Unprecedented numbers of Americans — including a growing share of Republicans and independents — recognize racial injustice as pervasive, support police reform and back the protests against police brutality. Apparently oblivious to this emerging consensus, Trump is trying to run a Nixon-style “law-and-order” campaign for the 2020 election. And it is killing him politically.
Trump has seen a dramatic and precipitous collapse among people who used to support him, according to an intensive study of swing voters by The New York Times. Critically, this shift does not seem to be driven by economic turbulence or perceived mismanagement of COVID-19; his approval ratings in these domains are equal to or higher than his general approval rating.
Instead, for many voters who have stuck by Trump up to now, the president’s grotesque response to the killing of George Floyd, and especially the protests that followed, seems to have been the main driver. Most Americans, among them significant numbers of Republicans and independents, agree that Trump is exacerbating racial tensions — and it is driving many towards his political opponents.
Critically, these defections are occurring primarily among white voters. According to the New York Times data, the president’s margins among minorities are holding fairly steady.
That Trump is seeing major losses among whites specifically as a result of his racially inflammatory behavior seems fairly inexplicable given the common narrative that whites who voted for Trump were motivated largely by racism. But in fact, that common narrative is wrong. The underlying dynamics at work here have been present from the outset: Trump’s racialized rhetoric and policies have always been a drag on his candidacy, including among many whites who have traditionally skewed Republican.
Here the reader may be perplexed. After all, there are thousands of studies ostensibly ‘proving’ Trump voters were driven by racism. Yet as I demonstrated in a recent paper for The American Sociologist, this literature is plagued by glaring errors, oversights and prejudicial study design.
In a typical example, an author purported to demonstrate that Trump voters were especially motivated by racial animus. In fact, his data showed that Trump voters were less “racist” than Romney voters — and less ‘authoritarian’ too (see pp. 7-11).
‘Driven’ by Race?
The ‘logic’ leading most scholars and journalists to associate voting for Trump with racism goes something like this: Trump said racist things. Most whites who turned out in 2016 voted for Trump. Therefore, Trump must have won those votes because of his racialized language. In making this leap, of course, people are ignoring the first rule of statistics: correlation does not imply causation.
To meaningfully talk about voters being ‘driven’ by race, we would have to be able to identify citizens who voted differently than they otherwise would have due to Trump’s positions on racial issues. At least one of two conditions would therefore have to be met, either they 1) voted when they otherwise would have abstained, specifically because of racial issues, or 2) they voted for a different party than they otherwise would have specifically as a result of racial issues.
Let’s start with the first condition. There’s no evidence that Trump mobilized people who would have otherwise sat the election out. Instead, voter turnout in 2016 ended up being the lowest in decades, as many who strongly opposed Hillary Clinton or the Democrats could not bring themselves to vote for Trump, either. White turnout was stagnant, with Trump actually winning a slightly smaller share of the white vote than Mitt Romney did in 2012 (driven largely by defections among white women).
Trump was able to win despite a lackluster performance among whites partly because he performed significantly better among Hispanics than Romney and got the largest share of the black vote of any Republican since 2004. Of course, Trump can probably claim little credit for this success with minorities — Democrats had been consistently shedding these voters for the better part of the preceding decade.
Nor did the GOP win a majority of white votes specifically because of Trump. Republicans have captured the lion’s share of white votes in every presidential election since at least 1972. The vast majority of those who voted for Trump in 2016 had supported Mitt Romney, John McCain and George W. Bush in previous years — all leaders who have condemned Trump’s rhetoric towards minorities.
Overwhelmingly, those who voted Republican in 2016 were not voting for Trump per se, but for the party (or against the Democrats). Had Trump critics Jeb Bush, Marco Rubio or John Kasich ended up as the 2016 nominee instead, they almost certainly would have cast their general election ballot the exact same way.
Indeed, if the primaries are any indication, most would have preferred to have been able to vote for some other Republican instead. Trump got the lowest primary vote share of any GOP nominee since 1968. He didn’t win because most Republicans wanted him to be the nominee, but because the anti-Trump vote was split between several other candidates that different voters found compelling (from February 1 through April 19, Trump typically won between a quarter to a third of votes in most contests; it was only after it became mathematically impossible for others to win that he began to pull in majorities, and even then, large shares of voters continued to cast ballots for other candidates in protest — even after those other candidates had dropped out!).
