fellow Americans, that’s an important thing to keep in mind.