(Settle in, folks, this is a long one.)
It was June 16, 2015. Reality television star Donald Trump announced his candidacy for president. For a supposed “populist,” everything about his declaration communicated gilded elite, as he descended a golden escalator at his ritzy Manhattan self-named tower. Yet while his gaudy displays of personal wealth might communicate “Republican,” his campaign, from the beginning, was anything but business as usual.
“When Mexico sends its people, they’re not sending their best,” Trump said during his announcement. “They’re sending people that have lots of problems, and they’re bringing those problems with us [sic]. They’re bringing drugs, they’re bringing crime, they’re rapists. And some, I assume, are good people.” Trump would, from the very beginning, say the quiet parts out loud, eschewing veiled appeals to racism, so-called “dog whistles,” for overt out-and-out racism.
This was, of course, a party built on a foundation of racism, fueled by defections from the Democratic Party following President Lyndon B. Johnson’s signing of the Civil Rights Act. Just think about how remarkable this realignment was—white southerners embraced “The Party of Lincoln”—the party of the very president that defeated them during the Civil War! And all it took was the open embrace of racism that Democrats had finally begun to excise from within. Upon signing the landmark civil rights legislation in 1964, Johnson told an aide, “We [Democrats] have lost the South for a generation,” or maybe he said “I think we just delivered the South to the Republican party for a long time to come.” Accounts differ. But it turns out that “a long time” would certainly have been more accurate. The Deep South remains fervently locked in the GOP’s hands nearly 60 years later.
Well, almost. Demographic and cultural changes are eroding the power of racism and xenophobia in key regions of the country—including parts of the South. And while its allure and power remain strong, especially in the hands of a skilled demagogue like Trump, it was exactly his bigotry that sowed the seeds of his defeat. And thus we are entering into a new American frontier—one in which our nation is becoming more diverse, more urban, and more educated, while non-college whites left behind in economically devastated rural America end up especially susceptible—and responsive—to a message of hate.
What Trump discovered, to everyone’s surprise (including his own), was just how many of these disaffected whites exist (I called them the “hidden deplorables”). And while they only seem to turn out when Trump is on the ballot, the ability of future right-wing demagogic populists to activate them will determine the direction of our country over the next generation. And Republicans will certainly be motivated to bring them out.
On paper, “billionaire playboy” didn’t seem to be quite the profile to activate this latent strain of virulent American racism, especially given the Republican Party’s supposed fealty to its core values: Family values, national security, and lower taxes. On the tax stuff? Sure! But Trump was openly flirting with our supposed Russian foes, begging them to interfere in the election on his behalf. Meanwhile, a steady stream of women accused Trump of unwanted sexual advances. Stormy Daniels, an adult movie actress, was paid hush money for a sexual dalliance while Trump’s wife, Melania, was at home with their infant. And no one had to take these women’s words of Trump’s transgressions. He himself admitted so on tape, saying “When you're a star, they let you do it. Grab 'em by the pussy. You can do anything.”
And as for the Republican fetishization of the military? Trump attacked Sen. John McCain, who spent five years as a prisoner of war at the infamous Hanoi Hilton, even refusing an out-of-sequence early release. As a result of repeated torture, McCain suffered lifelong physical disabilities, and yet Trump said, “He’s not a war hero. He was a war hero because he was captured. I like people who weren’t captured.”
Ultimately, it didn’t matter if Trump sold out to the Russians, or mocked war heroes, or sexually assaulted women, or even that he was a terrible businessman, once bankrupting a casino, a business that literally prints its own money. His litany of bankruptcies and unpaid laborers and broken marriages and affairs were utterly irrelevant. All that mattered was his proven and undisputed record as a rank bigot. He certainly wasn’t faking that part.
You don’t have to search far to find examples of Trump’s racist past. In 1973, Nixon’s U.S. Department of Justice sued Trump for discriminating against Black tenants, including lying about whether units were available to prospective Black renters. In the ‘80s, a former employee of Trump’s failed casino told the New Yorker that “When Donald and Ivana came to the casino, the bosses would order all the black people off the floor … they put us all in the back.”
The former president of that Trump casino, John O’Donnell, wrote about Trump’s criticisms of his accountants in his book Trumped! “Black guys counting my money! I hate it. The only kind of people I want counting my money are short guys that wear yarmulkes every day,” O’Donnell quoted Trump as saying. ”I think that the guy is lazy. And it’s probably not his fault, because laziness is a trait in blacks. It really is, I believe that. It’s not anything they can control.” Lest you think that this was a he-said, she-said situation, Trump admitted to it all, telling Playboy Magazine in 1997 that “the stuff O’Donnell wrote about me is probably true.”
Still, of all the documented instances of Trump’s racism and bigotry, none beat his sorry involvement in the Central Park Five case in 1989, in which a white woman was sexually assaulted while jogging through Central Park. Five young Black men were arrested, charged, and convicted of the crime. The four youngest, all 14 years old, served six to seven years in juvenile facilities, while the oldest, who was 16 at the time, was tried and sentenced as an adult and served 11 years. They were all exonerated when another inmate in the prison system confessed to the attack, further confirmed by DNA evidence.
Throughout the racially charged investigation and case, Trump used his perch and money to fuel the flames of racial division, even running a full-page ad in four NY City newspapers—The New York Times, The Daily News, The New York Post and New York Newsday—demanding to “BRING BACK THE DEATH PENALTY.” For 14 year olds. "[NYC is] ruled by the law of the streets, as roving bands of wild criminals roam our neighborhoods, dispensing their own vicious brand of twisted hatred on whomever they encounter,” the ad claimed. "At what point did we cross the line from the fine and noble pursuit of genuine civil liberties to the reckless and dangerously permissive atmosphere which allows criminals of every age to beat and rape a helpless woman and then laugh at her family's anguish?"
Given chances to recant, given the exoneration of the five youths, Trump has steadfastly refused to concede that he was wrong, even claiming on the eve of the 2016 election that “they admitted they were guilty.” Of course they hadn’t, professing their innocence throughout their trial and incarceration. Yet that episode displays the seeds of what would become his signature campaign style, saying in that ad, "They should be forced to suffer and, when they kill, they should be executed for their crimes. They must serve as examples for their crimes. They must serve as examples so that others will think long and hard before committing a crime or an act of violence."
