In September, Zack Beauchamp penned a long essay for Vox entitled “White Riot: How Racism and Immigration Gave Us Trump, Brexit, and a Whole New Kind of Politics.” Fiddling with a heap of sociological data purporting to show a strong correlation between “racial resentment” and right-wing voting, Beauchamp announced that “racism, Islamophobia, and xenophobia” are what unite “far-right politicians and their supporters on both sides of the Atlantic” — “far right,” in the case of the United States, being Donald J. Trump.
Beauchamp’s was one of the more substantive contributions to what long ago became liberals’ preferred explanation for Trump’s inexplicable ascendancy: He is a racist and rose by exploiting widespread, latent bigotry — and now he has ridden that wave all the way to the White House. This narrative was refined to its purest form on Election Night, when Slate columnist Jamelle Bouie, ever a hammer looking for a nail, suggested that Trump’s campaign and victory were a recapitulation of the assault that “angry, recalcitrant whites” waged on blacks in the wake of Reconstruction.
Election Day numbers, unsurprisingly, tell a very different story.
Start with the exit polls. Nate Cohn, data whiz for the New York Times’ “Upshot” blog, rightly observed that “Trump won the presidency by riding an enormous wave of support among white working-class voters.” But it’s where those white voters are from that is striking. “Youngstown, Ohio, where Mr. Obama won by more than 20 points in 2012, was basically a draw. Mr. Trump swept the string of traditionally Democratic and old industrial towns along Lake Erie. Counties that supported Mr. Obama in 2012 voted for Mr. Trump by more than 20 points.” Similar pockets can be found across the country. Maine’s second congressional district, which allocates its own electoral vote, opted for Obama by eight points in 2012, but went for Trump by twelve. And, of course, these sorts of counties were instrumental in helping Trump flip recently Democratic states. Among the 2012 blue states that ended in Trump’s column were Wisconsin, which last voted for the Republican nominee in 1984, and Iowa (also 1984, but with an anomalous Bush victory in 2004). As of this writing, Michigan, which Obama won by almost ten points in 2012, has not been called, but Trump holds a small lead.
How did this happen? Exit polling, taken together with turnout numbers, suggests different causes in different places. In Wisconsin, for example, Trump performed about as well as Mitt Romney in 2012. However, Clinton voters did not materialize; compared with President Obama in 2012, she received approximately 200,000 fewer votes in Wisconsin. Milwaukee alone dropped 40,000 Democratic votes. Consider that 80,000 Democratic votes also failed to show up in Detroit, and it’s clear that part of the reason for Trump’s victory in these states was Clinton’s failure to hold on to Obama’s voters among urban blacks. Something similar seems to have played out in North Carolina, despite the fact that Clinton outperformed Obama in mostly black Guilford, Mecklenburg, and Wake counties (where Democrats claimed that Republicans were engaging in “voter suppression”).
The scene elsewhere looks different. Clinton outperformed Obama in Pennsylvania, running up large margins in the suburbs of Philadelphia; she won Chester County, which Obama lost, by 25,000 votes. But she still failed to carry the state. Why? Thousands of white working-class voters who live in those industrial counties Cohn identified and who voted Democratic in 2012 voted Republican in 2016. Trump also appears to have mobilized white working-class voters who did not vote last cycle but did so this time in numbers sufficient to build up large vote totals in the state’s most reliably Republican areas.
All of this complicates the simple narrative of racial bigotry, and it becomes especially improbable as the lens widens. Take the nation as a whole, and exit polls suggest that Trump actually underperformed Mitt Romney among whites, 58 to 59 percent, while outperforming him among minorities — by two points among blacks and Latinos, and by three points among Asians. Meanwhile, the white share of the national vote decreased, from 72 to 70 percent.
As sociologist Jonathan Haidt has observed, if we define the white working class by level of education rather than income, it is clear that this voting bloc has been steadily tilting away from Democrats and toward Republicans since the 1970s. Al Gore lost whites without a college degree by 17 points, and John Kerry lost them by 23 points. Barack Obama, however, outperformed Kerry by five percentage points — meaning that, if white working-class voters are closet racists, many of them took a sudden holiday from their bigotry in November 2008.
What accounts for this long-term shift, and why do Bouie, Beauchamp, and others on the left think that the white working class misperceives its own economic interests?
Bouie, writing at Slate after the election, theorized that Barack Obama’s success among future Trump voters was simply a subtler form of racism. Part of the challenge of being black in America, he suggested, is “navigating the reality of white Americans who show kindness and care in one breath and say nigger in the other” (italics in original). Such people surely exist, but this is armchair psychologizing posing as explanation.
Beauchamp at least marshals some statistics to his armchair. In 1971, political scientists David O. Sears and Donald R. Kinder proposed the idea of “racial resentment,” a subtle form of racial prejudice that blends hostile attitudes toward black Americans with values characteristically associated with political conservatism, such as an emphasis on self-reliance. Beauchamp cites work in this vein by, among others, Hamilton College government professor Philip Klinkner, whose examination of recent American National Election Study data showed that support for Trump was, as Klinkner wrote at Vox, most firmly “rooted in animosity and resentment toward various minority groups, especially African Americans, immigrants, and Muslims” — not, among other popular hypotheses, in economic anxiety. This lines up with a substantial volume of empirical data that finds that high racial-resentment scores as measured by the standard battery of tests are positively correlated with conservative policy preferences.
