On Tuesday night, Donald Trump beat Hillary Clinton to become our 45th President. Trump won in the Electoral College and Clinton won the popular vote by a significant margin. Turnout was just shy of 130 million voters—almost identical to 2012— so Trump won despite a weak ground game and Clinton being heavily favored in state and national polls.
What the heck happened?
The short answer is that the polls were off because they assumed the Obama Coalition, the force of people of color/women, would more or less hold for Clinton. The Coalition did hold together for Democratic women but not men, which means that the polls were way off due to unexpected gender bias on the Left.
Context
Gender scholars in political science have long identified a strong bias against female candidates due to the masculinity the office embodies and gender biases in media coverage that diminish female candidates such as more negative coverage, a focus on dress and appearance, and outright sexism. Clinton’s bid was the tenth time a woman ran for the office, and she was faced the same gender bias in media coverage as the previous nine, with the added layers of hyper-sexist new media and social media. Furthermore, because leadership is defined as masculine, female presidential candidates face a double-bind where if they follow the norms of leadership, they are seen as too aggressive, but if they act in ways that are expected of women, they are seen as weak, incompetent leaders.
It is no wonder, then, that 7% of Americans say they will not vote for a female presidential candidate, and 26% are “angry or upset” at the idea of a woman in the White House. Some Americans genuinely fear a female president.
Clinton faced some world class sexism in this election, but it was ignored by many on the Left because we tend to see sexism as being less prevalent or less important than other systems of oppression. But we don’t live in a post-sexist society, and much of the public dialogue about Clinton has been explicitly and implicitly sexist, starting with the primary.
Public dialogue about Clinton has been so sexist that many only see her as a caricature— a corrupt, conniving, shrill villain, instead of the staid person and candidate she is. The double standard is highlighted when we imagine the short life the faux email “scandal” would have had if Joe Biden were running, or how unsuitable a Clinton candidacy would be were she with her third husband; a husband who had violated U.S. immigration laws; on whom she had cheated; had she refused to release her tax returns; had 13 allegations of sexual assault; and joked that she likes to grab men by the dick without their consent. This election was anything but a fair race.
Given what we know about gender bias and the presidency, it is empirically absurd to project outcomes without taking latent sexism into account, which is what pollsters did. It is also absurd to talk about the outcome of an election with the first female candidate without discussing gender, which many analysts are now doing.
The Data
When it comes to gender, most women identify as Democrats, and this year saw the largest gender gap in history with a 12-point advantage for Clinton (see table below).
When it comes to race and gender, a majority of white women voted for Trump, while most women of color voted for Clinton (see table below).
This snapshot data, which many have reported on, is highly misleading when it comes to understanding what happened in the 2016 election. White women did not cause the election upset. They did not abandon Clinton. She actually succeeded in winning the vote of more white women – a 1% increase over 2012 (see table below). I am heartened that so many journalists have woken up to the fact that a majority of white women vote Republican for a variety of reasons (e.g., racism), but pollsters knew this and accounted for it in their projections.
This table does reveal an obvious pattern that explains the upset. With the exception of Latinas, women held their numbers in the Obama Coalition, while men in the Coalition broke to vote for Trump and third party candidates in surprising numbers. Overall, men’s support for the Democratic candidate dropped from 45% in 2012 to 41% in 2016, while women’s support held firm overall.
Percentage of the Vote for the Democratic Candidate: 2012, 2016
| 2012 | 2016 | |
| White Men | 35% | 31% |
| White Women | 42% | 43% |
| Black Men | 87% | 80% |
| Black Women | 96% | 94% |
| Latino Men | 65% | 62% |
| Latino Women | 76% | 68% |
| Others | 66% | 61% |
A 4% drop in support from men on the Left is significant. We are still waiting on state-by-state voter demographic information, but had men voted for Clinton in the same numbers as they voted for Obama in 2012 in key swing states, she would have won the election.
Other Factors
Other factors came into play in the election as well. We know that while voter suppression in the form of ID laws and a reduction in polling places demobilized votes on the Left. But experts disagree about whether it was enough to account for Clinton’s loss.
Racism accounts for Trump’s candidacy. Trump rose to political prominence hocking the racist birther lie. He is the answer to 8 years of a Black presidency. We know this from his rhetoric (he hasn’t been shy about it), the racism he encouraged at his rallies, and his appointment of alt-right hero Steve Bannon as his chief strategist. Research on Trump’s supporters finds that racial resentment, not economic anxiety, is the driver. Exit polls show that Trump voters were older, white, and well-off, and their primary concern was the deportation of immigrants.
But racism doesn’t explain the election upset in terms of numbers because it played a predictable, predicted role in the election. Trump’s support with whites looks similar to Romney’s support from this group in 2012.
A factor that did contribute to the upset is third party voting, but it is also likely a reflection of deep-seated gender bias. Nearly three times as many people voted third party in 2016 compared to 2012 (6.9 million compared to 2.4 million). This pattern only makes sense if there was a particularly charismatic third party candidate (there wasn’t) or both major party candidates were seen as equally awful. Throughout the campaign, Clinton and Trump were framed with false equivalency, aided by false balance in the press that simply would not have worked if Clinton were a man. Many Americans unconsciously dislike ambitious women, so it was easy to hyperbolically paint Clinton as a villain for mundane political trespasses.
Eleven days prior to the election, FBI Director James Comey reignited fears about Clinton’s emails with a bafflingly vague letter to Congress that possibly affected the outcome. Comey admitted there was nothing to the “scandal” two days prior to the election, but as Trump campaign manager Kellyanne Conway noted, “the damage is done.” Clinton’s polls dropped precipitously in the final week, and while we can’t know for sure whether the 13% of voters who were undecided the week before the election were influenced by the Comey letter, most of them broke for Trump. The margin of late deciders for Trump was especially large in the key states of Wisconsin, Minnesota, and Florida.
As noted above, the Clinton email story is not a real scandal, but it sustained the gendered #CrookedHillary frame that dominated the way we think about the first women in a general presidential election. It was easy for many to buy into the Hillary hype about corruption because Americans have always seen ambitious women as corrupt, vile, unhealthy, and villainous.
Next Time
I’ve had a knot in my stomach for over a year about Clinton. I didn’t dare believe she would make it past the primary because we’ve been here before. I had even published a book about the impossibility of a female presidency that predicted Clinton’s loss in 2008. I started trusting the polls in the general election because they were so positive and her opponent was/is a buffoon. I knew better than to entirely trust them because pollsters generally underestimate sexism, so I halved the margins, which were still enough pre-Comey’s October Surprise.
I trusted the polls through #Shrillary, #Killary, smile more, smile less, a “horse on her way to the glue factory,” Clinton using the bathroom is “gross,” she doesn’t have the stamina, she doesn’t look “presidential,” “when she walked in front of me, believe me I was not impressed,” “bitch,” “whore,” “lock her up,” “nasty woman,” “kill the bitch.”
I trusted the polls because I believed that men on the Left were with us in this struggle, but in the end, too many left us to vote for a racist, sexist, xenophobic, hyper-masculine demagogue. This was not a victory on the Right, but a failure of the Left.
In the future, when the stars align and we get a female candidate who is the most qualified candidate in history with remarkably high name recognition, she will face the same tired sexism that upset this election. We need to remember that every woman who runs for the presidency becomes a villain.