That dystopian Trumpian worldview was apparent in his 2015 book Crippled America. “Look at the state of the world right now. It’s a terrible mess, and that’s putting it mildly,” read the book jacket. “There has never been a more dangerous time.” And of course, it was explicitly reflected in that presidential announcement, in which the nation was being invaded by Mexicans “bringing drugs, they’re bringing crime, they’re rapists.” His first television ad of the campaign warned of mass hordes of brown-skinned immigrants invading our southern border, with the narrator saying, “He'll stop illegal immigration by building a wall on our southern border that Mexico will pay for." It didn’t matter that the images were actually of Moroccans crossing the border into the bordering Spanish enclave of Melilla in 2014, on the opposite side of the globe. (Spain has two such tiny cities on the African mainland.) Trump didn’t even need to traffic in reality. Whether it was 14-year-old Black kids or Moroccans desperate to enter the European Union, Trump would use whatever tools at his disposal to amp up racial fear and resentment.
Unfortunately, it was the right message for far too many people, giving Trump his narrow presidential victory in just a handful of key states. And leaving much of the nation shocked—how could this huckster two-bit idiotic bigot con this many people? How could we go, overnight, from having twice elected our nation’s first Black president, to someone who explicitly ran against our nation’s diversity? It was small comfort that his Democratic opponent, Hillary Clinton, had won the national popular vote by 3 million. The fact that nearly 63 million Americans had voted for him was distressing enough.
Yet almost immediately, there was hope: 4.6 million Americans joined the nationwide Woman’s March on January 21, 2017, believed to be the largest demonstration in American history. The 500,000 who marched in Washington, D.C., was double the crowd size for Trump’s inauguration. Unlike past liberal protest movements that fizzled from infighting and lack of singular focus, Trump was an easy villain to rally against. Almost a week from the Woman’s March, Trump signed an executive order banning foreign nationals from seven predominantly Muslim countries from entering the United States, while also blocking all refugees (including from Syria, then suffering an unimaginable catastrophe). Protesters again took to the streets as lawyers rushed to airports to help stranded Muslim travelers. Here was a different America than the one that had elected Trump—rallying unconditionally on behalf of the nation’s most vilified religious group.
This new wave of progressive activism created a new movement in Democratic-led cities to begin purging their public spaces of racist Confederate traitors, slave owners, and other stains on our national history. In response, on August 12, 2017, white supremacists, neo-Nazis, and Klansmen marched on Charlottesville, Virginia, protesting a proposal to remove a statue of Confederate general Robert E. Lee from a city park. One angry racist protester rammed his car into a group of liberal counterprotesters, murdering a 22-year-old woman, Heather Heyer. The following day, another round of progressive national protests and vigils erupted, as this growing “resistance” movement amplified the removal of Confederate monuments and other symbols of hate and racial division throughout the country. Trump responded by claiming there were “very fine people, on both sides." And to be clear, that included the Nazi side.
If Trump’s racism was expected, the national reaction was encouraging. Polling consistently showed him in negative approval territory—a place he would inhabit his entire presidency. An analysis by political data site FiveThirtyEight determined that Trump was the most unpopular first-year president since the advent of polling (which was 1945). It wasn’t even close. Trump’s average approval was 40% favorable, 55% unfavorable, or net favorability rating of minus-15. The next worst president was Gerald Ford in 1974, with a 44-39 favorability rating, or plus-5. In other words, Trump was 20 points more unpopular than the second-most unpopular president in polling history!
His numbers weren’t just atrocious. Trump was doing nothing to improve on them. From his first day, he was 100% fixated on speaking to his base of support, rather than expanding and growing it. It was a curious strategy, given that his first victory had been so tenuous—not only did he lose the popular vote by three million, but flip 77,744 total votes in Michigan (10,704), Wisconsin (22,748), and Pennsylvania (44,292), and Hillary Clinton would’ve won the presidency. Trump had no national mandate, but he governed as he did. And almost from the start, there was one demographic group that was becoming increasingly uncomfortable with Trump’s appeals to racism, bigotry, and xenophobia.
Enter the college-educated suburban white woman.
Democrats made big gains in Suburban America in 2016 despite nominating Hillary Clinton, an unfairly maligned and polarizing candidate. In 2016, Republican nominee Mitt Romney defeated Barack Obama by almost 16 points in Texas. Four years later, Hillary Clinton only lost the state by 9 points as suburban voters (and mostly women) shifted over to the Democratic party in suburban Dallas and Houston. In Georgia, the historically GOP bastion of Cobb County in suburban Atlanta went from a 12-point Romney win in 2012 to a two-point Clinton win in 2016, presaging Georgia’s eventual Democratic shift in 2020.
As the full scope Donald Trump's horrific presidency came into focus, that suburban shift among those college-educated suburban white women only accelerated. The first outward sign of this trend came in Georgia's 6th congressional district, an Atlanta suburban district long considered safe territory for the Republican Party. The district’s congressman at the time, Rep. Tom Price, had been nominated by Trump to run the Department of Health and Human Services. First elected in 2004, with zero Democratic opposition, Price easily won reelection—when opposition even bothered showing up. In 2016, even as the state’s suburbs had begun to move, he still won comfortably with 62% of the vote.
So when that seat opened up thanks to Trump's nomination, there was little reason for Republicans to fear losing their hold on the seat. But those suburban trends were well on their way to reshaping the region's politics, and the resistance geared up immediately to try and take advantage of the situation, attempting to deal Trump a black eye. Daily Kos was the first to spot this opportunity, raising the first million dollars for the Democratic nominee, Jon Ossoff. Even as tens of millions flooded into the district ahead of its special election, few expected the contest to be close. Yet on June 26, 2017, the political world was rocked when Republicans held on by the narrowest three-point margin against Ossoff (who is so close to winning a Senate special election in Georgia next month). The days of 35-point Republican victories was over, as would the GOP’s hold on the seat—as Democrats picked up the seat in the 2018 midterms, and held it in 2020 in the face of Republican gains elsewhere.