But what “racial resentment” actually indicates is disputable and disputed. In their recent working paper “Conservatism, Just World Belief, and Racism: An Experimental Investigation of the Attitudes Measured by Modern Racism Scales,” Riley K. Carney and Ryan D. Enos, both of Harvard’s government department, compared rates of “resentment” toward blacks with rates toward non-blacks, using a simple test of their own design. They found that conservatives express similar rates of racial resentment regardless of the racial group at issue. Conservatives are about equally likely to resent whites and blacks — or Bhutanese or Nepalese or Lithuanians. According to Carney and Enos, what racial-resentment tests actually seem to show is the predilection of conservatives toward “just-world belief,” the “general political orientation that perceives the world as consisting of people who work hard and those who do not.” Conservatives, the authors suggest, tend to “apply this same just-world belief regardless of whether they are asked about Blacks or non-Black groups.” Carney and Enos make allowance for the reality of racial animus and acknowledge that just-world belief may cause conservatives to ignore genuine historical and contemporary injustices to black Americans, but their findings clearly subvert simple-minded narratives such as Beauchamp’s — to wit, that racial resentment means “fear [of] the demographic trends that would further erode the foundations of white privilege.”
Haidt and others offer a more nuanced and satisfying account. “When working-class people vote conservative, as most do in the U.S., they are not voting against their self-interest,” Haidt wrote in the Guardian in 2012; “they are voting for their moral interest” — that is, they are more interested in voting for “a moral vision that unifies a nation and calls it to greatness” than in electing politicians who promise the protection or expansion of specific programs, as has been the main pledge of Democrats in recent cycles. First Things editor R. R. Reno has made a similar suggestion with a different vocabulary in interpreting Trump’s appeal: Our “politics of prosperity,” says Reno, which prioritizes economic gain, is giving way to a “politics of meaning,” a desire for cultural consensus and solidarity.
Haidt elaborated this view in The American Interest. Nationalists “feel a bond with their country, and they believe that this bond imposes moral obligations both ways: Citizens have a duty to love and serve their country, and governments are duty-bound to protect their own people.” When this bond is widespread, it facilitates “a shared sense of identity, norms, and history [that] generally promotes trust.” “Globalism,” by contrast, with its eschewal of national loyalties in favor of diversity, immigration, and cosmopolitanism — literally, “world citizenship” — fails to recognize the importance of this shared sense of identity. And while patriotism can verge into something ugly, globalism’s dedication to embracing “difference” often decays into the labeling of any skeptics as racists — on the whole, says Haidt, “a shallow term when used as an explanation.” Summarizing the moment, Eric Kaufmann, a demographer at the London School of Economics, writes that rapid ethnic change has separated “those who prefer cultural continuity and order from novelty-seekers open to diversity.” This thesis provides a much more durable explanation for what happened this year than does racial animus.
There is also an explanation from a hard-headed political perspective: Democrats are losing white working-class votes because they have given up entirely on trying to win them.
Writing in the New York Times in 2011, Thomas Edsall previewed the Obama campaign’s reelection strategy:
All pretense of trying to win a majority of the white working class has been effectively jettisoned in favor of cementing a center-left coalition made up, on the one hand, of voters who have gotten ahead on the basis of educational attainment — professors, artists, designers, editors, human-resources managers, lawyers, librarians, social workers, teachers, and therapists — and a second, substantial constituency of lower-income voters who are disproportionately African American and Hispanic.
The Clinton campaign pinned its hopes on the same coalition this year, underestimating (as did many number-crunchers on both sides) the extent to which Obama’s victories relied on white working-class support in the Midwest and Northeast. (One of the few people who did seem to recognize the need to retain these voters was Bill Clinton, but the campaign largely ignored his intuition.)
The Democrats’ failure should not be overly surprising. It aligns with their ideology. Increasingly under the influence of its most socially “progressive” elements, the Democratic party has adopted the mantra that — as signs at recent anti-Trump protests declared — “the future is brown” and held it up ipso facto as a marker of moral progress. Actress Lena Dunham, a Clinton supporter, recently filmed a video celebrating the eventual “extinction of white men.” The result is a party largely convinced that appealing to the white working class is not only demographically silly but morally regressive.
Democrats’ alarming racial opinions do not excuse Donald Trump’s. The president-elect’s campaign occasioned, and was entangled with, a virulent strain of white racism and anti-Semitism; Trump’s rhetoric about minorities ranged from careless (his approach to urban blacks) to slanderous (Mexican “rapists”); and his treatment of Judge Gonzalo Curiel was abominable. These are facts conservatives should not forget as they look for areas in which to make common cause with President Trump and to expand the electoral appeal of conservatism, especially among disaffected minorities.
But Trump’s abundant flaws should not be automatically ascribed to his voters. The same people who voted for Trump also elected Tim Scott and Will Hurd and Mia Love. They gave Republicans the legislatures in Minnesota and Iowa. They gave a Republican the governorship of Vermont. Democrats who hope to recoup those losses should start by trying to understand why they happened. The answer isn’t black and white.