In fact, the most common negative word associated with Trump last cycle, even among Republicans, was “racist” (followed closely by “sexist”). As political scientist Matt Grossmann put it, “His negative statements about minority groups were recognized by voters — but not positively” (subsequent research has affirmed this finding).
In a nutshell, the first condition for claiming that Trump voters were driven by racism seems to fail in light of the available data. Let’s turn to the second: did voters switch from other candidates to Trump because they supported his racialized policies or rhetoric?
One famous study claimed just this — purporting to prove that Obama-Trump switchers were driven by a desire to preserve white privilege (reacting against “status threat”). One problem with this narrative, of course, is that many of the most critical Obama-Trump switchers in states like Pennsylvania, Wisconsin, Ohio, etc. were not whites, but African Americans and Hispanics (again, Republicans increased their margins with these demographics in 2016 while holding steady with whites). To put it mildly, these voters probably did not switch out of a desire to shore up white privilege/ white supremacy.
The study itself was riddled with glaring errors that somehow made it past peer review — and was widely cited in the media and academia, with almost no one noticing the problems. For instance, the author claimed Trump voters were especially hostile to immigration. In fact, her data showed that Trump voters were warmer on immigration than Romney voters, and that immigration was on balance a losing issue for Trump (see pp. 20-26).
Immigration remained a losing issue for the GOP through the midterms. According to a post-election report by The Winston Group, a Republican polling company, Trump’s hardline posture against refugees and immigrants in the leadup to the midterms was a factor in Republicans losing the House. Critically, the attrition the GOP experienced in the 2018 midterms came primarily from more wealthy, educated, urban and suburban whites.
Trump’s problems with white voters have only deepened since. For instance, many have claimed that the white working class was especially important for securing Trump’s 2016 victory. In truth, the role of these voters has often been greatly exaggerated — but regardless, Trump is seeing erosion among these voters (especially working class white women and young people). Nearly 80 percent of white evangelicals supported Trump in 2016; only 62 percent approve of him now. In 2016, Trump won white senior citizens by huge margins. This year, he’s running neck-and-neck with his Democratic rival among seniors.
That is, to the extent that Trump’s racialized rhetoric and policies are getting people to vote for a different party than they otherwise would have, it’s because he has been convincing Republican-leaning whites to vote Democrat!
In fact, not only are whites turning on Trump more than any other group, his racially inflammatory words and actions seem to be leading some to rethink their own racial identity. Research by political scientist Ashley Jardina shows that identification with whiteness has actually declined over Trump’s tenure. Whites have also grown less likely to express prejudiced attitudes about blacks and Hispanics.
In short, to the extent that Republican votes are being ‘driven’ by race, it seems to be in the opposite way from what the prevailing narrative would suggest. Rather than energizing people who would have otherwise sat out, Trump seems to have suppressed turnout in 2016 among white Republican-leaning voters. Rather than driving defections toward Republicans through his racialized rhetoric, he seems to be driving white Republican-leaning voters into the arms of his opposition.
Trump Doesn’t ‘Get’ His Own Base
Of course, it is not just journalists and scholars who misunderstand the situation. Despite mocking academia and the media as biased, the president seems to have fully bought into their caricature of the Republican voter. As Vox’s Jane Coaston aptly put it, “Trump campaigns like a Manhattan liberal parody of a conservative.” He keeps giving his voters what he thinks is red meat, and they continue to recoil in horror and abandon him for it.
This is a dynamic that has been present since the 2016 primaries, but which people have been unable to recognize up until now because they’ve been so firmly locked into the idea that racism was the key to Trump’s success. Instead, Trump may lose in 2020 because his racialized policies and rhetoric have alienated key white constituencies that previously skewed Republican.
When Mitt Romney lost in 2012, the GOP conducted an intensive ‘autopsy’ to find out what went wrong. One of their core findings was that the party needs to be more welcoming and inclusive, and to strike a more positive and hopeful message. It turns out, this is not just important for attracting minority voters – building this kind of ‘big tent’ may be even more important in helping them retain bourgeois whites.