The two factors underlying these shifts have been the increasing diversification of suburban America, and a dramatic shift toward the Democrats among college-educated voters. Whereas immigrants once settled in urban America, as of 2010 over half of all new immigrants settled in Suburbia, and, ironically, did so in predominantly red states in the Sunbelt and rust belt states—Ariziona, Florida, Texas, Pennsylvania, and Ohio. Meanwhile the non-white suburban population has exploded in key Metropolitan regions, driven both by low-income service workers being pushed out of increasingly expensive inner cities, as well as wealthy people of color moving to more upscale suburbs. In the end, we have a different picture of suburban America than that of 50 years ago, when whites fled the cities to suburban enclaves out of fear and racism. In 1980, most non-white metropolitan people lived in cities. In 1990, over half of Asians in a metro region lived in the suburbs. Latinos reached that milestone in 2000, and Black metro residents did so in 2010. In fact, the largest 100 cities collectively lost 300,000 Black residents between 2000 and 2010. As a result, 81% of the suburban population was white in 1990, according to U.S. Census Bureau. By 2010 it was down to 65%. We still don’t have the results of the 2020 census, but you can assume that number is even smaller today.
But race alone didn’t explain the political shifts we were witnessing in these suburbs. And in 2016, it was clear what was happening—education had become as important in predicting a person’s vote as race, sex, and geography. It's hard to believe now, but in 2012 Republican Mitt Romney won college-educated voters by a 51-47 margin. In 2016, Trump lost those same voters by a 49-44 margin—a 9-point shift in a single presidential election cycle. That disparity was particularly stark among white voters—those with college degrees voted for Trump by a three-point margin, 48-45. Those without a college degree? 66-29, or a whopping 37-point margin. If you were to assume that those college-educated whites live mostly in cities and suburbs, while the non-college ones live in rural America, you’d be right. A Department of Agriculture report found that nearly 35% of urban adults had at least a bachelor’s degree, while the same was only true for 20% of rural adults.
Meanwhile, as educated voters moved strongly in the Democrats’ direction, so did non-college voters move toward the GOP. Trump won non-college voters 51-46 in 2016, a reversal of Democrat Barack Obama’s 53-46 margin in 2008—a 12-point shift in just two presidential cycles. The end result has been extreme national polarization along educational lines. Of the top 20 states with the highest percentage of college graduates, Hillary Clinton won 23 of them in 2016—the exceptions being Kansas (which is trending Democratic, but lacks a large urban center—Kansas City is across the border in Missouri), Montana, and Utah (for religious reasons). Meanwhile, of the bottom 30 states, Trump won 27 of them, the exceptions being Maine, New Mexico, and Nevada.
Pollsters, assessing their polling misses of 2016, one in which Trump exceeded the support predicted in polls, breathed a sigh of relief—all they had to do was factor education into their models, and all would be fine again! How could they have foreseen this educational realignment?
And thus we entered the 2018 election cycle, with the nation beset by racial turmoil and violence, and Donald Trump stocking that conflict with his very real ability to tickle that racist corner of the conservative lizard brain.
Republicans had long suppressed overt appeals to racism by using coded language, such as President Ronald Reagan’s “welfare queens,” or his announcing his presidency in Philadelphia, Mississippi—the place where civil rights freedom riders James Chaney, Andrew Goodman, and Michael Schwerner were murdered by the KKK. That’s not surprising given that Reagan was being advised by Lee Atwater, who later had stints running George H.W. Bush’s campaign and the Republican National Committee. “Y'all don't quote me on this,” he once told an interviewer, off the record. “You start out in 1954 by saying, ‘Nigger, nigger, nigger.’ By 1968 you can't say ‘nigger’—that hurts you. Backfires. So you say stuff like forced busing, states' rights and all that stuff. You're getting so abstract now [that] you're talking about cutting taxes, and all these things you're talking about are totally economic things and a byproduct of them is [that] blacks get hurt worse than whites.”
Trump stopped being abstract.
Indeed, a 2017 study by researchers Matthew Luttig, Christopher Federico, and Howard Lavine found that they could alter a Trump supporter’s view of housing policy merely by using images of Black people instead of white ones. “Among citizens with favorable views of Donald Trump, black racial cues increased opposition to mortgage assistance, anger at such assistance, and the tendency to blame policy targets for their own plight. In contrast, among citizens with unfavorable views of Donald Trump, black racial cues had the opposite effect: decreased opposition to mortgage assistance, anger, and individual blame,” they concluded. Supporters of Hillary Clinton weren’t affected by racial cues.
Another study, this one by political scientists Brian Schaffner, Matthew MacWilliams, and Tatishe Nteta in 2018, laid waste to the fiction that white voters had supported Trump out of some sort of “economic insecurity,” a risible notion given the GOP’s opposition to social net programs like Social Security, Medicare, Medicaid, and the Affordable Care Act. “[M]ost of the divide [between college-educated and non-college whites] appears to be associated with sexism and denial of racism, especially among whites without college degrees,” concluded the authors. “Attitudes on race and gender were powerful forces in structuring the 2016 presidential vote, even after controlling for partisanship and ideology.”
So 2018 quickly set itself up as a test of sorts—was this really the racist and bigoted America of Donald Trump and his base, or was it the pluralistic, diverse and tolerant America represented by its cities and, increasingly so, its suburbs. Trump clearly believed the latter, and doubled down on the very rhetoric and tactics that won him the White House in 2016. The Republican Party played along. It had no choice for two simple reasons: one, Trump had eschewed traditional politics by relying on his instincts and had won, and two, if they didn’t, a well-timed angry tweet from Trump would unleash all manners of pain. It simply wasn’t worth it for Republicans.
As in 2016, public polling suggested a resounding Democratic victory in the cards, but most people were gun shy after Trump dramatically outperformed those numbers during his own campaign. Pollsters thought their educational level adjustments had fixed things, but no one would really know until the new models were field tested with an actual election. The political and media class were cautiously optimistic about the polling industry’s ability to properly gauge the election, but the 2016 failures had one certain effect—it allowed Trump and his supporters to merely shrug off bad numbers as fake polls from the fake news, supposedly designed to depress Republican turnout. Thus, Trump could guide his party to run the campaign he wanted it to run, unencumbered by any evidence that it might be backfiring.
And it was a racist campaign, underpinned by two major scare tactics: hyperventilating hysteria about a caravan of Honduran migrants making its way to the Mexican-United States border, and the pervasive presence of menacing-looking Salvadoran MS-13 gang members in Republican TV and online ads, and direct mail.
The caravan was a perfect foil for Trump—the threat of that Moroccan horde from his first campaign ad made real, just south of America. The reality was tragically pathetic—4,000 rag tag refugees fleeing violence and economic devastation in their own land, desperate for any hope for a better life. The trip, by foot, was hundreds of miles, and as such, slow and treacherous. They had no papers, and thus no real hope to cross the border. They certainly weren’t going to slip through unnoticed. Trump couldn’t contain his glee, writing dozens of tweets attacking the caravan, such as, “Many Gang Members and some very bad people are mixed into the Caravan heading to our Southern Border. Please go back, you will not be admitted into the United States ... This is an invasion of our Country and our Military is waiting for you!” and “I am watching the Democrat Party led (because they want Open Borders and existing weak laws) assault on our country by Guatemala, Honduras and El Salvador, whose leaders are doing little to stop this large flow of people, INCLUDING MANY CRIMINALS” and “Anybody entering the United States illegally will be arrested and detained, prior to being sent back to their country!”
At one point, afraid that perhaps the fear factor wasn’t ramped up high enough, Trump even tweeted that “Criminals and unknown Middle Easterners are mixed in [the caravan].” (They weren’t.) The Thursday before the election, Trump actually called a press conference to declare that he was mobilizing the U.S. military to secure the border, a $200 million deployment without any practical application—other than helping set Trump’s narrative of a nation under siege. And the day before the election, Trump declared at a campaign rally in Ft. Wayne, Indiana, “Democrats are inviting caravan after caravan of illegal aliens to pour into our country, overwhelming your schools, your hospitals, and your communities. If you want more caravans, if you want more crime, vote Democrat tomorrow. ... If you want strong borders and safe communities, no drugs, no caravans, vote Republican.”
Of course it was all ridiculous posturing, but Trump’s base ate it up. Finally, someone was standing up to the “illegal” hordes. And Fox News happily fanned the flames. From mid-October to Election Day, the channel featured a “FOX NEWS ALERT” daily, and likely hourly, with such breaking news as, “The migrant caravan in Central America is growing,” as Fox & Friends co-host Steve Doocy breathlessly announced after one such alert interrupted his show. On another episode, guest co-host Pete Hegseth said, “when you see a lot of young men carrying the flag of their country to your country to break your laws, it looks a lot more like an invasion than anything else.” It was always growing. Fox News host Laura Ingraham declared that media was “highlight[ing] the wonderful camaraderie and welcoming spirit of the masses bent on breaking into our nation.”
A study by Media Matters, a media watchdown organization, found that Fox News had devoted over 33 hours to caravan coverage in the lead-up to the election—coverage that weirdly stopped the day after the election! Election eve the story got zero minutes on Fox, and on Election Day, less than five minutes. “Every election, there are a series of issues that rise to artificial highs and then, once the votes are cast, settle back down to normal noise,” former George W.Bush press secretary Ari Fleischer told the Associated Press. “Both parties do it; this isn’t some trumped up phony issue. The caravan will be back in the news once it gets closer to the border.” That didn’t happen, not returning to Fox News until Trump needed a new phony crisis to try and get funding for his border wall.
Even as Republicans fearmongered on the caravan, they realized that the narrative had a fatal flaw—those brown people were really far away. While the story appealed to conservative voters’ desire to “protect America,” it didn’t really hit them at home. Thus, they had a parallel story to sell—that of terrifying Salvadoran MS-13 gang members IN YOUR BACKYARD!
As far as gangs are concerned, MS-13 are bit players, a mere street gang having neither the resources nor clout of the Mexican cartels, the Italian mafia, or Russian and even Kosovan gangs in the United States. According to FBI statistics, there are between 8-10,000 MS-13 gang members in the United States, or less than 1 percent of all gang members. Florida International University researcher Jose Miguel Cruz called the gang’s presence in the United States “a federation of teenage barrio cliques that share the MS-13 brand” with no national or transnational leadership to corral them together. Their efforts to traffic narcotics across the US border have proven comically inept. InSight Crime, a publication tracking organized crime, even had a story titled, “5 Times the MS13 Tried—and Failed—to Become Drug Traffickers.” The article’s conclusion? “Melgar Díaz is at least the fifth gang leader who has failed to corral the MS13 into a united front across the United States to sell drugs on a massive scale,” its author wrote. “Of course, the gang sells drugs on a local level, but efforts to depict them as a drug trafficking organization, as US authorities want to describe them, are simply hyperbole.”
Still, if a ragtag crowd of desperate refugees is fodder for conservative fearmongering, MS-13 provides even better material. Their motto is “Kill, Rape, Control,” and prison photos of MS-13 gang members covered in elaborate demonic-looking tattoos are genuinely terrifying. No one would want to run into these thugs in a dark alley, or a lit one for that matter. And if they were too inept to forge international drug routes, who cares! If MS-13 were only responsible for around 35 annual homicides (according to one right-wing anti-immigrant organization), out of over 17,000 annual homicides, that was similarly irrelevant. They certainly looked like they could threaten people and their loved ones.
Furthermore, an Albanian or Russian gang member, or even worse, an Italian one, would do little to create the GOP’s dominant anti-immigrant rhetoric. MS-13 happened to be conveniently Latino, thus making it easier to equate immigration with death, violence, and fear. It was Republican catnip. Trump, himself, certainly couldn’t resist, saying about the gang, “They have transformed peaceful parks and beautiful quiet neighborhoods into bloodstained killing fields.” (They have. In El Salvador.)
A study by the Wesleyan Media Project found that one-quarter of all television ads run by Republican candidates warned against immigrant violence, heavily featuring imagery of MS-13 gang members. In the U.S. Senate race in Nevada, Republicans attacked the Democratic nominee, Rep. Jackie Rosen, for voting against “getting tough” on immigration. “Gangs like MS-13 exploit our broken immigration system and commit terrible crimes, horrific crimes,” thundered the ad. Another one in Arkansa’s 2nd congressional district, for Republican Rep. French Hill, claimed that “MS-13, the most dangerous gang infiltrating America, but Washington liberals want to get rid of ICE, the police enforcing our immigration laws and protecting our border from MS-13.” In Northern Virginia, Republican incumbent Rep. Barbara Comstock’s campaign claimed her Democratic challenger, Jennifer Wexton, was “out-of-touch [...] with the violent MS-13 gang threat and the victims they brutally target.”
There’s a third prong to this anti-immigrant rhetoric, all working synergistically with these appeals to national security and personal safety—the economic argument. You see, in the Republican telling, it’s all those “illegal” immigrants who are stealing your jobs, bankrupting your government, and taking advantage of American hospitality. A NY Times reporter, talking to diners in Mahoning County, Ohio, reported, “In a county that is 89 percent white and less than 2 percent Hispanic, they spoke of undocumented immigrants bankrupting Sun Belt hospitals, dragging down wages and burdening taxpayers.” That kind of logic is ridiculous on its face, of course. “He’s allowing these workers to say, ‘I don’t have a good job because of these immigrants. That’s not true. But he’s got a voice,” A local Democratic leader told that NY Times reporter. “There’s an underbelly of America that America doesn’t want to accept about itself, and he speaks to it.” And it works! In 2012, President Barack Obama carried Mahoning by over 28 points. Hillary Clinton won it only by only 3 points. It should come as no surprise that just 24% of people living in the county have college degrees. If the county was a state, it would rank 45th in college attainment.
Each one of these three prongs reinforced the other, giving conservative voters a grand unified theory as to why their lives were miserable, and Republicans gave it their all. Yet when the votes were counted after the 2018 midterm elections, Rosen had won, Wexler had won, and Democrats as a party enjoyed historic gains up and down the ballot. House Democrats picked up 41 seats and the majority—the third largest since the late ‘70s. Democrats won the national popular vote 53-45, or 9 million votes—the largest gap in a midterm election in American history. Democrats flipped seven governorships, six state legislatures, and hundreds of state legislative seats. While Democrats didn’t retake the Senate, they gained two seats in a map that had strongly favored Republicans—Democrats were defending 26 seats, compared to the GOP’s nine. By the math, America was back, a stunning repudiation of Trumpism—and one that continued into 2019 gubernatorial victories by Democrats in crimson-red Kentucky and Louisiana.
So the year 2016 had to be a historical aberration, right? A look at the polls confirmed that Trump, indeed, remained deeply unpopular outside of his conservative bubble. Early 2020 matchups against potential Democratic opponents had him trailing badly. And most importantly, demographic trends that had long predicted Democratic dominance were finally starting to appear in actual elections.
According to the U.S. Census Bureau, total turnout in 2018 was 53.4% of the voting-eligible population (which includes people who aren’t registered to vote, but are able to do so), the highest in 100 years. It was the first time since 1982 that the number had exceeded 50%. Turnout among eligible young voters (18-29 years old) was 36%, which yes, lagged all other age groups, but was dramatically higher than the 20% who voted in the 2014 midterm elections. Compared to that previous midterm election, the Black vote was up 11 points, while both Latinos and Asians were up 13.
Among the white vote, the non-college vote was down nearly three points, to 39% of all voters, from 42.8% in 2014. It had been 50% in 2006. Meanwhile white college grads slightly increased their numbers from 33.5% to 33.8% of all voters, while non-white voters went from 23.7% in 2014, to 27.2%. This was an electorate unlike any America had seen, browner and more educated than ever before.
Whatever suburban gains Democrats had made in 2016 were now supercharged. Of the 41 House seats Democrats picked up, 38 of them were in suburban districts, Republicans had held 69 suburban seats before the election, they were down to 31. And Trump’s bigotry was undoubtedly a driving factor. “[College-educated suburban white women] view this literally as a crisis,” Christine Matthews, a Republican pollster, told Vox. “The Trump presidency is a crisis to democracy, our values, our morality. It is making women physically sick. That is the word they use all the time — the word is ‘nauseous ... When Bush did something, nobody said they felt nauseous.”
While Trump’s actions were morally revolting to these suburban voters, it also had a practical effect too—his obsession with immigration, with airing his petty grudges, and with other culture war issues, proved to suburbia that their particular economic concerns didn’t matter to him—concerns that looked nothing like rural America, where jobs had fled overseas and meth and opioid usage devastated communities. Suburbanites didn’t fear losing their jobs, they feared not having affordable health care, affordable child care, or affordable college. They worried about taking care of aging parents and clean air and water. They cared about climate change. They weren’t afraid of immigrants, because immigrants cleaned their homes, cooked their food, drove their Ubers and Lyfts, and handled much of the service sector’s jobs. They were also close enough to actual urban centers (and played and worked in them) to know that they weren’t the MS-13-infested crapholes Republican ads claimed they were.
Hence, while Trump won suburbs 49-45 in 2016, per the exit polls, it was a 49-49 tie in the national House 2018 exit polls. And while Trump won white women in 2016 by a 52-43 margin, it was a 49-49 tie in 2018. And 51 million Americans still voted for Republican candidates, the same ones screaming about Honduran caravans and MS-13 gang members, yet as a whole, America had resoundingly rejected that appeal to bigotry. For a moment, it seemed as if Trump’s 2016 victory was, indeed, an aberration.
But, it turned, out, it wasn’t that simple.
One of the dirty secrets about polling is that public opinion just doesn’t change that much. Take marriage equality, for example. Per Gallup polling, only 27% approved of gay marriage in 1996, and didn’t hit majority support (53%) until 2011. That is, support increased 26 points in 15 years, or about 1.7 points per year—and that’s considered a massive sea-change in public opinion!
Yet because different polls, using different methodologies, taken at different times might show significantly different numbers, people assume that a typical polling trendline is a metaphorical mountain range, full of peaks and valleys, as people bounce all over the place.
But public opinion doesn’t change. Ask yourself, when was the last time you changed your mind about a major issues, such as Trump’s job performance, or abortion, or new tax cuts for billionaires? Odds are, you haven’t, not in a long time, and there’s nothing special about you. Furthermore, we are so ideologically polarized, that it is even more rare for people to change their minds. In 2016, researchers Jonas T. Kaplan, Sarah I. Gimbel, and Sam Harris took 40 liberals and put them under an MRI scanner. They then challenged their strongly held political and non-political beliefs, and watched the effect on different regions of the brain. “The brain’s systems for emotion, which are purposed toward maintaining homeostatic integrity of the organism, appear also to be engaged when protecting the aspects of our mental lives with which we strongly identify, including our closely held beliefs,” they concluded. In other words, once we hold a belief, any challenge to that belief isn’t intellectual, it cuts to our very own emotional sense of self.
Our opinions are our identities, and particularly so when discussing politics.
There are no shortage of psychology experiments showing this dynamic in action, like a 1975 Stanford experiment in which participants were shown 20 suicide notes, half real, and half fake, and then asked to guess which ones were which. Some students were told they had been fantastically accurate, guessing 24 of 25. The others were told they pretty much sucked, getting only 10 right. However, that wasn’t true, and they were then told it wasn’t true. Knowing that their initial assessment had been a lie, they were then asked to guess how many they actually did get right. Those who were originally told they had done well, guessed as though they had done well. Those who were originally told they were terrible at the task, guessed that they did poorly. “Once formed, impressions are remarkably perseverant,” the researchers concluded. Now take that proclivity among us humans to cling to our beliefs, and overlay partisan and social media on top of that, and people have even more opportunities to fall into ideological echo chambers that further reinforce their ideological rigidity.
My polling firm Civiqs has a daily Trump approval rating tracker (Civiqs.com), which means we asked respondents every single day of his presidency whether they approved of him or not. The trendlines are essentially two parallel horizontal lines, with just the barest up- and downticks (caused by statistical noise, the “margin of error” you hear about in polling). On the day he took office, we pegged Trump’s job approvals at 42 approve, 51 disapprove. On Election Day, it was 42 approve, 55 disapprove—and that’s after 300,000 Americans died of COVID-19 and other disasters. Now, five weeks after undermining American democracy with his coup attempt? He’s actually up a tad, 43-55!
People don’t change their minds. That’s why Trump’s 2016 supporters stuck with him no matter how many times reporters asked them to reassess. They stuck with him when the factories didn’t come back, and when China responded to Trump’s trade wars by shifting their agricultural buying to Brazil, and when hundreds of thousands suffered mostly preventable deaths as their president suggested injecting Clorox. And they stuck with him in 2020. That was disappointing, for sure, given four years of never-ending chaos, incompetence, bigotry, and in his last year, mass death, but at least there was some logic to it. Abandoning Trump would mean they had been wrong the first time, and no one likes to admit they were wrong. No one likes to change their mind.
What was unexpected, and a gut-punch to so many, was that Trump got 11 million more votes than in 2016. Assuming some dropoff, since a small number of people likely changed their minds, while others died or were incapacitated, that means that over 12 million people took a gander at Trump’s job performance over his first term and said, “yup, I want more of that.” The 2018 midterms had given us hope that Trump and his 2016 election was a historical aberration, and that the American electorate had swung strongly in the Democrats’ direction as a result. This year told us that was a pipe dream, and that a message of bigotry, xenophobia, and division didn’t just have strong political appeal, but it was one of the strongest political appeals—one so strong that Republicans couldn’t pull it off without Trump on the ballot.
Yes, Joe Biden won, seeing Trump’s 11 million new votes and countering with 16 million new ones of his own. The scale and scope of that accomplishment is nearly beyond comprehension. Take Georgia, for example. Trump won the state by 5% in 2016, and then added 372,000 new votes to that total, a breathtaking 17% increase in turnout. In any normal year, that would be significantly more than enough to win. Yet Democrats turned out nearly 600,000 new voters—roughly 30% more!—to score a razor-thin 12,000 vote victory.
In Arizona, Trump got 400,000 more votes in 2020 than he did in 2016, when he won the state by 3.5 points—a 25% increase in votes. Joe Biden got 500,000 more votes, a 43% increase, enough for an 11,000-vote victory. In Texas, Biden got 1.4 million more votes than Clinton did in 2016! Lucky for him, Trump got 1.2 million more votes on his own, and he held on by a little over 5%. Had Trump gotten 2016 turnout, Biden would’ve won Texas. We know why people turned out for Biden—Donald Trump was the single best argument for a Democratic vote in American history. But what exactly drew all that new Trump support? The answer, it seems, was racism, bigotry, sexism, and xenophobia—the very same factors that fueled the 2018 Democratic wave. How could that be?
Trump didn’t just campaign on those divisive themes. He reveled in them. He was never happier than when attacking Massachusetts Sen. Elizabeth Warren as “Pocahontas,” a racist slam at her belief of American Indian heritage. Or when he told the Democratic “squad”—Reps. Alexandria Ocasio-Cortez, Ilhan Omar, Ayanna Pressley, and Rashida Tlaib—to “go back and help fix the totally broken and crime infested places from which they came,” and led chants of “send her back” directed at Omar at a campaign rally. He suggested that then Sen. Kamala Harris “doesn’t meet the requirements” to be vice president because she was Black and South Asian Indian, an echo of his past ”birtherism”—the claims that Barack was born in Kenya, thus couldn’t serve as president. He called the SARS-CoV-2 coronavirus the “china virus” or the “kung flu.” And just like in 2018, MS-13 made its appearance in ads. One tweet from the Trump campaign featured two tattoo-clad gang members with the caption “I’M ON TEAM JOE! Thanks for pledging to not deport us!”
But nothing dominated Trump and GOP rhetoric than the Black Lives Matter protests and calls to “defund the police” that swept the nation after the murder of George Floyd by Minneapolis police. Just like the Honduran caravan, this was a ready-made moment for both Trump and his state media at Fox News, personified by Mark and Patricia McCloskey—the a-hole millionaires in a walled-off corner of St. Louis, hysterically waving their guns as peaceful Black Lives Matter protesters walk by minding their own business. For someone as skilled as Trump at tickling the racist corners of the conservative id, he couldn’t have invented a better boogeyman.
Remember how public opinion doesn’t change? Well, that’s not always true. Sometimes, public events can deliver information so dramatic and shocking, that it can knock people out of their existing frame of reference. Floyd’s murder, and those of Ahmaud Arbery and Breonna Taylor around the same time, delivered a jolt to the American psyche. White people could no longer pretend violence against Black Americans didn’t exist, or was only isolated in nature, or only happened against people who “deserved it.” “Silence is complicity” became a theme.
Civiqs had been tracking attitudes toward Black Lives Matter for years. On the eve of Floyd’s murder, only 32% of white Americans supported the movement, while 38% opposed it. Twenty-eight percent of them didn’t even bother to have an opinion about it. Practically overnight, the number of whites supporting the movement to 43%, opposition dropped to 34%. It seemed like real progress! But that was before Trump and Fox News fully engaged. And they both did so with true gusto.
Indeed, it was as if Trump challenged himself in the run-up to the election to find new ways to attack and demonize the Black Lives Matter movement every single day. “Left-wing mobs have torn down statues of our founders, desecrated our memorials and carried out a campaign of violence and anarchy,” Trump said in one speech. “Whether it is the mob on the street, or the cancel culture in the boardroom, the goal is the same: to silence dissent, to scare you out of speaking the truth and to bully Americans into abandoning their values.” He attacked efforts to teach children about this nation’s legacy of slavery as “a form of child abuse.” He claimed at an Atlanta rally, “The stated goal of BLM organization, people, is to achieve the destruction of the nuclear family, abolish the police, abolish prisons, abolish border security, abolish capitalism, and abolish school choice—that's what their stated goals are." When New York City announced plans to paint a “Black Lives Matter” mural on Fifth Avenue in front of his Manhattan headquarters, a triggered Trump tweeted that it would “denigrat[e] this luxury avenue,” and called it a “symbol of hate.”
As protests erupted after every brutal police killing of yet another unarmed Black person, Trump and his campaign moved swiftly to weaponize them to their advantage. After a series of protests in Philadelphia, Trump claimed the city was “torn up by Biden-supporting radicals.” In one DC protest, Black protesters tackled and subdued a white radical clad in black who was using a hammer to break up the curb into projectiles to throw at police, then dragged the man to the police line requesting he be arrested. Ten days later, a selective slice of that video made it into a Trump ad claiming “chaos in the streets” that would proliferate if Biden won because he wouldn’t be able to “stand up to the radical leftists fighting to defund and abolish the police.” A group of Black protesters scuffling with a white man was exactly the exclamation point the Trump campaign wanted to broadcast, so much so, that the footage made it into at least one other ad.
Indeed, Trump campaign adviser and 2016 campaign manager Kellyanne Conway couldn’t contain her glee at the protests, telling Fox & Friends, “The more chaos and anarchy and vandalism and violence reigns, the better it is for the very clear choice on who’s best on public safety and law and order.” As Biden responded, “[Trump] views this as a political benefit to him. You know, he’s rooting for more violence, not less, and is clear about that.”
In that Civiqs polling over attitudes toward Black Lives Matter, an interesting dynamic emerged. White support for the movement ebbed from its high of 43% down to 39%—still well above the pre-Floyd level of 32% (remember, public opinion usually shifts much more slowly than that). But remember all those people who were undecided about BLM? Well, Trump and Fox News made sure they would turn in opposition. The number of undecideds dropped 15 points, from 27% to 12%. Meanwhile, opposition rose 14%, from 34% to 48%—a mirror-like inverse. And as might be expected, lower-education whites were far more susceptible to this fearmongering. By Election Day, non-college whites opposed BLM by a net 14 points—50% opposed, 36% supporting. And while still a disappointment, whites with a college degree only opposed BLM by a net 5 points, 46% opposed, 41% supporting.
Did this drive conservatives to the polls in those shocking numbers? Undoubtedly, but the impact of the Black Lives Matter movement wasn’t one-sided. Among Joe Biden’s 16 million new votes was record turnout from core Democratic constituencies. Among 18-29 year olds, 25 million voted, or about 53%—an all-time record. This age group, disproportionately Asian, Black and Latino, voted for Biden by a nearly 2-1 margin. Black voters over the age of 65 gave 95% of their vote to Biden, as did 90% of Black women, 84% of Black men, 70% of Asians and Pacific islanders, 69% of Latina women, and 59% of Latino men. White voters? Biden only got 44% of white women, and 39% of white men.
These numbers had an impact on the key battlegrounds. Those massive voter gains in Arizona and Georgia we discussed earlier were driven in large part by Latinos and American Indians in the former, and Asian and Black voters in the latter. In fact, Asian turnout in Georgia grew by 91% compared to 2016 . Biden doesn’t win Michigan, Pennsylvania, and Wisconsin without strong Black turnout in Detroit, Philadelphia and Pittsburgh, and Milwaukee, respectively.
Yet Biden’s large 7-million strong national popular vote victory and 306-232 Electoral College victory masked just how close we were to a second Trump term. Earlier we discussed how Trump’s 2016 victory hinged on 77.744 votes in just three states: Michigan, Pennsylvania, and Wisconsin. Well, turns out that Biden’s hinged on just 76,514 votes in Arizona (10,457), Georgia (11,779), Nevada (33,596), and Wisconsin (20,682). Flip those four states, and Trump wins the Electoral College 275-263. Fact is, we don’t have a real democracy, and the United States system of electing our president and Senate over represents small, rural, predominantly white states. Hillary Clinton should’ve won in 2016. We shouldn’t have had to sweat the results in four states given Biden’s massive national popular vote lead.
And while we could breathe a sigh of relief that Trump wouldn’t be back, at least not for another four years, that Republican voter surge cost Democrats downballot, including several competitive Senate seats, 10 House seats, and countless state-level races—despite polling suggesting Democrats being poised for another 2018-sized victory. Remember those problems with 2016 polling? The problems pollsters thought had been solved by modeling for educational attainment? All seemed great in 2018, when polling accurately predicted and gauged the size of the Democratic wave. Polling had even been good in 2019, when Democrats won governor races in the Republican bastions of Kentucky and Louisiana. Yet here we were again, dealing with the same crap as 2016. It turns out, something else was afoot.
Put simply, Trump was able to turn out people that hadn’t voted before, and people didn’t even know they existed. That’s why they weren’t polled—they were a hidden demographic. And they only turned out for Trump. Some conservative pollsters were calling these the “shy Trump vote,” but there’s nothing shy about them. “The hidden deplorables aren’t Republican,” I wrote. “They aren’t even conservative. They’re apolitical, otherwise ignoring politics, because their lives legitimately suck. They live in meth country, with dim job prospects (in fact, those two factors are highly correlated). Institutions have failed them—corporations abandoned them for cheaper labor overseas, government seems and feels distant, and it’s certainly not improving their lives. Cities feel like walled gardens—unattainable, unaffordable, yet that’s where all the jobs are, the culture, the action. These deplorables have been left behind. So their attitude? ‘Fuck them all.’”
The NY Times’ David Brooks came to a similar conclusion, talking about those left behind in an information economy that had created a new wealthy class concentrated in urban centers, “While these cities have been prospering, places where fewer people have college degrees have been spiraling down: flatter incomes, decimated families, dissolved communities.” As a result, those forgotten rural regions of the country have rallied around those speaking to their grievances. “This precarity has created, in nation after nation, intense populist backlashes against the highly educated folks who have migrated to the cities and accrued significant economic, cultural and political power,” he wrote.
Conservative pollster Daniel Cox offered further support for this theory of the alienated Trump supporter. “In our pre-election survey on the strength of Americans’ social networks, we found that nearly one in five Americans (17 percent) reported having no one they were close with, marking a 9 percentage point increase from 2013,” he wrote. “What’s more, we found that these socially disconnected voters were far more likely to view Trump positively and support his reelection than those with more robust personal networks. This was especially true among white voters even after accounting for differences in income, education level, and racial attitudes. Sixty percent of white voters without anyone in their immediate social network favored Trump, compared to less than half (46 percent) of white voters with more robust social ties.” In other words, the more alienated and alone a white person was, the more likely they were to support Trump. This was such a strong factor, in fact, that among white people with strong social bonds, less than half supported Trump.
Thus imagine that rural white person locked out of the prosperity of elite, urban America. They can’t afford to buy into those economic and educational opportunities. Their youth are abandoning their cities and towns for greener pastures, leaving them alone, bitter, angry at their losses. They are aging, further adding to their isolation, as their children move away and focus on their own children. Democrats talk about making college more affordable, or even of forgiving college debt, but that does nothing for them. In fact, they believe their taxes are now paying for others to have better educational opportunities than they or their families have. The jobs have long left, and whatever hope Trump gave them of a manufacturing and coal revival are dead, squashed by those coastal elites. Alcoholism and drug addiction is rampant. A study by University of North Carolina-Chapel Hill sociologist Charles Kurzman found that “Counties with the most clandestine drug lab busts — averaging one or more per year since 2012, according to addresses listed online by the Drug Enforcement Agency—supported Donald Trump over Hillary Clinton by a margin of more than 2 million votes, according to preliminary returns. Trump lost the popular vote in the rest of the country [by 3 million votes].”
Life in Trump country is legitimately awful, hopeless, so it’s easy to see how receptive those voters might be to appeals to racism and xenophobia. Think about Mahoning County, Ohio, which we discussed earlier. Hillary Clinton won by just over 3 points after Barack Obama had won it by over 28 points in 2012. Turns out, Donald Trump won it by two points in 2020, the first time a Republican carried it since Richard Nixon’s 1972 49-state landslide victory. The difference? Biden actually got 1,000 more votes than Hillary Clinton did in 2016, about 57,000. But Trump went from nearly 53,000 in 2016 to just shy of 60,000 in 2020.
To reiterate, because it’s important, there’s no sign that these Trump-only voters—whether you want to call them “shy Trump voters” or “hidden deplorables”—are conservative, or Republican, or even ideological. To the contrary, by all appearances they have no policy considerations beyond the overt expression of bigotry against anyone deemed an “other”—be it Black and brown Americans, immigrants, or even “liberals” in general (or like they say in their online forums, “libtards” or “democRATS”). Trump literally begged them to turn out in those 2019 governor races to no avail.
"Here's the story," Trump said at a Kentucky rally on election eve, 2019. "If you win, they are going to make it like, ho hum. And if you lose, they are going to say Trump suffered the greatest defeat in the history of the world. You can't let that happen to me!" Democrats defeated the incumbent Republican governor the very next day, in a state that Trump would win by over 554,000 votes in 2020, 62-36%. Just a couple of months later in Louisiana, Trump suffered a second such embarrassment. “You will deliver a powerful rebuke to the socialists trying to demolish our democracy,” he said at a rally in northern Louisiana’s Bossier City, a Republican stronghold. “You got to give me a big win, please, O.K.” Again, Democrats won the race in a state that Trump would win the next year by roughly 400,000 votes, 58-49%.
These Trump-only voters weren’t interested in electing those Republicans, no matter how much Trump begged. Indeed, they aren’t really interested in building anything—their towns and homes are decaying, fading into irrelevance. All that matters is finding some purpose to make life bearable, and the Republican Party isn’t delivering it. But Trump? He put brown kids in cages. He sent federal troops against the Black Lives Matter “mobs.” He nominated judges hostile to a woman’s right to have agency over her body. And above else, he tore down norms, traditions, the DC bureaucracy, the media, even his friends the second they were no longer useful to him. He was the personification of their rage made real, in the Oval Office itself.
Perhaps the clearest example of this dynamic was in the “jungle primary” for Georgia’s special Senate election. In this system, all candidates, from all parties, ran on the same ballot line, and the top-two vote getters advanced to a January 2021 runoff election. Democrats coalesced around a single candidate, Rev. Raphael Warnock of Atlanta’s Ebenezer Baptist Church, Rev. Martin Luther King’s church. Republicans, on the other hand, had two candidates locked into a tight battle to advance to the runoff—appointed Sen. Kelly Loeffler, the incumbent, and Trump loyalist Rep. Doug Collins. Indeed, there had been drama when Loffler was appointed to the vacated seat in December 2010 over Trump’s objections.
The supposedly moderate Loeffler was chosen by Georgia’s Republican governor Brian Kemp in a bid to stem losses in the suburbs, yet the Collins jungle primary challenge forced her to try and outflank Collins to his right, and the way she did so was quite masterful: She ran an ad claiming that she was “More conservative than Attila the Hun” and had a ”100 per cent Trump voting record.” Now, think about the Huns. There are no Hun ruins to visit, no Hun art, no Hun literary or philosophical tradition. No Hun contributions to world European or world civilization. In fact, all the Huns did were rape and pillage throughout Europe, and using that fear of further destruction to extort the Roman Empire into near insolvency. It’s no surprise that this appeal to destruction would appeal to Trump’s Republican Party. What is more surprising, perhaps, is that the Kelly Loefflers of the Republican Party, supposed moderates, can no longer resist the siren song of Trump’s bigoted appeals.
The 2018 elections gave us hope that Trump’s successful appeal to bigotry had been an accident, a historical aberration. The year 2020 proved otherwise. There is a receptive American constituency for racial, ethnic, and religious hate and divisiveness, one that is otherwise disengaged from mainstream American society and institutions.
The big question for the coming years is whether Republicans can bring out these Trump-only voters, even when Trump is not on the ballot. Can they replicate his success in using appeals to racism, sexism, and bigotry in order to generate that turnout? They haven’t so far, even when Trump has begged them to vote. America’s direction in the coming years will be determined by that answer.
The outcome of the Georgia Senate runoffs will give us a big clue as to which direction we’re headed.